azra Špago - УКИМ spago.pdf · university “ss. cyril and methodius“ skopje . faculty of...

UNIVERSITY “Ss. CYRIL AND METHODIUS“ SKOPJE FACULTY OF CIVIL ENGINEERING

Azra Špago

METHODOLOGY OF GEOTECHNICAL MODELLING OF THE CARBONATE ROCK COMPLEXES

- SUMMARY OF DOCTORAL THESIS -

Skopje, 2010.

METHODOLOGY OF GEOTECHNICAL MODELLING OF THE CARBONATE ROCK COMPLEXES

Abstract: This thesis proposes methodology of modelling carbonate rock massifs as a wide basis for building complex constructions such as dams, tunnels and underground excavations, and the like. The goal was to undertake, in a certain way, necessary systematization of knowledge and research in this field and to propose new as well as modify already existing procedures and methods of modelling. Proposed physical and analytical models were formed based on researcher’s own and gathered laboratory and field investigations of carbonate rock massifs on the localities of dams “Salakovac” and “Grabovica” in Bosnia and Herzegovina and “Sveta Petka” in Republic of Macedonia, while with physical models, it was also conducted a modification of already existing models for other rock massifs (GSI classification according to Hoek at al etc.). For forming of analytical regressive models, we used a program packet Microsoft Excel and RockLab. The work also gives an example of forming of numerical model and conceptual interaction matrix by using a programme packet Examine 2D in order to show that by proper choice and varying of main elements of diagonal matrix, which include qualities and conditions of rock massifs in one hand and qualities of engineering activities, in the other hand, we can realistically examine and anticipate interaction rock – structure that is stress-strain behaviour of rock massif. Key words: geotechnical model, carbonate rock complexes, physical model, GSI classification, analytical model, extrapolation of results, numerical model, conceptual interaction matrix.

2

CONTENT

1. INTRODUCTION ......................................................................................... 6 1.1. Thesis objectives ............................................................................... 9 1.2. Research methodology ...................................................................... 9 1.3. Content of the orginal version of the thesis in bosnian language ..... 11

2. THE TOPIC AND REVIEW OF RECENT RESEARCH .......................... 13

3. BASIC THEORETICAL SET UP ON GEOTECHNICAL MODEL .......... 16 3.1. Definition and structure of the geotechnical model .......................... 16 3.2. Relations between geological, engineering-geological, geotechnical models and model of interaction ................................................................. 18 3.2.1. Engineering geological sections (EGS) ........................................... 21 3.2.2. Integral engineering - geological sections (IEGS) ............................ 22 3.2.3. Engineering geological model (EGM) .............................................. 24 3.2.3.1. Engineering - geological model per parameter of velocities of longitudinal elastic waves (vl) ..................................................................... 24 3.2.3.2. Engineering-geological model per deformability parameter ........... 26 3.2.3.3. Engineering geological model per parameter of shear strength .... 28 3.2.3.4. Engineering-geological model per parameter of permeability ........ 28 3.2.4. Geotechnical model in narrow sense (GM) ...................................... 29 3.2.5. Model of interaction - model of stress-strain behavior ..................... 30

4. PROPOSED METHODOLOGY FOR GEOTECHNICAL MODELLING OF CARBONATE ROCK MASS ......................................................................... 34

4.1. Overview, analysis and comparison of the test results of strength of monolithic samples of carbonate rock mass from different locations in Bosnia and Herzegovina and Republic of Macedonia ................................ 34 4.2. Forming the possible physical models for analyses of carbonate rock masses by applicability of the Geological Strength Index (GSI) classification ............................................................................................... 45 4.2.1. Introductory notes ............................................................................. 45 4.2.2. Carbonate complex from the location of “Salakovac”, “Grabovica” and „Sveta Petka” dams .................................................................................... 46 4.2.3. Modification of GSI classification for carbonate rock mass ............. 48 4.2.4. Proposing of combination of GSI classification with state of karstification or with total value of carbonate rock mass porosity ............... 55

3

4.3. Formation of analytical models for prediction of shear strength parameters and deformability of carbonate rock masses for different GSI values ......................................................................................................... 59 4.3.1. Introductory notes ............................................................................. 59 4.3.2. Analytical models for prediction of modulus of deformation for carbonate rock masses based on empiric term Hoek, Carranza-Torres and Corkum, 2002 and Hoek and Diederichs, 2006 .......................................... 59 4.3.3. Analytical models for prediction of cohesion for carbonate rock masses based on empiric term Hoek, Carranza-Torres and Corkum, 2002 ................................................................................................................... 63 4.3.4. Analytical models for prediction of angle of internal friction for carbonate rock mass based on empiric term Hoek, Carranza- Torres and Corkum, 2002. ............................................................................................ 67 4.4. Forming analytical models for establishing correlative dependences between quality of rock mass (RMR and GSI), dynamic (vl, and Edyn) and static characteristics (D and E) of carbonate rock masses based on the results of field tests on location “Salakovac” and “Grabovica” dam ............ 71 4.4.1. Introductory notes ............................................................................. 71 4.4.2. Analytical models for carbonate complex from location of the “Salakovac” and “Grabovica” ...................................................................... 71 4.4.3. Comparation of correlative dependences established based on investigation results from the “Salakovac” and “Grabovica” dam with existing correlations for carbonate complex ............................................... 81 4.4.4. Overview on possible polysemics on implementing formed dependances .............................................................................................. 86

5. GEOTECHNICAL MODELS AND METHODOLOGY FOR FORMATION OF CONCEPTUAL, ANALYTICAL AND NUMERICAL MATRIX OF INTERACTION ........................................................................................... 90 5.1. General information about formation of conceptual model of interaction ................................................................................................... 90 5.2. Formation of the numerical model and interaction matrix for carbonate rock massifs with Examine2D software package ....................................... 94 5.2.1. Introductory remarks......................................................................... 94 5.2.2. Applicated field testing methods of stress state around hydrotecnical tunnel in carbonate rock massifs ................................................................ 94 5.2.3. Numerical model of stress-strain behaviour around hydrotechnical tunnel in carbonate rock massive formed with aid of software package Examine2D ................................................................................................. 96 5.2.3.1. Software package Examine2D ...................................................... 97 5.2.3.2. Results and comments .................................................................. 99 5.2.3.3. Analysis of influence in the matrix of interaction .......................... 103

4

6. CONCLUSIONS AND RECOMMENDATIONS ....................................... 109

7. APPENDIX: ............................................................................................. 113 7.1. Table of disturbance factor Df wich is necessary in term for Hoek-Brown failure criterion ............................................................................... 113 7.2. Table for MR -relation between deformation modulus of rock mass and modulus of intact part of rock mass irm EE / ( term Hoek and Diederichs, 2006) ........................................................................................................ 113

8. LITERATURE .......................................................................................... 116

LIST OF SYMBOLS ................................................................................... 121

5

1. INTRODUCTION

1. INTRODUCTION It is generally known that, given the current level of development, geotechnics as a complex and interdisciplinary theoretical-practical discipline offers solutions to numerous problems of importance for modern civil engineering and mining as well as problems related to preservation of environment, through estimation of natural and technogenous hazards, risks and other technical issues. It can be stated that there are two key moments is in modern geotechnics:

• emergence of new and specific requirements in research, design, construction and auscultation of artificial structures and interventions that contribute to ever increasing impact on geological environment;

• development of computer technology and therefore almost unlimited possibilities of simulation of various complex processes in soil and rock, using numerical procedures.

Both of those moments require that great attention be paid to development of new and modification of the existing methods of terrain modelling. The first aspect leads to the emergence of new and complex types of influence and therefore the complex reaction of geological environment. Second aspect imposes requirements for finding a method for reliable definition of the quantitative parameters that would act as a realistic input data for complex stress-deformation simulations. It can be stated that now days it is more difficult to define reliable input data for numerical analysis then to perform a complex and extensive simulations. Such simulation would offer little meaning if performed without realistically input data. Since the main interest in geotechnics is geological environment, which is immensely complex and since every attempt of simplification, to a greater or lesser extent translates to abstraction, idealization, and generalization, the need for modelling in geotechnics is perhaps more pronounced than in other related sciences. Geotechnical modelling is therefore a normal consequence of the need to comprehensively and realistically examine the properties of the natural environment and its response to the changes that occur as a result engineering activities. Only a few decades (40 years ago) primary field of interest in this area were the laboratory and site experiments whose results were very frequently included into design solutions without paying too much attention to the properties and behaviour of the terrain as a whole. With the development of computers and parallel with that, of numerical methods, the interest shifted towards the modelling of interactions and again with insufficient attention to the properties of the natural environment. Given these introductory remarks, this paper will propose some procedures of geotechnical modeling of carbonate rock massif. This term has a broad meaning, ie, massif can appear in different forms. Without going into details, it

6

1. INTRODUCTION

should be noted that the carbonate rock mass consist of calcite and dolomite minerals and genetically belong to sedimentary and metamorphic rocks. Sedimentary rocks composed of calcite is called limestone while sedimentary rocks comosed of dolomite are called dolomite. Marble is a metamorphic rock built of crystalline calcite grains, or much more rarely dolomite (dolomite marble). It is formed by cross-crystalization i.e. metamorphosis of limestone and dolomite. It is obvious that there is a genetic cause-effect relationship with carbonate complexes, which in some way affect certain similarities in their mechanical and hydraulic behaviour. The idea that the problem of geotechnical modelling of so called carbonate rock mass be analyzed in this doctoral dissertation comes from following aspects:

• specific characteristics and conditions of carbonate rock massif;

• practical aspects.

In fact, carbonate complexes have certain specific features that make them somehow different compared to other types of litological solid rock masses. In other types of rock masses it is common that their mechanical and hydraulic behavior depends on the properties and conditions of monolithic parts on one side and condition of discontinuity on the other side. With carbonate complex in some cases the additional influence can have state of karstification and state of disintegration which manifests itself as a lack of mass in discontinuous medium. Conceptual scheme of interaction which generally presents impacts of characteristics of monolithic rock parts, structure od jointed carbonate rock massif and the state of karstification on certain aspects during the construction of an engineering facility is given in Figure 1.1. Practical aspects of problem analysis are arising from the basic forms of interaction shown in Figure 1.1. They are encountered during construction of any engineering structure in carbonate areas, and their type, intensity and impact on the environment must be defined separately for each structure. It is very important that for treating this problem particular methodology be adopted, which should cover the most important aspects of the interaction which in turn is unthinkable without formation of the appropriate geotechnical model. Practical aspects of the problem analysis arise from the fact that the areas of Bosnia and Herzegovina, approximately one-third the space, Macedonia and the entire Balkans is characterized by the presence of carbonate complexes, where large number of complex engineering structures have been constructed and it will continue to be the case in the future. In this context, their modeling is still a practical and scientific challenge.

7

1. INTRODUCTION

Characteristics of monolithic rock

parts

Characteristic structure and texture and lithological composition can affect the state of jointing of massif in a given field of stress.

Lithological composition of carbonate rocks affects the type and possible conditions for development of karstification.

Monolith strength affects the conditions of work execution, massif strength, etc.

Fi, j

Micro-jointing characteristics influence deformability and strength of monolith.

Jointing characteristics

Jointing characteristics affect the potential development and state of karstification.

Jointing characteristics define possible types of excavation instabilty and other conditions for construction

si, j

Development of karstification process over a period of time has the influence on change in monolyth characteristics.

Development of karstification process over a period of time affects changes of jointing characteristics (by increasing the dicontinuity sizes)

State of karstification

Kartstification condition defines possible problems during works, affects the water permeability,deformability and other conditions for construction

Ki, j

Excavations can partially affect the change of monolith characteristics.

Excavations affect formation of additional jointing

Construction can cause a change of karstification condition, underground water inflow, change of porosity within caverns filled with clay

Construction of the structure

Ci, j

Figure 1.1. Conceptual matrix of interaction with four main elements in diagonal of

the matrix which are important for the carbonate massif.

8

1. INTRODUCTION

1.1. Thesis objectives

General objectives:

• Contribution to the methodology of geotechnical modelling by proposing the procedures for modelling carbonate rock massif as widespred base for the construction of complex engineering structures (dams, tunnels and other underground excavations, etc.).

Specific objectives:

• Analysis of the results of own and collected laboratory and field test of important engineering-geological and geomechanics characteristics carbonate rock massif in the Balkans region.

• Forming of physical models for the analysis of carbonate rock massif throught the application of modified scheme, according to the model of previous experience in the definition of so-called "Geological strength index - GSI" classification.

• Forming of analytical models for prediction of shear strength parameters and deformability of carbonate rock massif for different GSI values.

• Forming of analytical models which establish correlative dependance between the quality of rock massif (RMR and GSI), dynamic (vl i Edyn) and static characteristics (D and E) of carbonate rock massif on the basis of the field tests results performed on the dam sites „Sveta Petka“, R. Macedonia, and „Salakovac“ and „Grabovica“, Bosnia and Herzegovina.

• Forming of numerical model and conceptual matrix of interaction with three main elements in diagonal of the matrix performed through a software package Examine2D.

1.2. Research methodology

In accordance with prevously defined issues and with clear objectives in mind, work is completed according to the following methodology:

• Collection of available carbonate massif test results in the Balkans region, particularly laboratory and field test results of strength, deformation, water permeability and other parameters.

• Specific laboratory testing for a specific purpose in this thesis, performed in Geotechnical laboratory of Civil Engineering Faculty in Skopje.

9

1. INTRODUCTION

• Statistical analysis and comparison of data collected from the literature and data collected through research and tests performed for purposes of this thesis.

• For establishing of a physical model through the GSI classification, results of geological and engineering-geological investigation of the carbonate complex from sites areound „Salakovac“ and „Grabovica“ dam on the Neretva river in Bosnia and Herzegovina and „Sveta Petka“ on the River Treska in R. Macedonia are used, as well as own research results. Also, the analysis relative to the formation of similar models for other solid rock was performed by Hoek, E., Marinos, P. and Benissi, 1998; Marinos P. and Hoek, 2000; Hoek, E., Marinos, P. and Marinos, V., 2005; Marinos, P., Hoek, E. and Marinos, V., 2006.

• Forming analytical models for prediction of shear strength and deformability parameters of carbonate rock massif for various physical models defined by GSI values using a software package RockLab which aplies expressions Hoek, Carranza-Torres and Corkum, 2002 and Hoek and Diederichs, 2006.

• Analytical models for the establishment of correlation dependency between the quality of rock massif (RMR and GSI), dynamic (vl i Edyn) and static characteristics (D and E) of carbonate rock massif were formed on the basis of rock massif test results combined field static and dynamic methods at the locations „Salakovac“ and „Grabovica“ dam on the river Neretva, results of the detailed classification of the rock massif around the measuring point with RMR (GSI) classification, where the obtained results were extrapolated in order to make results obtained on individual test sites applicable to the entire area of the dam. The extrapolation procedure applied here is based on regression analysis, which aims to establish a connection between the observed results and corresponding analytical-mathematical expression, ie. regression model. Also, for the purposes of verification of the obtained analytical model, they have been compared with models formed for carbonate complexes, done by other authors for other localities.

• Defining the methodology of implementation of specific numerical models, where method is given so previuosly defined analytical models are used appropriately in interaction analysis of stress-deformation states for the example of hydropower tunnel. The obtained results were compared with results of site tests of stress conditions around the same hydropower tunnel of Fatnica-accumulation Bileća, Bosnia and Herzegovina, within the hydropower system on Trebišnjica situated in compact bank limestones and dolomites. The formation of this numerical model is performed through a software package Examine2D.

• A critical review of the results obtained in this dissertation, as well as definition of conlusions at the end of the thesis.

10

1. INTRODUCTION

1.3. Content of the orginal version of the thesis in bosnian language

The orginal version of the thesis on bosnian language consists 8 chapters, list of symbols, review of figures and tables. Chapter 2. explains the topic and provides an review of recent research in the area treated in this thesis. Chapter 3. outlines theoretical basis necessary for the better understanding of the proposed methodology of geotechnical modelling of carbonate rock complex. Geotechnical model and its structure are defined and connection between certain types of geotechnical models is shown. Further, discontinuity, heterogeneity, anisotropy and natural stress are explained as a general physical-structural properties of rock masses, as well as state of karstification and disintegration which are essential characteristics of carbonate sediments. When designing large and complex construction projects on or in rock massif the interaction between the structure and the rock massif is particularly examined and it is therefore necessary to define the deformability, mechanical strength and stress conditions, and methods of their investigations. This chapter provides an overview of the classification system used in rock mechanics and also describes in detail the classification based on Rock Quality Designation RQD - system, Geomechanical Classification System, or RMR – Rock Mass Rating system, multiparameter classification of terrain as the working environment developed based on data from the Republic of Macedonia, Jovanovski, 2001, ERMR (Excavation Rock Mass Rating) and Geological Strength Index - GSI classification, which are important for this work. This chapter describes the applied extrapolation of results procedure based on combined application of regression analysis and extrapolation of results on the engineering geological models. Chapter 4. shows the proposed methodology for carbonate rock complex modeling by physical, analytical and numerical models. This chapter provides concrete results of own investigation, a series of statistical analysis, regression dependencies, and other essential elements for the formation of geotechnical models. Chapter is adequately illustrated with photographs, where each physical model is presented visually. Chapter 5. provides an overview of the methodology of forming conceptual, analytical and numerical interaction matrix as a procedure that should follow the methodology of modeling the carbonate massif in further development of geotechnics. In the chapter 6. entitled Conclusions and Recommendations conclusive considerations on the proposed methodology is given. Further, guidelines for geotechnical modelling of carbonate rock complex are recommended, a critical overview of the proposed methodology in terms of its application in various phases of testing, designing and construction of the structure and its application to different types of buildings and massif rock. Finally, directions and opportunities for further research arising from this thesis are proposed. 7. Chapter contains Appendices (Table of rock massif disturbance factor D, necessary for defining the Hoek-Brown failure criteria, Table for the MR ratio

11

1. INTRODUCTION

of rock mass modulus of deformation and modulus of rock mass monolith part irm EE / that is used in the application of Hoek and Diederichs, 2006

expression). Chapter 8. outlines a list of abbreviations, literature and project documentation used. At the end there is a list of symbols used throughout this thesis as well as the overview of presented figures and tables. 1. Introduction, 2. The topic and recent research, 3.1. Definition and structure of the geotechnical model, 3.2. Relations between geological, engineering geological, geotechnical models and model of interaction, 4. Proposed methodology for geotechnical modelling of carbonate rock mass, 5. Geotechnical models and methodology for formation of conceptual, analytical and numerical matrix of interaction, 6. Conclusions and recommendations, 7. Appendix, 8. Literature and List of symbols are chapters from original version that given in this shorter version (summary) of thesis in English language.

12

2. THE TOPIC AND REVIEW OF RECENT RESEARCH

2. THE TOPIC AND REVIEW OF RECENT RESEARCH The problem of geotechnical modelling was highlighted in the past few decades when designing the most sensitive and most complex engineering structures such as concrete dams, deep tunnels and other underground excavations, large bridges ie where parameter extrapolation methods are the most developed. Positive experience with high dams design in our surrounding have shown that the results of rock massif properties and condition testing can be usefully depected through engineering-geological cross-section model. This methodology is mainly based on the assumptions given by Kujundžić, 1973, it is constantly expanded over time, and in this context it is pertinent mention the works Kujundžića and Petrovića, 1980; Lokina and Čolića 1980, 1990 and 1996; Lokin, Lapčević, Petričević, 1989; Čolić, Manojlović, 1983; Jovanovski, Gapkovski, Petrevski, 1996; Jovanovski and Gapkovski, 1995, 1998; Jovanovski et al., 2000; Jovanovski, Gapkovski, Ilijovski, 2002, 2003, 2004; Ilijovski, Jovanovski, Velevski, 2004; Ilijovski, 2005; Jašarević, 1999 et al. More recent developments in geotechnics give great importance to the interaction of the geological environment and engineering activities, with the highlight works Pavlović, 1992, 1995, 1996. Problems of general physical and structural properties, especially discontinuity were considered by Deere, 1967; Mihailov et al, 1972; Annon, 1970, Guy, 1973, 1978; Rolofs and Beyer, 1981; Panzera, 1977, 1979, 1983 etc. By the scale effect, the size of the observed area and relation effect were considered by Müller, 1969; Kujundžić, 1977, 1983; Herget, 1988; Charrua-Graca, 1990; Rocha, 1974, Martin et al, 1990; A. P. Cunha, 1990; Rac, M. V., 1968; Lapčević, 1996, 2005, etc. Site testing methods of deformation of rock massif and shear strength were developed and perfected by Kujundžić and his colleagues from the Institute "Jaroslav Černi“, Belgrade, 1965, 1966, 1974, 1977, 1983. Contribution to defining the shear strength of rock massif through empirical failure criteria is given by Hoek and Brown, 1980, 1983, 1988. Application of the method to rocks of poor quality demanded further changes (Hoek, Wood and Shah, 1992) and even the development of a new classification based on the application of so-called Geological Strength Index (Hoek, Kaiser and Bawden, 1995, Hoek, 1995, Hoek and Brown, 1997, Hoek, Marinos and Benissi, 1998; Marinos and Hoek, 2000, 2001, Hoek, Carranza-Torres, Corkum, 2002; Marinos, V., Marinos, P. and Hoek, E., 2005, Hoek, E., Marinos, P and Marinos, V., 2005; Marinos, P., Hoek, E. and Marinos, V., 2006, Hoek and Diederichs, 2006; etc.). Contribution of Barton and Chobeau, 1994; Bieniawski, 1993 should also mentined. Hypotheses about the primary stress conditions given by Heim, 1878; Terzaghi and Richart, 1952; Herget, 1973; Brown and Hoek, 1978; Hoek, 2000; Sheorey, 1994, etc. While the method of test on „in situ“ stress conditions research was developed by Hast, 1958, Leeman et al, 1966, 1968, Kujundžić et al, 1977, 1979, 1980, 1983 etc. Methods for measuring the

13


induced stress state have been developed by Mayer at al, 1951; Tincelin, 1951; Kujundžić et al, 1977, 1979, 1980, 1983 etc. Classification system developed in the field of rock mechanics that needs to be highlighted is Geomechanical Classification – Rock Mass Rating system (Bieniawski, 1970, 1973, 1974, 1975, 1976, 1979, 1989); RSR - Rock Structure Rating (Wickham, Tiedemam i Skinner, 1972 i 1974); ); Q system - Rock Mass Quality (Barton, Lien i Lunde, 1974), multiparameter classification system which can be used for main type of excavation problems, studied by Jovanovski, 2001 and called ERMR ( Excavation Rock Mass Rating) etc. Computers development in recent decades has contributed to the development of numerical calculation method in rock mechanics which enabled new and wider possibilities of stress and deformation calculation. This had significantly stimulated the development of rock mechanics as scientific and technical disciplines as well as the wider application of research results into practice. Following are used: Boundary method, in which only the boundary of the excavation is divided into elements, and the interior part of the rock massif is represented mathematically as an infinite continuum; Domain method, in which the interior part of the rock massif is divided into geometrically simple elements with appropriate characteristics. Ones to be highlighted are: the finite element and finite differences methods which treat rock massif as continuum. Distinct elements is also a method of domain which models each rock block as a single element. These two groups of analysis can be combined into hybrid models in order to minimize disadvantages and maximize advantages of each method. Today there are many different software, based on the above mentioned methods for the analysis of stress-deformation behavior of rock massif, its stability and other geotechnical problems (SWEDGE, SLIDE, EXAMINE, FLAC, PHASE, UNWEDGE, PLAXIS etc.). Also, in the field of numerical modelling, there are a large number of works that are based on the analysis of stress-deformation state of solid rock mass, and in this framework for carbonate complexes (eg Jašarević, Kovačević, Miščević, 1995; Hoek, 2000; Carranza, C.T. and Fairhurst, C., 2000; Dawson, E. et al, 2000; Lolino P. at al., 2004). It is clear that above authors as well as many others have made a significant effort in defined states and properties of rock massif. On the other hand it is also clear that this area is practically inexhaustible for further scientific work, especially in the field of geotechnical modelling in general sense. In light of this it is worth mentioning that during work on the dissertation, some results have been published at the expert meetings (Krvavac, Jovanovski, Gapovski and Ilijovski, 2006 and Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008). Based on this sublimate presentation I came to the conclusion that systematization of knowledge in mentioned field is necessary in some way as well as the definition of procedures and methodology of modelling and their modifications. Hence the topic that is indisputable from both the theoretical and practical point of view. Thus this thesis describes a methodology that shows how it is possible to integrate all the approaches in solving different geotechnical problems. The emphasis was placed on the carbonate massif,

14


which does not mean that similar methodology can be applied to other litho logical types of rock masses.

15

3. BASIC THEORETICAL SET UP ON GEOTECHNICAL MODEL


3.1. Definition and structure of the geotechnical model

Under the term geotechnical model is considered the connection between theoretical and practical knowledge on characteristics of natural geological environment and changes in it made as consequence of engineering activity whose basic task is to formulate optimal technical solutions based on the defined interaction. This complex geotechnical model is consisted of three basic models (Pavlović, 1995, 1996):

• model of natural geological environment - engineering geological sections (EGS), integral engineering-geological sections (IEGS) and engineering geological models (EGM);

• model of engineering activity - geotechnical model in narrow sense (GM);

• model of interaction - model of stress-strain behavior.

These elements above are mutually inseparable and such as can only represent geotechnical model as a unit. The elements are in a very complex interactions of mutual dependence and influence. So the characteristics of engineering activity affects on the choice of significant parameters of the model of natural geological environment. On the other hand, model of natural geological environment which is very determent by geological conditions of area, affects on certain technical and constructive solutions that engineers make, and by that directly influence on important elements of it's model. On engineering activity’s model it affects in the way that can show some unreliable reaction of rock mass and then can change certain elements of model, searching for final solution. These changes can also influence on geological model, because new solutions demand other significant parameters. These interactions of certain elements of geotechnical model are shown in the figure 3.1. It is well known that rock mass is been characterised by large changelibility in time and space at every state and characteristic. If we look at rock mass by microscope and in it's natural size, we can see that it is mainly discontinues, heterogeneous, anisotropies and naturally stressed (prestressed), which is consequence of its' forming conditions, then additional tectonic and some others influences on it as well as natural physical and chemical properties of rocks. Because of that every attempt of representing it could be mainly it's idealization and generalization. So we, by this, get figure of geological model which is largely simplicity of realistic figure and reflection of the stage of knowing the issue itself. The correct characterization and secure forecast how will environment act mostly cannot be assumed.

16


Figure 3.1. Structure of the geotechnical model (Pavlović, 1996).

In modelling process, certain areas inside geological environment where rocks have similar characteristics are selected. However there are too many characteristics that completely characterise certain rock mass, which is way the claim for uniformity of all or most of the characteristics can't be achieved. So before some areas are selected, we choose one or few characteristic for which the uniformity of one area is demanded. We call these areas quasichomogenic zones and they represent the basic and constituve elements of geological model. Inside such zone some coditions or characteristics are the same in every point, and very different outside the zone. Each and every zone is determened by space limits and consists in

PRELIMINARY KNOWLEDGE OF

THE TERRAIN

GEOTECHNICAL INVESTIGATION

DESIGN BRIEFS

MODEL OF GEOLOGICAL

ENVIRONMENTT Š

MODEL OF ENGINEERING

ACTIVITY

MODEL OF INTERACTION YES

VERIFICATION

NO

MODEL OF GEOLOGICAL

ENVIRONMENTT

MODEL OF ENGINEERING

ACTIVITY

MODEL OF INTERACTION

NO

REPEATING OF THE PROCEDURE

YES

VERIFICATION

OPT

IMA

LIZA

TIO

N O

F TH

E M

OD

EL

17


some way characteristics which are important for study. The number of selected quasichomogenic zones does not only depends of characteristics' changelibility, but also of possibility and usefulness of research and the changing technology during construction works. The model will be more realistic, if the number of zones is bigger and their size is smaller. Beside that, by lowering the zone's size, we lower insecurity of extrapolation and range of possible variations to values of representative parameters. On the other hand demands related for further analyses of interactions are becoming significantly larger. 3.2. Relations between geological, engineering-geological, geotechnical

models and model of interaction

Model of natural geological environment replays all important elements of natural geological environment which are relevant for specific aspects of current issue (Pavlović, 1996). It is the main basis of geotechnical model and represents synthesis of results of very detailed and different fields' measurements. Least number of lab analyses, complex geophysical researches and use methods of mathematic statistic and theory of probability. All researched works are made by using most parameters and only one complex analyses introduce us to know what kind of rock mass we can use as base to geotechnical construction of large and complicated structure, first off all high dams, tunnels and other underground excavations, stanchion for bridges, towers for sports' objects etc. While making engineering-geological model, researches from field include some stratimetry as well sedimentary researches, then surveying of tectonic assembly with genetic classification of discontinuity, measuring space between discontinuity of the same families, determining morphologic and other characteristics which are applied and interpreted in rock mechanics, beside geological, geophysical and engineer-geological proceedings. Laboratory tests include all necessary analyses for defying the characteristics of main mass. Main point in preparing labs' and fields' researches is connecting them considering the existence of tectonic blocks, different stratigraphic belonging and litho facial characteristics of rocks, selecting rocks by types, by tectonic blocks and mainly, and determining all incoming results needed for classification for rock mass by proceedings recommended by International Society for Rock Mechanics (ISRM). Models of natural geological environment are done in form of engineering -geological sections (EGS), integral engineering-geological sections (IEGS) and engineering geological models (EGM). Engineering-geological sections (EGS) show results gained by measurements, investigations and researches in the nature and basically they present fact graphic picture of gained results. Integral engineering-geological sections (IEGS) present the final picture of all results. Engineering geological model (EGM) presents the certain simplified picture of terrain with bordered quasi-homogeny zones per parameters that are necessary for the dam projecting (Kujundžić, 1973).

18


Model of engineering activity - geotechnical model in narrow sense (GM) presents the rock mass divided on particular zones recommended for geotechnical intervention (excavation, consolidated injection, grout curtain, anchorage, drainage, etc.), (Kujundžić, 1973). Model of interaction - model of stress-strain behavior presents adequate substitution of real behaviour of rock mass in newly formed condition during working operations, by applying some of model approaches (Pavlović, 1996). Linear way to show phases during defining of model of interaction is given in figure 3.2. In the figure we can see that are model of engineering activity and model of interaction final phase of geotechnical modelling.

Figure 3.2. Linear phases in defining of model of interaction.

The adequate way to show relationship between different types of models is given in figure 3.3. by the engineering geology triangle and geotechnical triangle (partly modificated by Knill, 2003). These triangles show how in each specific case, model of geological environment defined with help of research, observation, experinece and intuition is basis for all other types of models, in order to get input data for further numerical analysis.

Figure 3.3. The relationship between the engineering geology and geotechnical

triangles and the relationship between geological, geotechnical models and numerical analysis (partly modificated by Knill, 2003).

Geological model (Geological profiles, EG sections)

Engineering - geological model (with separeted quasy-homogenity zones)

Model of engineering activity

Model of interaction - model of stress-strain behavior

Observation Experinece

Intuition Synthesis

Geological material and mass properties

Geological process

Geological model

Ground behavior

Geotechnical model

Numerical analysis

ENGINEERING GEOLOGY GEOTECHNICAL

Ground model

19


Physical models Mathematical models

Geological sections EG sections and models Maps Mock models

Analytic

Numerical

Named geotechnical model could be divided additionally on physical and mathematical (Figure 3.4.). Physical models in principal presents certain simplified reproduction of real conditions in terrain. Mathematical models are having one goal in defining certain properties and conditions of the terrain with the analytic relations or models of stress-strain behavior. In any case, physical models are basic for any mathematical model. Mathematical and especially numerical models get significant dimension when they have huge possibilities of the newest electronic computers in mind. Their application however can easily lead to fictitious understanding of the problem which would place a shadow on realistic mechanisms of their behavior. The difference considering some other methods is in that the true and clear results have been given. These results are usually well documented and shown particularly, so we have impression of real resolve of problem. That is way in numerical models, as well as in use of any other modeling method, the experience and free engineer's judgment have very significant part in marking of reality of gotten results. When computer's simulation of behavior is used by numerical methods, previously it must be clearly understood and feel all complexity of problem and possible mechanisms of reaction of rock mass. Mathematical models basically must satisfy two, on the first hand, contradictory apply: that they simulate as much realistic conditions of the field as they can, and the other one is to be as much simple as it can. They can be key of successful progress in methodology of geotechnical modeling only when they are used concretely, as part of complicated integral process. Their efficient and quality directly depend on income results as well as of degree of mathematical formulations that for real show how rock mass behave naturally. While we respect every advantages and possibilities, it must not be forgotten that how true facts of numerical methods depend on reliability and details' degree of engineer-geological model, as well how those methods are not the only ones neither they are always expectable and optimal. Beside all this, numerical methods should be economical, and their technical information precious enough, which can be checked only if we compare gotten results with realistic situations. Result of that should be their verification.

Figure 3.4. Geotehnical model divided on physical and mathematical model.

20


3.2.1. Engineering geological sections (EGS) Engineering geological sections are made as vertical sections of the terrain, and when necessary as transversal vertical and horizontal sections on different stages. Rarely they are made as oblique ones. Engineering geological sections are done by some parameters of investigation and consist show of all individual results of one certain kind of investigation. When we have high concrete dams, then EGS are usually done by further parameters of investigation:

• EGS per litho logy parameter;

• EGS per jointing (discontinuity) parameter;

• EGS per parameter of physical - chemical decompose of rock;

• EGS per parameter of velocities of longitudinal elastic waves (vl);

• EGS per parameter of electric resistance (ρ);

• EGS per deformability parameter (D, E, Edyn);

• EGS per parameter of shear strength of the rock mass (c, ϕ);

• EGS per permeability parameter (Lu). Figures 3.5. and 3.6. show the engineering geological sections for some parameters of investigations for dams „Salakovac” and „Sveta Petka”.

Figure 3.5. The engineering geological section (EGS) per litho logy parameters (1-4) and per permeability parameter of rock mass checked non-parameter testing (I-V) for the profile of the dam “Salakovac”, (Selimović, 2004).

21


Figure 3.6. The engineering geological section (EGS) per jointing parameter for the profile of the dam “Sveta Petka” where is: (al) alluvial sediment,(dl) diluvial sediment, (M) solid marble, (M’’) poorly schistose marble, (MS) carbonate schist, (MSS) carbonate muscovite schist, ДБ – bore pit, ГЛ and ГД – exploratory adits, ДИЛ-dilatometer, P.z.,- fault zones, P - fault, WTE – water table elevation (saturation line), (Final analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004).

3.2.2. Integral engineering - geological sections (IEGS) Integral engineering geological sections (IEGS) represent synthesis of all engineering geological sections which are made by each parameters of the investigation. So in that way all engineering geological sections are overlapped on one figure. Integral engineering geological sections overlapped in this way give complex showing of all information, gotten by measuring on diversion dam and their patterning in one logical unit.

22


IEGS represent the main document with complex results of researched works and base for design of the high dam project. It is done usually in transversal projection across the river that has been separated. Besides IEGS, texture report must be given, where except the explanation of the graphic figure where we have elementary information about characteristics of the researched rock mass and at the end, the conclusions and recommendations for projecting and building of the dam are formulated. IEGS also provides construction of the engineer-geological models (EGM) as well as geotechnical models (GMS). In the figure 3.7. is given IEGS for the arc dam “Mratinje”, the Piva river, Montenegro, (Kujundžić, 1973).

Figure 3.7. IEGS for the arc dam “Mratinje”, the Piva river, Montenegro,

(Kujundžić, 1973).

23


3.2.3. Engineering geological model (EGM) Engineering geological models (EGMS) are done by parameters needed for construction of the object, and somehow they usually represent simplified version of the enviroment itself, I mean the rock mass. In engineer-geological model the quasichomogenic zones are boredered by parameters needed for projecting with numerical characteristics of rock mass in every bordered quasichomogenic zone. The engineering geological models per following parameters are the most made:

• velocities of longitudinal elastic waves;

• deformability;

• shear strength;

• permeability;

• jointing;

• rock mass quality

and others models depending on what parameters are needed for the construction of the dam's project.

3.2.3.1. Engineering - geological model per parameter of velocities of longitudinal elastic waves (vl)

Using dynamic methods of investigation, velocities of longitudinal elastic waves (vl) are measured and with them we account dynamic modulus of elasticity (Edyn), with them is possible to select massif in quasi-homogeneity zones by appreciate showing. Using these models very quality conclusions can be done about characteristics of rock mass inside the zone that has been researched. This model compared with results of static examination of deformability, also establishing correlation dependents of type D = f (vl) is base for construction of model of deformability. Figure 3.8. shows engineering-geological model per the parameter of the velocities of longitudinal elastic waves for the “Salakovac” dam profile, which highlights five quasihomogenous zones according to the elastic properties parameter (vl, Edyn), obtained on the basis of investigation by refraction seismic. Quasi homogeneous zones according to the litho logical structure, water permeability and rock mass damage parameter are also classified in this model.

24


Figure 3.8. Engineering geological model of the dam „Salakovac“, with quasi-homogeneity zones per litho logy parameter (1-5), elastic characteristics (a-d), permeability (I-V) and coeficient of damage (ζ1- ζ3), where are: 1) diluvial material, 2) upper cretaceous – light grey marble limstones, 3) upper jurassic – bedded dolomites and limestone, 4) debris and stone deposit, 5) reversni fault I order – fault zone, 6) diversion tunnel at the right bank built, and at the left bank desinged, 7) depth of foundation of the dam, 8) bore pits,with marked depths of investigation by borhole dilatometer, 9) trial adit with places of investigations: a) shear strength on a system concrete block-rock mass, b) deformation characteristics of rock mass, 10) road tunnel, 11) fractures filled with clay and dolomitic gneiss,12) a number of shear discontiunities grouped in the space of 1 m separated aproximately 10 cm from each other, 13) fluctuations of water table elevation, 14) quasi-homogeneity zones per permeability, 15) quasi-homogeneity zones separated with a using of method of refractive seismic with values of vl and Edyn (Selimović, 2003).

Figure 3.9. shows engineering-geological model per the parameter of the velocities of longitudinal elastic waves for the “Grabovica” dam profile with highlighted dynamic elastic characteristics of the massif (vl, Edyn) according the quasi homogeneous zones

25


Figure 3.9. Engineering geological model of the dam “Grabovica”, with quasi-homogeneity zones per parameter of velocities of longitudinal elastic waves (vl) and dynamic modulus of elasticity (Edyn), where are: (1) debris, (2) gravel with sand and blocks, (3) gravel, sand and river’s boulders,(4) bedded limestone, dolomites and dolomite limestone, (5) local faults, (6) caverns,(7) joints, (8) geological borders,(9) borders of zones with different vl and Edyn,(10) water table elevation WTE, (11) bore pits for permeability,(12) entrance and exit object, (13) and trial adits, with the places of the geotechnical investigations, with hydraulic jacks (HJ) and others (Selimović, 2004).

3.2.3.2. Engineering-geological model per deformability parameter Forming of engineer-geological model by deformability parameter, which is usually expressed through modulus of deformations D, modulus of elasticity E, is based on use of methodolgy of combined static-dynamic methods. This methodology makes posible division of rock mass on the dam profile in certain number of quasichomogenic zones by deformability parameter. This kind of model that represents certain simplify, but simplify with enough cerrection serves as base for analysis of stress-strain behavior or for construction of the physical model, if experimental test of the object is recommended. Figure 3.10. shows EGM per deformability parameter for the profile of dam “Sveta Petka”. This EGM is constructed by models per parameter of velocities

26


of longitudinal elastic waves vl and models and sections per jointing (discontinuity) parameter, considering all results of geological, geophysical and geotechnical investigations. Gradation of parameters is based on defined interval of velocities of longitudinal elastic waves vl, for main zones, for which correlation dependties between vl, and the values of static compressive modulus of deformations and elasticity (D and E) and static shearing modulus of deformations (Ds) have been established.

Figure 3.10. Engineering-geological model per deformability parameter for profile of arch dam “Sveta Petka” where is: (al) alluvial sediment,(dl) diluvial sediment, (M) solid marble, (M ‘) poorly schistose marble, (MS) carbonate schist, (MSS) carbonate muscovite schist, ДБ – bore pits, ГЛ and ГД – exploratory adits, ДИЛ-dilatometer, P.z.,- fault zones, P - fault, WTE - Water table elevation (Final analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004).

27


3.2.3.3. Engineering geological model per parameter of shear strength During construction of engineering geological model per parameter of shear strength we can get two cases. The first one applies to jointed and poor rock massif where is possible to make bordering of individual quasi-homogeneity zones per parameter of shear strength, although given numerical results of characteristics of shear strength apply for whole quasi-homogeneity zone whether at every direction or some specific directions, if it is considered for rock mass which per parameter of shear strength show anisotropy. The second case applies to the stiff but jointed rock mass whose monolits have high values of modulus of deformation. In this case we first find certain number of types of discontinuty and then make compatible clasification of discontinuity. Also we must do examinations of the shear, wheter field's or lab's. We get values that apply for individual types of discontinuity and they are written with compatibile type of discontinuity in engineer-geological model. In the figure 3.10. on can see the quasichomogenic zones per parameter of static shear modulus of deformations (Ds) for the profile of dam “Sveta Petka”.

3.2.3.4. Engineering-geological model per parameter of permeability Engineering geological model per parameter of permeability forms on results of the investigation of permeability and theirs graphic figure on EGS per parameter of permeability. The lines of equally permeability mainly form from results gotten by direct examinations using Lugeon's experiment. They check on integral engineer-geological section, especially considering jointing because of their fitting in one logical unit. Then individual quasi-homogeneity zones mark and border where permeability has nearly equal value. Reserch phase of hydropower station “Salakovac” profile included water permeability investigation which formed the basis for classification of quasi homogeneous zone per parametar of permeabilty (Figure 3.11.) (Magdalenić, Raljević and Jurak, 1976; Selimović, 2004). This model was also used as a basis for design of injection and other works in the rock massif.

Figure 3.11. Engineering-geological model with quasi-homogeneity zones per parameter of permeability (Magdalenić, Raljević and Jurak, 1976) for

28


the profile of the dam „Salakovac” where are: (1) kvartar, (2) upper cretaceous limestone, (3) lower cretaceous limestone, (4) upper jurassic – limestone and dolomite, (5) faults, (6) bore pits, (7) Water table elevation WTE (min), (8) Water table elevation WTE (max), (9) permeability < 1 Lu, (10) permeability 1-5 Lu, (11) permeability 5-10 Lu, (12) permeability > 10 Lu, (13) barren drilling.

3.2.4. Geotechnical model in narrow sense (GM)

Geotechnical model, as said before, represents rock mass on dam division divided on individual zones inside where it is needed to project and do engineer works or geotechnical meliorations. For concrete high dams, in the geotechnical model, further zones border:

• zone of the excavation;

• zone of the consolidation grouting;

• zone of the grout curtain;

• zone of the anchoring;

• zone of the drainage.

Figure 3.11. Geotechnical model of the profile of the dam «Sveta Petka» where is: (al) alluvial sediment,(dl) diluvial sediment, (M) solid marble, (M') poorly schistose marble, (MS) carbonate schist, (MSS) carbonate muscovite schist, ГЛ and ГД – exploratory adits, P.z.,- fault zones, P -

29


fault, WTE - Water table elevation (Final analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004).

These zones, confirmed in geotechnical model and gotten by analysis of IEGS, as result of finished researched works, can be corrected during projecting process taking some constructive or working reasons. On the other hand, they can be corrected while constructing because the excavation is the best way to know rock mass. So the conclusion is that geotechnical model really represents a very solid forecast for projecting and constructing mentioned works.

3.2.5. Model of interaction - model of stress-strain behavior The present example concerns the numerical study of the behavior of natural karst caves at Castellana-Grotte, in the Apulian karst of Southern Italy by means of the Discrete Element Method (P Lollino. & M.Parise, 2004) . This research is focused on the behavior of one of the caverns (namely, the so called “Civetta” figure 3.12. and 3.13.), which can be considered as representative of the behavior of the most the caverns in the cave system. The observed behavior of the carbonate rock mass within the cave shows gradual propagation of fractures through the horizontal rock strata at the roof, until portions of the ceiling, ranging in thickness from a few centimeters to more than one meter, suddenly fall. This rock mass behavior at the top of large excavations or natural caves has been documented in the literatura as the the result of the transition for the lower strata at the roof from continuous elastic beam behavior to Voussoir beam behavior; the stability of the roof formed of vertically jointed beams is then controlled by the eventual formation of a sufficiently thick arch compressive. The aim of this research is to investigate the mechanism that leads to the long-term progressive collapse of portions of the cave roof. A numerical analysis has been therefore carried ou by means of UDEC 3.10 distinct element code (ICG 1999). The process of formation and propagation of cracks normal to the bedding planes due to tensile strength degradation of weathered limstone with time due to chemical and humidity weathering processes of the rock mass has been studied. Then, the response of a numerical model, which includes new vertical joints as an effect of tensile failure, has been evaluated.

Figure 3.12. Longitudinal cross-section of the Castellana-Grotte karst system.

30


Figure 3.13. Cross-sections of the Civetta cavern. In brief, the whole numerical analysis has been divided into four stage:

1. gravitational elastic equilibrium; 2. assignment of the real and unweathered material propreties; 3. gradual reduction of the tensile strength within the elements above the

cave roof; 4. analysis of the stability of a new model with vertical joints at the roof.

The initial configuration of the geometry modeled is shown in figure 3.14. (a). It is characterized by an 11m span horizontal roof, which underlies continuous rock beams delimited by persistent bedding planes. This configuration can be considered as representative of an intermediate stationary stage of the evolution process of the cavern, before that vertical joint openings and breakdowns take place within the roof. A finer schematization of the domain, with lower spacing of the bedding planes than the rest of the domain, has been modeled just above the roof to reproduce structural features closer to reality. Three monitoring points have been chosen along the cave roof to assess the trend of block displacements with timesteps. A linear elastic perfectly plastic constitutive model with a Mohr-Coulomb failure criterion fitting the peak strength envelope assumed and a tenssion cut-off reproducing the tensile strength as measured in the laboratory tests has been adopted for the intact rock. The model, which assumes real and unweathered material propreties under gravitational load (stage 2), results in a stable condition. Few tensile failures appear just above the roof and within a distance about 2 m from it (figure 3.14.(b)). The model does not show block displacements changing with timesteps. In the third stage of the analysis, a gradual reduction of the tensile strength has been applied to the elements above the roof and consequently tensile

31


failures start to propagate in the middle area of them. The model assuming a tensile strength of σ = 20 kPa results in a large development of tensile failures throughout the strata overlying the roof for a 5 m thick area according to a „dome“ pattern: short joints develop alonge the whole span, whereas longer persistent joints form in the middle area (Figure 3.14.(c)). This is consistent with the assumed transition from continuous elastic beam behavior to vertically jointed beam behavior. Consequently, the stability conditions of the new model in figure 3.14. (d) have been evaluated. In this model the mean spacing between the vertical joints has been assumed very low (about 40 cm) to account for the worst conditision for stability.

(a) (b)

(c) (d)

Figure 3.14. Initial configuration of the model (stages 1-3) (a), tensile failures after

stage 2 (b), tensile failures after stage 3 (c), configuration of the model in stage 4 (d).

The pattern of the calculated displacements vectors, which extend for 3-4 m above the roof, shows a clear deflection of the lower beams at the midspan (figure 3.15.). The „dome-shaped“ stress-loosened area, which is indicated by the opening of the horizontal joints subjected to zero normal stress, is also clearly shown in figure 3.15. The plot of the vertical displacements of the three monitoring points with timesteps indicate that the model does not reach a static configuration since the displacements do not tend to a constant value

32


(figure 3.16.). This result implies that for the assumed structural conditions, a sufficiently thick arch compressive is not formed in the jointed beams above the roof, and the cavern ceilling will consequently tend to collapse.

Figure 3.15. Displacement vectors and open Figure 3.16. Vertical displacement joints after stage 4. versus timesteps for three

monitoring points during stage 4.

The influence of the ratio between spacing of the horizontal bedding planes and the cave span at the Civetta cavern has been also analysed. Different configuration assuming the same vertical joint spacing (sv = 80 cm) and spacing of the horizontal bedding planes equal to sh = 20, 30, 40 and 60 cm have been modeled. The resulting midspan deflections of the different models are reported against numerical time in figure 3.17.

Figure 3.17. Midspan deflection versus timestep for models with different spacing of bedding planes.

The figure shows that assumed joint friction angle value spacing greater than 20 cm ensures stable conditions to the cave roof, whereas the model with sh = 20 cm produces a deflection continuosuly increasing with time, as for the case with sv = 40 cm. This implies that in the last case the vertically jointed beams are too slender to form arch compressive, in accordance with the guidelines by Diederichs & Kaiser (1999).

33

4. PROPOSED METHODOLOGY FOR GEOTECHNICAL MODELLING OF CARBONATE ROCK MASS

4. PROPOSED METHODOLOGY FOR GEOTECHNICAL MODELLING OF CARBONATE ROCK MASS Based on showing of significant characteristics and conditions of carbonate rock mass, given in previous chapters, it is obvious that there is a lot of knowledge and approaches which in one or another way have tangible hand barrow with made analyses in this work and with more or less success are applied in practice. On the other hand there is formidable space for development and modification of existed methods, in order to in more details comprehend carbonate massifs and at the same time it would serve as further development of methodology for its' geotechnical modelling. Having in mind these facts, in this chapter is shown proposing of methodology for modelling carbonate massive. The suggestion is based on existing data of characteristics and conditions of carbonate rock mass, mine own research, statistic analyses of parameters, and also series of made correlations for the most significant conditions and characteristics of these massifs. 4.1. Overview, analysis and comparison of the test results of strength of

monolithic samples of carbonate rock mass from different locations in Bosnia and Herzegovina and Republic of Macedonia

Although because of monolithic characteristics of examples on which compressive strength and tension is examined, its' practical use in solving engineer's problems is limited, values of these strengths can be useful parameters that indicate on the quality of rock mass. So it is very often use as enter information for classification of rock, or as value for defining some failure criteria, and this will be shown in next chapter 4.2. In order to know interval in where can vary values of strengths of monolithic parts of rock massif, in this chapter are shown some results of examination of uniaxial compressive strength of intact parts σci of carbonate rocks from the area in Bosnia and Herzegovina and Republic of Macedonia. In tables 4.1. and 4.2. is shown part of results of examinations of uniaxial compressive strength of intact parts σci of rock massifs which are taken from location of ″Salakovac″ dam on Neretva river, Bosnia and Herzegovina, and ″Sveta Petka″, on Treska river, R. Macedonia. In tables 4.2. it can be seen that the most frequent values of uniaxial compressive strength for rocks on the location of “Sveta Petka” dam are in diapason from σci = 40 - 50 MPa. These results are under usuall values for carbonate rocks, that can be explained by influence of microscopic deffects made with earlier tectonic influences, although if we look it macroscopic, the samples are in fresh state. Generally, by parameter of strength of monoliths, it can be concluded that these rocks are with medium strong (by Brown, 1981).

34


Table 4.1. Average values of uniaxial compressive strength σci vertical on

bedding in MPa, samples of limestone for “Salakovac“ dam, “Testing of uniaxial compresive strength vertical on bedding for quarry of Water power plant “Salakovac““, Sarajevo, 1970.

Condition Sample 1.

Sample 2.

Sample 3.

Sample 4.

Sample 5.

Sample 6.

Dry 171,5 162,3 152,3 150,6 158,2 153,2

Water logged 164,4 145,4 139,6 135,1 138,3 144,8

Frozen samples 156,4 137,7 127,1 129,2 136,8 143,3

Table 4.2. Values of uniaxial compressive strength σ ci of monolithic samples from

the location of “Sveta Petka“ dam, (Final analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004).

Number Drill hole Depth (m) Samples

dimension D/H (mm)

Unia. compressive strength σ ci

(MPa) 1. DB-8 21.10-21.30 71.5/177 .5 25.78

2. DB-9 25.2-25.40 71.3/177 35.56

3. DB-9 26.5-27.1 71/168 45.05

4. DB-9 27.80-28.00 71.3/178 50.84

5. DB-9 28.5-28.7 71.3/172 64.89

6. DB-9 29.40-29.60 71.5/178 62.76

7.

DB-9 30.10-30.30 71.4/177.5 54.45

8. DB-9 73.10-73.3 61/152 32.07

9. DB-10 25.10-25.30 83.7/ 209 27.26

10. DB-10 23.50-23.80 83.7/209.5 42.35

11. DB-11 33.70-33.90 71.3/178.5 33.06

12. DIL-2 25.20-25.40 61.4/153 35.12

13. DIL- 2 26.10-26.30 61.5/155 24.91

14. DIL- 2 32.60-32.80 61.5/153 .5 55.88

15. DIL- 2 37.60-37.80 61.5 /155 24.91

16. DIL- 2 38.5-38.7 61 .5 /156 75.96

35


17.

PS-1 32.20-32.40 71.4 / 178 .5 54.45

18. PS-2 31.60-31.80 71.3/179 41.33

21. PS-2 69.70-69.90 61.3/156 49.47

22. S-3 28.10-28.30 71.2/179 61.28

23. PS-4 25.40-25.60 61.5/153.5 41.20

24. PS-5 25.40-25.60 71/180 33.09

25. S-6 30.40-30.60 71.4/176 42.71

26. DVZ-1 17.30-17.50 61.5/154 26.59

27. DZ-1 28.20-28.40 71.2/ 179 44.96

29. DR-3 16.10-16.30 71.3/178.5 38.07

Average values od uniaxial compressive strengths of limestone for concrete's aggregation on “Grabovica“ dam on Neretva river were: in dry condition 172 MPa, water logged 149 MPa, and after freezing 125 MPa, (Selimović, 2004). From these results it is visible that the highest strengths are in dry, then in satured or water logged and in the end after freezing. Beside these results, on figures 4.1., 4.2., 4.3. and 4.4. are given histograms with statistic processed data. From figures 4.1. to 4.4. it can be seen that uniaxial compressive strength of carbonate rock massifs vary in large interval, that can be explained by the influence of cristalisation level, decomposition of monolithic parts, strength anisotropy of monolith, different factors of subjective nature etc.

36


Figure 4.1. Histogram of distribution of uniaxial compressive strength of intact parts of carbonate rock massifs, Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008.

Figure 4.2. Histogram of distribution of uniaxial compressive strength of monolithic samples from the location of dam “Sveta Petka“, Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008.

37


Figure 4.3. Histogram of distribution of results of Point load test of intact parts of carbonate rock massifs, Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008.

Figure 4.4. Histogram of distribution of results of Point load test for samples from the location of dam “Sveta Petka”, Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008.

38


In comparation with world's experiences, in table 4.3., are given certain tests results for different carbonate rocks. And for these cases it is clear that uniaxial compressive strength vary in large interval. This certainly imply that it is very hard to generalise results from diffrent locations and it also indicates on necessary of specific examination for every location. Table 4.3. Overview of diapason of variation of values for uniaxial compressive

strength for different types of carbonate rock massifs (Hawkins, 1998).

Direction of testing

Sample 1 (MPa)

Sample 2 (MPa)

Sample 3 (MPa)

Average values (MPa)

Limestone type “Shelly“

Parallel Vertical Oblique

94 119 109

107 139 109

115 131 121

105 131 113

Crinoids limestone


154 169 144

157 163 159

153 152 179

154 161 161

Oolitic limestone


118 207 165

156 142 141

206 156 151

179 176 166

Micritic limestone


217 222 225

225 233 216

211 244 235

218 233 225

Dolomite Parallel Vertical Oblique

227 242 270

258 229 259

269 248 254

251 240 261

As it said earlier (chapter 3.3.7.1. orginal version of the thesis on bosnian language) results of Point load test (IS(50) – Index of point load strength for sample with diametar d = 50 mm) is not very easy to correlite with value of uniaxial compressive strength. Even when these correlations are made, correlation factors for carbonate rock mass of different geological age, mineral structure, weathered and macro and microstructural characteristics vary in large interval. So this factor for marbles of “Sveta Petka” dam is 16,832 (figure 4.5.), for limestone of round area of Ohrid in R. Macedonia 39,508 (figure 4.6.) etc. This variation of results indicates the need of determening these correlations for every specific location itself. Table 4.4. Correlation between results of point load test IS and uniaxial

compressive strength of monolithic samples by different autours. Correlations Author

σc=(14.5-27)Is(50) for limestone

σc=(5-10)Is(50) for cretaceous and porous limestone

Romana (1999)

σc=23xIs(50) (r2 =0.75) for limestone and marlstone

Tsiambaos & Sabatakakis (2004)

39


Figure 4.5. Correlation between results of Point load test IS(50) and uniaxial

comressive strength of monolithic samples for carbonate complex of “Sveta Petka” dam.

Figure 4.6. Correlation between results of Point load test IS(50) and uniaxial comressive strength of monolithic samples for limestone from the location of Ohrid, R. Macedonia.

40


In testing of uniaxial compressive strength on samples with different volumes that is different sizes, new knowledge about scale effect are confirmed by which increasing the size of examined samples, average value of uniaxial strength actually stays unchanged, and only dispersion of results is decreasing i.e. the variation of uniaxial strength. Variations of test results are not linear by increasing sizes of examined samples, but bordered curves of dispersion asymptomatically are getting near central curve. From figures 4.7. and 4.8. it can be seen that representative elementary volume REV is cca 1150 cm3, if testing of uniaxial strength for carbonate complex of “Sveta Petka” dam is considered, and for calcit marbles of Quarry K-2 for “Kozjak“ dam, R. Macedonia, is cca 554 cm3. It is volume under which size of sample has influence to the values of strength.

Figure 4.7. Influence of scale effect on uniaxial compressive strength of

monolithic parts of carbonate rock massifs from the locations “Sveta Petka” dam.

Figure 4.8. Influence of scale effect on uniaxial compressive strength of

monolithic parts of calcit marbles from the Quarry K-2 for “Kozjak“ dam, R. Macedonia.

41


In table 4.5. are given tests results of tension strength of monolithic samples of rock from the location on “Sveta Petka” dam, by Brasillian method, (Finale analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004.). We can see that values of tension strength vary, where at lower values are considered for wetheraed rock variations, and higher values on fresh monolithic parts. Table 4.5. Tests results of tension strength on monolith by Brasillian method for

“Sveta Petka” dam.

Num. Drill hole Depth (m) D/h

[mm]

D/h

[mm]

D/h

[mm]

σt1

[MPa]

σt2

[MPa]

σt3

[MPa]

1. PS–4 28.60 – 28.80 61.5/30.3 61.4/30.2 61.5/30.2 6.76 9.11 –

2. S–6 25.60 – 25.90 71.5/30.4 71.1/30.7 71.1/30.4 7.06 11.25 9.52

3. DB–8 22.60 – 22.90 71.6/30.0 71.5/30.6 71.4/29.9 14.39 15.15 7.56

4. DM–1 18.30 – 18.50 61.4/31.1 61.5/30.9 61.5/30.9 3.43 0.33 6.79

5. DM–1 25.10 – 25.30 61.5/31.2 61.5/31.7 61.5/31.6 3.95 7.28 6.30

6. DZ–1 30.40 – 30.60 71.5/31.9 71.5/31.9 71.5/31.8 3.66 – 3.45

On figure 4.9., is given histogram with represented categories of carbonate rock mass by RMR system from the location of R. Macedonia and Bosnia and Herzegovina.

Figure 4.9. Histogram with overview of categories of rock mass by RMR system, Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008.

42


Because of the comparation, on figure 4.10., is given histogram for different locations from the world by Bieniawski (1993). From figures it can be seen that with carbonate rock mass mostly appear II and III category (values for RMR=50-70, good or fair rock).

Figure 4.10. Histogram with overview of categories of rock mass by RMR system (Bieniawski, 1993).

On the end, on figures 4.11. and 4.12., are given tests results by shearing along the discontinuity, performed because of needs for making the disertation, and certain comparations with tests made in location of R. Macedonia.

Figure 4.11. Comparison of allowed (by Barton) and measured angle of friction along the discontinuity for location “Sveta Petka” dam and “Tunel 1” from main road “Demir Kapija-Smokvica“.

43


Figure 4.12. Diagrams of the dependence of normal-tangential stress with overview of comparations of laboratory results with empiric curve by Barton and Chobeau.

If we make brief analysis of all results, it is clear that all kind of methods (empiric, analytic and numerical) can be used, but every generalisation can make certain engineer's risk in selecting input results for further analyses. This again confirms that every separated project requires very carefull investigation and modelling, and certainly that all previous experiences should be used with a dose of engineer's skepticism for goal to achieve optimal solution. Based on these facts, in this work are shown some analyses that can be usefull in geotechnical modelling, but it certainly can not offer universal solutions.

44


4.2. Forming the possible physical models for analyses of carbonate rock masses by applicability of the Geological Strength Index (GSI) classification

4.2.1. Introductory notes One of the most used criteria for analysing shear strength and deformability of rock mass is the one suggested from Hoek and Brown, described in chapter 3.3.7.3. of original version of thesis on Bosnian language. Implementing of the Hoek-Brown criteria on rock mass needs knowledge of the following parameters:

• the uniaxial compressive strength of intact rock elements that make up the rock mass σci ;

• the constant mi that defines the frictional characteristics of the component minerals in these rock elements;

• the Geological Strength Index GSI which relates the properties of the intact rock elements to those of the overall rock mass.

Uniaxial compressive strength of intact elements of rock mass σci and material constant mi is determined by laboratory but if there are not exist specific tests or project is in preliminary study phase then these parameters estimate from tables (Tables 3.8. i 3.9., chapter 3.3.7.3.3. of original version of thesis on Bosnian language). Physical models defined through GSI classification exist in practice for evaluation of strength and deformability in general case (for every type of rock mass) and flysch rock complex with poor quality. The classification were worked through by Hoek, Marinos and Benissi, 1998; Marinos, P. and Hoek, E., 2000; Hoek, E., Marinos, P. and Marinos, V., 2005; Marinos, P., Hoek, E. and Marinos, V., 2006 and others. Based on these works and regarding other analysis in the disertation, there has been try to modificate GSI values for carbonate rock complex. This is shown in Tables 4.4.i 4.5.

Models are formed based on researches of carbonate rock complex on the location of „Salakovac“ and „Grabovica“ dams in Bosnia and Herzegovina and „Sveta Petka“ dam in Macedonia, but it can implement on other carbonate complex especially the ones of the Dinaric and Pelagonic massif, of course respecting local geological conditions, and mostly it includes following cases:

• quasichomogenic and isotropic area which is case with very jointed carbonate complex on the one side and massive carbonate complex on the other side;

45


• quasichomogenic and anisotropic area which mostly relates to

bedded limestone;

• combination of isotropic and anisotropic area with karstification condition namely for changed rock mass with clayey filling of the discontinuity.

4.2.2. Carbonate complex from the location of “Salakovac”, “Grabovica” and „Sveta Petka” dams

Wider area of the “Salakovac” dam is built with sediment rock mass, jurassics, cretaceous, neogenes and from quaternary age (Geological report, Book 1, Final design, “Salakovac” dam, 1970). By litho logical point of view this area is filled with all types of carbonate rocks from strong limestone, then dolomite limestone to dolomite itself. It is defined that left bank of dam’s profile is built of extremely karstificated upper-cretaceous light-grey limestone and that dam’s profile on left bank is touched by one fault zone of larger dimensions, which is filled with schistose clayey material, while other fault is in the area of same left inclination (see chapter 3.2., figures 3.5. and 3.8.). The right bank is formed of bedded dolomite and limestone of upper-jurassic. In the river's channel on the right bank were observed diluvial materials, while on the left debris materials and stone’s deposit. Tectonic of the area is very complex, whereas faults and associated discontinuity systems in rock mass are much more expressed than blister forms made because of the tangential movements. Main characteristic of jurassic sediments of the right side of valley is stratification. Thickness of layers is to 1 m with bland downing towards Neretva River. Inter-bedded filling is made of carbonate film or with a lot of compressed clayey mass. Discontinuities are less expressed and are located on the area with width of 1 m and space of 10 cm between themselves. The most of discontinuity intersects layers squarely. It can be concluded that these rocks are very compact with less discontinuities and it was asserted during the excavations. Unlike of the right side of valley, jointing and cavernous on the left side represent the most significant engineer-geological characteristics. Those discontinuities, which extend almost aright on the river, have usually clayey debris as filling or discontinuities are opened without filling. Geological structure of the area of „Grabovica“ dam contains limestone and dolomite of upper-jurassic age and terrace sediments with sandy grawel of late quarternary age (chapter 3.2., figure 3.9.). Bedded limestone and dolomite represent basic rock, while quarternary sediments are filling in large the flume of Neretva river and partly valley as covering. Although every type of rock exists on both sides of the valley, it can tell that the right side is bulit of dolomite and left of limestone more. Analysing engineer-gelological researchs as well as tectonic of wider area, geologists divided dam's profile with two

46


tectonic blocks. Division is made in fault along the Neretva river where is formed the cavern is in the zone of under footing of engine-room and dam. Shear's discontinuities are expressed on both blocks of cavern in width of 2,5 m. The discontinuities are thick and lie in parallel towards caverna with abrupt degradations. Left and right from shear' s discontinuities zone the rock is with more quality. Cavern is filled with dark grey and grey sandy clayey, clayey sand or grawel in higher tickness (Geological bases, Water power plant „Grabovica“ – Preliminary design, 1970). On the „Grabovica„ dam's profile, bedding is unequal, dolomite are more banked on the right bank where thickness of banks is approximates 1 m. On the left bank bedding is more expressed, but layer's thickness is also unequal, so there are more thick layers about 20-30 cm between banks. This can not be very favorable, because we must take note that planes of bedding are also planes of discontinuity, but in the other hand large banks make rock on an average more compact. Bedded areas are rough and without any interlayer fillings, so it can expected to have strong holding of rocks during the excavation, especially on the right bank where orientation of layers' incline is adverse from the incline of slope. Lower angles of inclination of layers (which are very blend than incline of slope) make the left bank more stable. The position of disconituity is much more significant for stability of rock during excavation. Jointing is significant characteristic of carbonate rocks from which the „Grabovica„ dam is built. On the right bank jointing extend with directions and inclines which are almost the same as direction and inclines of slope. This is very unfavorable if we see it isolated and only from the safety point of view for excavation of the foundation, otherwise there is no special meaning to it, considering that those directions are adverse to the directions of layers and in that way isolate rock in some miniature paralel-pipedic blocks which are then interlocked mutually, so that its' wider edge have softer incline, adverse to incline of slope. Some of these discontinuities are filled with dolomite grus. Bigger discontinuities are shear's discontinuities around cavern. On the left bank discontinuities are equally expressed as on the right bank. Every discontinuity has abrupt incline, that is angles of inclination. Regarding stability of excavation, these discontinuities do not have particular good position, because with bedding rocks are divided into blocks and its' longer axes have direction and decline compliant to slope. Concerning the profile of „Sveta Petka“ dam, it's geological structure is built by precambrium and rifecambrium metamorphic complex (massif marbles, poorly schistose marbles, carbonate schist, carbonate muscovite schist, etc.) Quartenary sediments are represented as aluvial sediments bulit with sand and marble's blocks and diluvial sediments built of dust sandy debris which is clayey or there are marble's blocks with size of 2 m as the ones noticed in the flumme of The Treska river (Finale analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004.). The left bank of profile is built with off white massive marbles, which are noticed under quartenary sediments of flumme of the river. The right side of profile is located on massive to less shisostose calcit and dolomite calcit marbles (chapter 3.2., figure 3.6.). Bedding is more expressed on the right bank than the left, with very abrupt declines from 80 to 85°. Noticed discontinuities can clasify on

47


faults and faults zones, discontinuities with lenght of 30 m, discontinuities with lenght from 5 to 30 m and lenght from 1 to 5 m. Fault's zones located on the right bank form it more defective in tectonic sense then on the left bank. On the right bank we noticed four and on the left three main discontinuities systems. An average space between these systems is from 600 to 2000 mm. Discontinuitiy' walls are mostly rough and rarely plane or smooth. Morphology of the walls is rather altered to smooth, altered and altered to serrate. The walls are fresh to less limonited and often there is carbonate fillings against walls, rarely we can see clayey.

4.2.3. Modification of GSI classification for carbonate rock mass As said above in point 4.2.1., within making analysis in this work, effort is made for modification of GSI classification for carbonate rock mass. Whereat were analysed cases of rock mass without presence of bedding or foliations (Table 4.4.). Table 4.4., shows formed physical models for carbonate rock mass defined through GSI values. Description corresponds carbonated complex, where were not present large caverns. Determining of GSI values is based on knowing of two fundamental parameters of geological process: structure of rock mass that is the blockiness i.e. interlocking of intact rock elements and surface condition of the discontinuities indicated by joint roughness and alteration. These two parameters are determined by visual examination. Beside this general description, model is supplemented with values of uniaxial compressive strength of monolithic parts of rock mass and interval of spacing of discontinuities, that makes approach and classification much easier. Based on lithological characteristics of rock, surface condition of the discontinuities, some box is chosen in Table 4.4. then we determine average value of GSI. You should not try to be very precisely. If we estimate that value of GSI is between 50 and 60, it is more realistic than to have GSI = 55. Different case or example is shown in Table 4.5., where can be seen formed physical models by GSI classification for carbonate rock mass with distinct bedding. And for this case GSI classification is based on knowing two main parameters of geological process of rock mass, that is thickenss of strata and surface condition of the discontinuities. In Table 4.5. is selected the most probable zone, where it can be seen cases of transversal isotropy, especially when there is dethacing along planes of bedding and when other systems of discontinuities are not expressed. In order to have better illustration of determination of GSI values, some characteristic examples of carbonate rock mass with analogus values of GSI are given in the Figures 4.13. (a)-(l).

48


Table 4.4. Possible physical models for carbonate rock mass defined by modificated GSI classification.

Note: 1) N/A is related to cases which are not able in practice.

2) Surface conditions of discontinuities:

Very good -Very rough, fresh unweathered surfaces

Good - Rough, slightly weathered surfaces

Fair - Smooth, moderately weathered and altered surfaces

Poor-Very smooth or highly weathered surfaces, occasionally slickenside with compact coatings or fillings with angular fragments

Very poor -Very smooth slickenside or highly weathered surfaces with soft clay coatings or filings

49


Table 4.5. Possible physical models for carbonate rock mass defined by modificated GSI classification for rock mass with distinct bedding.

Note: 1) N/A is related to cases which are not able in practice.

2) Surface conditions of discontinuities:

Very good -Very rough, fresh unweathered surfaces

Good - Rough, slightly weathered surfaces

Fair - Smooth, moderately weathered and altered surfaces

Poor-Very smooth or highly weathered surfaces, occasionally slickenside with compact coatings or fillings with angular fragments

Very poor -Very smooth slickenside or highly weathered surfaces with soft clay coatings or filings

50


(a) (b)

(c) (d)

(e) (f)

51


(g) (h)

(i) (j)

(k) (l)

Figure 4.13.

52


Figure 4.13. (a) Blocky, interlocked marbles with three discontinuity sets, blocks from 0,6-1,0 m, rough, slightly weathered surfaces, exit of tunnel near „Sveta Petka” dam, GSI = 65-70. Figure 4.13. (b) Blocky, interlocked massive limestone, exit of tunnel “Demir Kapija”, blocks from 0,6-1,5 m, rough, slightly weathered surfaces, GSI = 70-80. Figure 4.13. (c) Banked interlocked limestone on the location “Bijela”, on the left bank of Neretva river, main road Sarajevo-Mostar, banks 0,6-1,6 m, with rough surface, GSI = 60-70. Figure 4.13. (d) Blocky or very blocky interlocked limestone with three discontinuity sets, „Salakovac“ dam, blocks from 0,2-0,7 m, with rough surface, GSI = 55-65. Figure 4.13. (e) Bedded, interlocked dolomites and dolomite's limestone, railway tunnel near „Grabovica” dam, layers are with thickness 0,2-0,5 m, with rough surface, GSI = 55-60. Figure 4.13. (f) Very blocky, interlocked limestone near tunnel „Grabovica”, blocks from 0,2-0,6 m, moderately weathered and altered surfaces, GSI = 40-50. Figure 4.13. (g) Bedded marbles near „Matka” dam, moderately weathered surfaces, blocks 0,2-0,6 m, with GSI = 40-50, mostly isotropic transversal. Figure 4.13. (h) Blocky, interlocked marbles with three discontinuity sets, on the slope of road Rosoman - Pletivar, R. Makedonija, blocks 0,2-1 m with moderately weathered surface, GSI = 50-55. Figure 4.13. (i) Very blocky to disturbed limestone near “Salakovac” dam, on main road Sarajevo-Mostar, angular blocks 0,06-0,6 m, with smooth weathered surfaces, GSI = 30-40. Figure 4.13. (j) Very blocky limestone, on main road Sarajevo-Mostar, angular blocks 0,2-0,4 m, with weathered surfaces, GSI = 35-40. Figure 4.13. (k) Disintegrated, poor interlocked marble rock mass near fault which is mixture of angular and rounded pieces on the road of dam „Sveta Petka”, GSI = 20-25. Figure 4.13. (l) Totally disintegrated rock mass, partly under cavern, on the road of dam „Sveta Petka”, GSI = 15-20. Showed figures cover only part of possible forms of carbonate rock mass, but is very illustrated and can serve as example of estimating of GSI value for single types of these massives. After values for σci, mi and GSI are defined, as it is described above, it is value of the mechanical characteristics of rock mass next to determine by using Hoek and Brown failure criteria or some other criteria. It is very important to have in mind that there can be certain unequal with anisotropic mass. Anyhow with anisotropic mass, deformability and shear’s strength are usually expressed on different ways (Figure 4.14. (a) i (b)). To illustrate abilities for application of GSI system to bedded rock mass, hypothetical example of evaluation of parameters of shear's strength and

53


deformability for the same value of GSI is shown on figure 4.15. (a) and (b), whereat we have alteration of uniaxial compressive strength vertical and parallel on the plane of bedding.

(a) (b) Figure 4.14. Dependence of modulus of elasticity E from angle β beetwen stress

and plane of bedding for unaxial test by Pinto, 1970, Barla, 1974 (a), shear's strength on the „segments“ of rock mass in dependance from incline of strata and directions of shear, Langof, 1983 (b).

σci = 20 MPa σci = 40 MPa GSI = 45 GSI = 45 Mohr-Coulomb: Mohr-Coulomb: c = 0.264797 MPa c = 0.370872 MPa

ϕ = 47.3629 ° ϕ = 52.2126 ° Erm = 2236.5 MPa Erm = 4473 MPa Figure 4.15. Evaluation of shear's strength and deformability for values of GSI = 45,

and value of σci = 20 MPa parallel with plane of bedding (a), evaluation of shear's strength and deformability for values of GSI = 45, and value of σci =45 MPa vertical with plane of bedding (b).

54


Figures show that with different strengths, measured parallel and vertical on the bedding, parameters such as shear strength of rock mass and deformability are higher for cases that are vertical on the bedding. On the other hand this is not case with field's measuring. In fact, in practice we have different situation where deformability modulus for the case of parallel on bedding should be higher than for the case of vertical on bedding. This indicates on possible error if the classification systems and failure criteria are used automatically and uncritical in anisotropic mass, that should be considered while making practical analysis. Real solution can be found if classification or evaluation of GSI values is made separately for parallel and vertical cases to the foliation, and this could be one of directions in further research in these thesis.

4.2.4. Proposing of combination of GSI classification with state of karstification or with total value of carbonate rock mass porosity

In order to foreknow all mechanical and hydraulic characteristics of carbonate massives and especially karstificated rocks, it is very important besides quality of rocks to know state of karstication. For example, massives with similar geometrical aspects can show different mechanical and hydraulic characteristics. Example for aspects of natural constructions of carbonate massives with and without impressive karstification, is shown on figures 4.16. (a) and (b).

(a) (b)

Figure 4.16. Schematic hydro-geological profile of the terraine: I - hydro-geological insulator; AZ – aquifer zone; ZAA – zone above aquifer; CAZ – constant aquifer zone; ZF – zone of fluctuation; CP capillary zone; TZ – transitional zone; OBZ – ore-bearing zone (a), schematic hydro-geological profile of the terrain with karst aquifer (b) Lapčević, 1994, 2005.

55


From presentation above, it is clearly seen difference in hydro-geological characteristics of the terrain with karstification and without it. Level of aquifer zone with it’s height can effect on construction of underground works in particular, and it can be very different for one or other showed case. Also, underground water because of its’ chemical and physical activity has large effect on mechanical characteristics of carbonate rock mass. Underground water makes erosion and alterations on rock by circulating and chemical activity. It takes out particles and wash out discontinuities. So, while making proposition for prediction of state of karstification, there are considered known criteria elaborated for cases of decomposition of rock mass, which are also usable for soluble rock. Suggestion is that in practice is used classification of carbonate rock mass mainly on V classes by this parameter. Namely, in practice on consider VI classes, but in mine opinion classes V and VI for carbonate rocks can combine, Figure 4.17.

Figure 4.17. Classes of rock mass by criteria of level of surface decomposition or level of solubleness, by Little,1969; Fookes and dr., Dearman, 1974; Bell, 1993; Lapčević, 2005.

Classes V and VI are not typical for soluble rocks and in carbonate rocks would correspond „terra rosa” (red soil made by decomposition of carbonate rocks). Class IV – is very soluble rock, more than 50% of rock can be soluble and taken from massive and in discontinuities can be found small amount of residuum. Class III - is middle decomposed rock, near 50% of rock can be soluble, but rock's structure stays untouched. Class II – is poor decomposed rock, with opened discontinuities, and mass itself is poor soluble under discontinuities. Class I – is fresh indissoluble rock with compressed discontinuities.

56


Having this in mind and within analyses for dissertation, it appeared to me to combine classification system where it would consider parameters such as quality and state of karstification. This is shown on figure 4.18.

Figure 4.18. Proposed system of combination of rock's quality and state of

karstification (total porosity of rock mass n).

On figure it shows that theoretically speaking there are 25 combinations of classes, but some cases are not able in real (N/A), and some are rarely able (R/A). This means that there could select 13 main combinations. It should also note that in carbonate rock mass is very difficult to evaluate, in real, total porosity which is criteria for classification. In jointing rocks with less karstification there can be used parameters of jointing (RQD) for estimating of effective porosity (Figure 4.19.). In karstificated rocks with large caverns it is recommended to make relatively selecting of quasichomogenic zones by state of karsification. It is very important here to predict well size or volumen of possible interaction of zone with object. After defining possible zone it could follow selecting different quasichomogenic zones by combined system shown on figure 4.18.

57


Figure 4.19. Correlation between RQD parameter and effective porosity for profile

of „Sveta Petka” dam.

It is noted that zone of interaction is depending from size and characteristic of object and that for the same massive there would be different zones of interaction for bore hole, tunnel or dam. The figure 4.20. shows an example how to appreciate zone of interaction for same rock mass in dependence from size of engineer's object.

Figure 4.20. Overview of relatively selecting of zones of interaction for same rock

mass:(A) If bore hole is made, cavern is not in zone of interaction. (B) If larger tunnel is made, cavern is in zone of interaction.

Based on showings and analyses in this chapter, it can be shortly noted that phase of preparation of real physical model is main for any further analytic or numerical analysis. Mistakes made while selecting physical model, can drastically effect on conditions in practical construction, more than any complicated numerical simulation, which can be formal, if it is not based on real physical model.

58


4.3. Formation of analytical models for prediction of shear strength

parameters and deformability of carbonate rock masses for different GSI values

4.3.1. Introductory notes

In this chapter will be shown proposition of procedure of making analytical models that will be used for prediction of possible intervals of modulus of deformation D, cohesion c and friction angle ϕ, depending on uniaxial compressive strength of intact elements of rock mass σci for physical models of carbonate rock complex which are made through GSI classification(chapter 4.2.). These analytical models will be made using software RocLab. RocLab is software programme for determining parameters of shear strength and deformability based on Hoek-Brown failure criterion. In this way, determined characteristics of rock mass use as input information for numeric analysis, that need knowledge of material characteristics in the way to implement analysis of stability or stress and deformations. Applying this kind of procedure is specially adequate in cases where fields investigation for determine rock characteristics take time, money or where reliability of the investigation results is questionable. So in these cases regression models have been suggested to valuate characteristics of carbonate rock mass complex based on GSI classification.

4.3.2. Analytical models for prediction of modulus of deformation for carbonate rock masses based on empiric term Hoek, Carranza-Torres and Corkum, 2002 and Hoek and Diederichs, 2006

On the figure 4.21. are shown several examples of analytical models for prediction of possible interval of modulus of deformation D of carbonate rock mass depending on uniaxial compressive strength of intact part of rock mass σci for different values of GSI. For analysis of modulus of deformation D software Rocklab has used term (Hoek, Carranza-Torres and Corkum, 2002):

for σci ≤ 100MPa →

−

−= 40

10GSIci

rm 101002

D1Eσ (MPa),

(4.1.)

for σci > 100MPa →

−

−= 40

10GSI

rm 102D1E , (MPa)

59


In term 4.1. modulus of deformation is marked as Erm (rock mass modulus of deformation) based on English term and not D as it is case in some other parts of this work and generally in literature of our speaking area and on the other hand to distinguish it of term for D (disturbance factor). In this analysis it says that value of disturbance factor is D = 0, so during excavation or blasting there is minimal deformation to rock mass. In case where there is deformation of rock mass RockLab gives different values for term D from 0 to 1 depending on size of deformation itself and if there are slope or tunnel ect. (see Appendix 7.1.). For deformation modulus D double logarithmic (POWER) model of regression is formed which in general form is: b

ciD a= ×σ .

60


Figure 4.21. Analytical models for prediction of possible deformation modulus interval D of carbonate rock mass depending on uniaxial compressive strength of intact part of rock mass σci for different values of GSI.

Newer version of RockLab software uses term 4.2 which is given by Hoek and Diederichs, 2006, and which require knowing of deformation modulus of intact parts of rock Ei:

+−

+=−+ 11/)GSID1560((irm e12/D102,0EE , (MPa) (4.2.)

On figure 4.22. is shown model for prediction of deformation modulus D for solid marbles from location of dam “Sveta Petka” on the river Treska established by term 4.2.

Figure 4.22. Comparation of values of rock mass modulus of deformation Erm for solid marbles from location of dam “Sveta Petka” on the river Treska estimated by Hoek and Diederichs term (2006) for deformation modulus of intact parts of rock mass Ei = 20.000 MPa, Ei = 30.000 MPa, Ei = 40.000 MPa, Ei = 50.000 MPa i Ei = 60.000 MPa and distrubance factor Df = 0,8 with values of deformation modulus gotten by field investigation through hydraulic flat jack.

61


On monolithic samples taken from different depth of bore hole on the dam location we have got modulus of deformation by lab deformability test. These modulus of deformation are gotten in wide interval from 20.000 to 85.000 MPa (Finale analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004.). So in that way deformation modulus were valued by Hoek and Diederichs, 2006, for values of deformation modulus of intact parts of rock mass Ei = 20.000 MPa, Ei = 30.000 MPa, Ei = 40.000 MPa, Ei = 50.000 MPa and Ei = 60.000 MPa. From diagram on figure 4.22. it can be seen that results gotten by “in situ” investigation of deformation modulus with hydraulic flat jack mainly match with those valued by term 4.2. Some examples of specific test of modulus of deformation D and modulus of elasticity E to monolithic samples from dam „Sveta Petka” are shown on figures 4.23. (a) i (b).

(a)

(b)

Figure 4.22. Results of deformability tests on the samples from dam “Sveta Petka” in three cycles for sample S-6 (a), and sample DB-8 (b).

62


In case when we are not able to use values for deformation modulus of intact parts of rock mass Ei or to be able to prepare undisturbed samples for measuring values of Ei, and we do have lab defined values of uniaxial compressive strength of intact parts of rock mass σci then we can use term:

cii MRE σ⋅= (4.3.)

where MR is relation between deformation modulus of rock mass and modulus of intact part of rock mass irm EE / , which is represented by Table for different kind of rock mass (see Appendix 7.2.) Model formed by terms 4.2. and 4.3. for carbonate complex of dam “Sveta Petka” is given on figure 4.45., chapter 4.4. (average value of uniaxial compressive strength σci = 44 MPa, distrubance factor Df =0,8, which are values for dam „Sveta Petka“, value of MR as relation between deformation modulus of rock mass and modulus of intact part of rock mass irm EE / = 850 – value for the marble) and is showing good overlapping with curve which is gotten by extrapolation of the investigation results on the location of dam „Sveta Petka“.

4.3.3. Analytical models for prediction of cohesion for carbonate rock masses based on empiric term Hoek, Carranza-Torres and Corkum, 2002

On figure 4.23., 4.24. and 4.25. are shown examples of correlative dependence between cohesion c of carbonate rock masses and uniaxial compressive strength of intact part of rock mass σci for different values of GSI. For calculation of cohesion c of carbonate rock masses RockLab software used term (Hoek, Carranza-Torres-Corkum, 2002):

' ' a 13max 3max(1 2a)s (1 a)m (s m )ci b bci ci

c' a 13max(1 a)(2 a) 1 (6a m (s m ) ) /(1 a)(2 a)b b

ci

σ σ − σ + + − +σ σ =

σ −+ + + ⋅ + + +σ

(4.4.)

Where Rocklab manages to find out these kind of models for general case, tunnel and slope. The case is analysed where value of deformation factor Df = 0. If the tunnel is a case, then supposed height of overlay is 50 m and with slope it is taken that slope has 50 m height. Unit weight of rock is acquired as γ = 26 kN/m3 while material constant mi = 10. The most convenient regression

63


model that is gotten for cohesion of carbonate rock mass c is linear regression model: abc ci += σ .

Figure 4.23. Analytical models for prediction of possible interval of cohesion for carbonate rock mass in dependence of uniaxial compressive strength of intact part of rock mass σci for different values of GSI in general case.

64


Figure 4.24. Analytical models for prediction of possible interval of cohesion for carbonate rock mass in dependence of uniaxial compressive strength of intact part of rock mass σci for different values of GSI in general in case of tunnel with overlay of 50 m.

65


Figure 4.25. Analytical models for prediction of possible interval of cohesion for carbonate rock mass in dependence of uniaxial compressive strength of intact part of rock mass σci for different values of GSI in case of slope with height of 50 m.

66


4.3.4. Analytical models for prediction of angle of internal friction for carbonate rock mass based on empiric term Hoek, Carranza-Torres and Corkum, 2002

Angle of internal friction φ in general case does not depend on uniaxial compressive strength of intact part of rock mass σci.. On figure 4.26. is shown correlative dependence between angle of internal friction for carbonate rock mass φ and value of GSI. As we can see the most favorable regression model is linear regression model: b GSI aϕ = × + .

On figures 4.27 and 4.28 examples are shown of formed correlative dependence between angle of internal friction for carbonate rock mass φ and uniaxial compressive strength of intact part of rock mass σci for different values of GSI in cases of tunnels and slopes. RockLab software used term (Hoek, Carranza-Torres and Corkum, 2002) to estimate angle of internal friction φ for carbonate rock mass:

' a 13max6a m (s m )b b' 1 cisin ' a 13max2(1 a)(2 a) 6a m (s m )b b

ci

σ − ⋅ +σ −ϕ =

σ − + + + ⋅ + σ

(4.5.)

The case which is analyzed has value of disturbance factor Df = 0. Tunnel has supposed overlay height 50 m, while slope has height of 50 m also. Unit weight of rock mass was γ = 26 kN/m3 and material constant mi = 10. The most favourable regression model gotten for angle of internal friction φ is double logarithmic regression model (POWER): b

ciaϕ = ×σ .

Figure 4.26. Analytical model for prediction of angle of internal friction φ for

carbonate rock mass dependence of values GSI for general case.

67


Figure 4.27. Analytical model for prediction of angle of internal friction φ for carbonate rock mass in dependence of uniaxial compressive strength of intact part of rock mass σci for different values GSI in case of tunnel with overlay of 50 m.

68


.

Figure 4.28. Analytical model for prediction of angle of internal friction φ for carbonate rock mass in dependence of uniaxial compressive strength of intact part of rock mass σci for different values GSI in case of slope with height of 50 m.

69


As I know this is the first case where transformation function for parameters of deformability and shear strength were implemented in this way which can be seen as contribution of this dissertation. My opinion is that this approach can help to find faster choice of parameters for analysis in phase of preliminary design. Based on transformation functions in every specific case you can estimate parameters, which later in research and construction process it can be checked through specific field investigation.

70


4.4. Forming analytical models for establishing correlative dependences

between quality of rock mass (RMR and GSI), dynamic (vl and Edyn) and static characteristics (D and E) of carbonate rock masses based on the results of field tests on location “Salakovac” and “Grabovica” dam

4.4.1. Introductory notes Unlike analytical models formed in chapter 4.3. based on empiric terms Hoek, Carranza-Torres and Corkum, 2002, then Hoek and Diederichs, 2006, in this chapter we shall form analytical models based on tests results of rock mass combined static and dynamic methods on the locations of dams “Salakovac” and “Grabovica” on the Neretva river. While forming it, there has been used extrapolation method which was described in chapter 3.5. (original version of thesis on Bosnian language) so in this way the results gotten by singular measuring places were extended to the whole area of the dam. These models are especially significant because they were established by the characteristics of rock mass which are determined in situ investigation, while with the objects such as dams, deformability has the main or primary importance when analysing interaction of the system dam-rock mass.

4.4.2. Analytical models for carbonate complex from location of the “Salakovac” and “Grabovica”

On figure 4.30. are shown correlative dependences (analytical models) between static modulus of deformation D and elasticity modulus E for carbonate complex on the “Salakovac” dam location, figure 4.30. (a) and the “Grabovica” dam, figure 4.30. (b). The dependence is established for the “Salakovac” dam based on the results gotten through 4 vertical and 2 horizontal hydraulic flat jack located on the right bank of the Neretva river in bedded dolomites and dolomite limestone and on the left bank in karstificated banked limestone (Testing of deformability of rock mass by hydraulic flat jack, Finale project, The dam “Salakovac”, 1970). In establishing correlative relations for the “Grabovica” dam were used results of the testing with vertical hydraulic flat jack located in bedded dolomites and dolomite limestone on the right bank of the Neretva river (Testing of deformability of rock mass by experiments of pressure “in situ” in large scale). Analysing gotten results, as the most acceptable one was regression model bEaD ×= , so in that way for the relation of static deformation modulus D and elasticity E were gotten terms in following form:

9451,0587,0 ExD = (GPa) (Salakovac) (4.6.)

71


6471,06194,1 ExD = (GPa) (Grabovica) (4.7.)

Determination coefficients are R2 = 0,9453 and R2 = 0,8826 that apply on strong connection between examined parameters.

(a)

(b)

Figure 4.30. Correlative dependence between static modulus of deformation D and elasticity modulus E for carbonate complex from the “Salakovac” (a) and “Grabovica” dam (b).

Correlative relations between velocities of longitudinal elastic waves vl and dynamic modulus of elasticity Edyn for rock mass of “Salakovac” and “Grabovica” dam are shown on the figure 4.31. (a) i (b). Velocities of longitudinal elastic waves vl for carbonate complex of the “Salakovac” dam

72


are gotten by tests results with micro seismic method and based on them with using terms of elasticity theory dynamic elasticity modulus were estimated.

(a)

(b)

Figure 4.31. Correlative dependances between dynamic elasticity modulus Edyn

and velocities of longitudinal elastic waves vl for carbonate complex from the “Salakovac” (a) and “Grabovica” (b) dam.

In general velocities and dynamic elasticity modulus are smaller on the left than the right bank of the river, because of the strong karstificated limestone of the left bank. The highest values were gotten in flume itself by measuring velocity of longitudinal waves upright to the river. Diapason of velocities of longitudinal elastic waves vl is from vl = 2470 to 5050 m/s, which indicates that it is very undestrubed to very strong rock mass. Their values are

73


increasing through depth in general, where normal velocities are usually above 4000 m/s, and that indicates less undestrubed rock mass. Smaller values were gotten in the zone of reversed fault where velocities of longitudinal elastic waves vl are only vl = 2850 to 3000 m/s (Results of seismic examination in the profile of „Salakovac“ dam 1968-1972). Values of dynamic charateristics for dolomite and dolomite limestone of „Grabovica“ dam are gotten based on seismic examination between drilling on different ground levels. Diapason of velocities of longitudinal elastic waves vl is from vl = 2260 to 5830 m/s, which indicates that it is very undestrubed to very strong rock mass (Seismic examination in the profile of „Grabovica” dam, 1975). Analysing gotten results, as the most convenient is regression model

ldynbE a x v= , so in this way for the velocities of longitudinal elastic waves vl

and dynamic elasticity modulus Edyn are gotten terms in following form:

2,2105dyn lE 1,7533 x v= (GPa)(Salakovac) (4.8.)

1,9993dyn lE 2,2514 x v= (GPa),(Grabovica) (4.9.)

with determination coefficient R2 = 0,9908 and R2 = 0,9999 which indicates on strong connnection betweeen examined parameters. Correlations between static modulus D and E, velocities of longitudinal elastic waves vl as well dynamic elasticity modulus Edyn for rock mass of the “Salakovac” dam were determined based on comparative static-dynamic investigations that is seismic investigation around hydraulic flat jack (Figures 4.32., 4.33., 4.34. and 4.35. (a)), while establishing these kind of regression models for the „Grabovica“ dam were used investigation results by vertical hydraulic flat jack and seismic investigation between drill holes in different ground levels (Figures 4.32., 4.33., 4.34. and 4.35. (b)) Using direct correlation method following regression models and correlations coeffiecient are gotten:

4,2922lD 0,0099 x v= (GPa), R2 = 0,9025, (Salakovac) (4.10.) 1,7693

lD 0,6685x v= (GPa), R2 = 0,9412, (Grabovica) (4.11.) 4,5261

lE 0,0136 x v= (GPa), R2 = 0,9479, (Salakovac) (4.12.) 2,6233

lE 0,2965x v= (GPa), R2 = 0,9535, (Grabovica) (4.13.) 1,2961

d nD 0,0592x E γ= (GPa), R2 = 0,8601, (Salakovac) (4.14.) 0,9008

d nD 0,3032x E γ= (GPa), R2 = 0,9249, (Grabovica) (4.15.) 1,4596

d nE 0,0636x E γ= (GPa), R2 = 0,951, (Salakovac) (4.16.)

74


1,322

d nE 0,0964xE γ= (GPa), R2 = 0,9451, (Grabovica) (4.17.)

(a)

(b)

Figure 4.32. Correlative dependances between static deformation modulus D and velocities of longitudinal elastic waves vl for carbonate complex from the “Salakovac” (a) and “Grabovica” (b) dam.

75


(a)

(b)

Figure 4.33. Correlative dependances between static elasticity modulus E and

velocities of longitudinal elastic waves vl for carbonate complex from the “Salakovac” (a) and “Grabovica” (b) dam.

76


(a)

(b)

Figure 4.34. Correlative dependances between static deformation modulus D and

dynamic elasticity modulus Edyn for carbonate complex from the “Salakovac” (a) and “Grabovica” (b) dam.

77


(a)

(b)

Figure 4.35. Correlative dependances between static elasticity modulus E and

dynamic elasticity modulus Edyn for carbonate complex from the “Salakovac” (a) and “Grabovica” (b) dam.

78


Regression models dependences between velocities of longitudinal elastic waves vl and static elasticity modulus D with quality of rock mass by RMR system of the dams „Salakovac“ and „Grabovica“ are shown on figures 4.36. and 4.37. (a) and (b). Models are formed based on the classification of rock mass by RMR system around area of measured places previously given static and dynamic characteristics.

(a)

(b)

Figure 4.36. Correlative dependances between quality of rock mass by RMR

system classification and velocities of longitudinal elastic waves vl for carbonate complex from the “Salakovac” (a) and “Grabovica” (b) dam.

79


(a)

(b)

Figure 4.37. Correlative dependances between quality of rock mass by RMR

system classification and static deformation modulus D for carbonate complex from the” Salakovac” (a) and “Grabovica” (b) dam.

Analysing these results by direct correlation method are gotten following regression models:

1,6825lRMR 5,1772xv= , R2 = 0,9716 (Salakovac) (4.18.)

2179,1lvx4537,9RMR= , R2 = 0,9905 (Grabovica) (4.19.)

RMR0657,0ex1369,0D= (GPa) R2 = 0,9304 (Salakovac) (4.20.)

80


RMR0295,0ex6963,1D= (GPa) R2 = 0,9219 (Grabovica) (4.21.)

Determination coefficients that are gotten indicate on strong connection between examined parameters. Lower values of RMR and vl are referring on category of very poor to poor rock mass (20 – 30 RMR and vl mostly from 2470 – 3000 m/s for the “Salakovac” dam and 25 – 30 RMR and vl mostly 2850 – 3000 m/s for the „Grabovica” dam). Other class of parameters' value fits fair to good rock mass (42 – 69 RMR and vl from 3500 – 5050 m/s for the “Salakovac” dam, and 44 – 69 RMR and vl from 3500 – 5830 m/s).

4.4.3. Comparation of correlative dependences established based on investigation results from the “Salakovac” and “Grabovica” dam with existing correlations for carbonate complex

It is very significant for these kind of researches to make comparation of correlations established by our own investigation with correlative dependences that are established from different authors. It is recommended to make comparation with models gotten on similar terrain, firstly, we think on litho logical and structural-tectonic conditions and that results are gotten by the same or similar investigation method. Analytical models that are gotten from the locations of the “Salakovac” and “Grabovica” dams were compared with the ones from the location of „Sveta Petka” (Finale analysis of investigation and testing for concrete arch dam “Sveta Petka”, The Treska river, 2004.; Ilijovski, 2005; Krvavac, Jovanovski, Gapovski, Ilijovski, 2006; Jovanovski, Krvavac-Špago, Ilijovski, Peševski, 2008), and with results of Kujundžić and Petrović, 1980, because these results are the most comprehensive, the most representative and rather compatible. Correlative dependences between deformation modulus D and quality of rock mass RMR (GSI) for the “Salakovac”, “Grabovica” and “Sveta Petka” dams were also compared with known correlations of Serrafin and Perreira, 1983; Jovanovski and Gapkovski, 1998; as well Carranza-Torres, Corkum, 2002 and Hoek and Diederichs, 2006. The Figure 4.38. shows that for the same values of elasticity modulus E, values of deformability modulus D from the locations of the “Salakovac”, “Grabovica” and „Sveta Petka” dams are something lower then it shows in correlations by Kujundžić and Petrović. This can be explained because of specific litho logical structure and structural-tectonic characteristics mentioned locations as well as characteristics of applied investigation methods. By comparing correlative dependances between velocities of longitudinal elastic waves vl and dynamic elasticity modulus Edyn for the „Salakovac“, „Grabovica“ and „Sveta Petka” dams, we can conclude that there is good overlapping between formed curves (Figure 4.39.).

81


Figure 4.38. Comparation of correlative dependences between deformation modulus D and elasticity modulus E from the location of „Salakovac“ dam

9451,0Ex587,0D = (GPa) and “Grabovica” 6471,0Ex6194,1D = (GPa) with correlative dependences from the location of „Sveta Petka” dam 27,1Ex0327,0D = (MPa) and results of investigation of Kujundžić and Petrović 189,1Ex0925,0D = (MPa).

Figure 4.39. Comparation of correlative dependences between velocities of longitudinal elastic waves vl and dynamic elasticity modulus Edyn from the location of „Salakovac“ dam 2105,2

ldyn vx7533,1E = (GPa) and “Grabovica” dam 9993,1

ldyn vx2514,2E = (GPa) with correlative dependences from the location of “Sveta

Petka” dam 2,469dyn lE 0,000041x v= (GPa).

Figure 4.40. shows well overlapping between correlative curve for deformation modulus D and velocities of longitudinal elastic waves vl whiche were established for the location of „Sveta Petka“ dam and investigation results of Kujundžić and Petrović (midlle curve). Deformation modulus D values for the same velocities of longitudinal elastic waves vl of the location “Grabovica” dam are higher and of the location of “Salakovac” dam are lower with regard to middle curve by Kujundžić and Petrović. However these values are also approaching to middle curve by increasing velocities of longitudinal elastic waves vl and stay under upper and lower limited curve, which Kujundžić and Petrović established for similar locations.

82


Figure 4.40. Comparation of correlative dependences between deformation modulus D and velocities of longitudinal elastic waves vl from the location of „Salakovac“ dam

2922,4lvx0099,0D = (GPa) and „Grabovica” dam 7693,1

lvx6685,0D = (GPa) with correlative dependences from the location of „Sveta Petka” dam 373,3

l9 v103,4D ××= −

(MPa) (vl u m/s) and results of Kujundžić and Petrović 883,2lvx0171,0D = (MPa) –

upper limited curve, 773,2lvx0131,0D = (MPa) – middle and

2,848lD 0,0071x v= (MPa) – upper limited curve.

Correlative curves of elasticity modulus E and velocities of longitudinal elastic waves vl (Figure 4.41.) which were gotten for the location of “Grabovica” dam and curves of Kujundžić and Petrović (middle curve) show good overlapping. Values of elasticity modulus E for the same value of velocities of longitudinal elastic waves vl form for the location „Sveta Petka“ dam are higher, while values from the location of „Salakovac“ dam are lower regarding middle curve by Kujundžić and Petrović. Every correlative curve is under upper and lower limited curve also by Kujundžić and Petrović.

Figure 4.41. Comparation of correlative dependences between elasticity modulus E and velocities of longitudinal elastic waves vl from the location of „Salakovac“ dam

83


5261,4

lvx0136,0E = (GPa) and „Grabovica” dam 6233,2lvx2965,0E = (GPa) with

correlative dependences from the location of „Sveta Petka” dam 663,2l

6 vx106,3E −×= (MPa) (vl u m/s) and results of Kujundžić and Petrović 425,2

lvx558E = (MPa) – upper limited curve, (MPa) – , 332,2

lvx446E = middle and 395,2lvx264E = (MPa) – lower

limited curve.

The values which were gotten for deformation modulus D for equal values of dynamic elasticity modulus Edyn are higher for the location of “Grabovica” dam then for “Salakovac” dam (Figure 4.42.), which can be attributed to litho logical and structural-tectonic characteristics of the locations as well as to the characteristics of used investigaton methods.

Figure 4.42. Comparation of correlative dependences between deformation modulus D and dynamic elasticity modulus Edyn from the location on “Salakovac” dam

2961,1ndEx0592,0D γ= (GPa) and “Grabovica” dam 9008,0

ndEx3032,0D γ= (GPa). Figure 4.43. shows correlative depandances between static modulus E and dynamic elasticity modulus Edyn for “Salakovac” and “Grabovica” dam and correlative curves of Kujundžić and Petrović based on which we can conclude that there is well overlapping between formed curves.

Figure 4.43. Comparation of correlative dependences between dynamic elasticity modulus D and dynamic elasticity modulus Edyn from the location on “Salakovac”

84


dam 1,4596dynE 0,0636x E= (GPa) and “Grabovica” dam 1,322

dynE 0,0964x E= (GPa)

with correlative curves of Kujundžić and Petrović 1,172dynE 0,05x E= (MPa).

Correlative dependances between quality of rock mass RMR and velocities of longitudinal elastic waves vl show good overlapping for the “Salakovac”, „Grabovica” and „Sveta Petka” dams (Figure 4.44.).

Figure 4.44. Comparation of correlative dependences between quality of rock mass RMR and velocities of longitudinal elastic waves vl from the location on “Salakovac” dam 1721,18519,9 lvxRMR= and “Grabovica” dam 2179,14537,9 lvxRMR= with correlative

dependances from location of the „Sveta Petka” dam 4979,16848,5 lvxRMR= . On the figure 4.45. it can be seen, that there exist well correspondence between formed correlative dependances for deformation modulus D and quality of rock mass RMR from the location of „Salakovac“ dam, „Sveta Petka“ dam and correlations given by Jovanovski and Gapovski, and Jašarević. Also curve from the location of „Grabovica“ dam is approaching these curves by increasing values of RMR. On the other hand, term gotten by Serrafim and Perreira shows some higher values of deformation modulus D for equal value of RMR. Analytical models formed by term of Hoek, Carranza-Torres, Corkum, 2002, and Hoek and Diederichs, 2006, for the location on „Sveta Petka“ (average value of uniaxial compressive strength σci = 44 MPa, distrubance factor Df = 0.8, which are values of the „Sveta Petka“ dam, value of MR relation between deformation modulus of rock mass and modulus of intact part of rock mass irm EE / = 850 – value for marbles) show well correspondence with curve formed by extrapolation of „in situ“ gotten results on the location of „Sveta Petka“.

85


Figure 4.45. Comparation of correlative dependences between quality of rock mass RMR (GSI) and deformation modulus D from the location on “Salakovac” dam

RMRexD 0657,01369,0= (GPa), „Grabovica” dam RMRexD 0295,06963,1= (GPa) and “Sveta

Petka” 0,0703RMRD 0,1104 x e= dam with correlative dependences formed by Serrafim and Perreira 40/)10(10 −= RMRD (GPa), Jovanovski and Gapovski 9,3669,1 RMRD −= (GPa), Jašarević )081,0407,4( RMReD += (GPa), then Hoek, Carranza-Torres and Corkum (2002) and Hoek and Diederichs (2006) (average value of uniaxial compressive strength σci = 44 MPa, distrubance factor value Df = 0.8, which are values of the „Sveta Petka“ dam, value of MR relation between deformation modulus of rock mass and modulus of intact part of rock mass irm EE / = 850 – values for marbles).

4.4.4. Overview on possible polisemics on implementing formed dependances

Advantage of analytical models showed on item 4.4.2. and 4.4.3., is that can make predictions and extrapolations relatively quickly and accurately and in that form are very appropriate for practical using. Whereat, when prediction of the parameters is made in this way, it needs to mention that different in situ test are made with different levels of vertical stress, for different strengths, anisotropy and etc. And that is one of the reasons why we have some deviations with curves from different locations. With a view to see possible reasons of deviation on figures 4.46., 4.47., 4.48. and 4.49. are shown typical diagrams „pressure-deformation“, formed on base of tests with borehole dilatometer on profile of the „Sveta Petka“ dam.

86


Diagrams shown on figures 4.46., 4.47., 4.48. and 4.49. are basic for estimating of deformation modulus and elasticity modulus and imply not only to its' value than also to dependence of the modulus on pressure itself, so the point is rock mass „strengthhening“ or „softening“ regarding to pressure. This is very significant for making choice of numerical models for stress-strain behavior.

Figure 4.46. Diagram made by „in situ” test of the poor quality marbles (RMR=25-30) in bore hole DZ-1 on the profile of „Sveta Petka” dam.

Figure 4.47. Diagram D=f(p) made for the same interval where it can be seen decreasing of deformation modulus D with increasing of pressure p.

87


Figure 4.48. Diagram made by „in situ” test of marbles (RMR=25-30) in bore DZ-2 on the profile of „Sveta Petka” dam.

Figure 4.49. Diagram D=f(p) made for the same interval where it can be seen

increasing of deformation modulus D with increasing of pressure p.

88


Even for the same zone of investigation it is possible to get large differences in gotten values because of anisotropy of deformability (Figure 4.50.).

Figure 4.50. Comparation of deformation modulus D for two different directions on same depth with different levels of pressure (differences imply on anisotropy of deformability).

It is clear that if we combine empiric an field's methods, we can succesfully cover a lot of cases which are important for project analysis, but it is also clear that examples on the figures always have to be carefully used, reexamined and attentively engaged in geotechnical models. The conclusion is that verification of the parameters on higher levels of projection or construction is very significant aspect that has always have to be in mind while analysing capital objects.

89

5. GEOTECHNICAL MODELS AND METHODOLOGY FOR FORMATION OF CONCEPTUAL, ANALYTICAL AND NUMERICAL MATRIX OF INTERACTION

5. GEOTECHNICAL MODELS AND METHODOLOGY FOR FORMATION OF CONCEPTUAL, ANALYTICAL AND NUMERICAL MATRIX OF INTERACTION 5.1. General information about formation of conceptual model of

interaction

It is generally known that before any engineering activity rock mass is found in certain geological conditions and it has the appropriate properties and conditions. Based on the defined properties and the state of the geological environment a corresponding geotechnical models can be formed in order to assess the optimal conditions and possibilities for application of specific supporting systems, technology and dynamic plan for execution of excavation and meliorative activities. On the other hand, the influence of engineering activities on the geological environment largely depends on the technology applied and the works dynamics, final characteristics of the facility and its contents. The area in which the interaction between the environment and structure is pronounced is called zone of interaction or zone of co-action. There are different forms of interaction but not all of them are always present in particular case and do not have the same importance. Different structures will cause different interaction in the same geological environment and vice versa, i.e., the same structures will cause different forms of interaction in different environments. Magnitude of stress-strain interaction is caused by stress redistribution and primarily depends on the rock mass properties, natural strain, size of the structure and load transferred from structure to the neighbouring rock mass. In this context, on the basis of defined geotechnical models one needs to make forecasts of possible forms of interaction and it is sure that all three components of complex geotechnical model i.e. model of the natural geological environment, model engineering activities and stress-strain model can be used. Having in mind all the analysis in the dissertation, it is obvious that there is a huge number of possible interactions. This knowledge needs to be regulated and selection of relevant design parameters essential for a given problem needs to be established. Further text provides examples of analysis based on the so-called interaction matrix concept (Hudson, 1993, 1997). The basic idea is that the main characteristics of the rock mass and structure are placed in diagonal of the matrix. Interaction matrix by its nature can be conceptual, analytical or numerical. One example of a conceptual matrix is given in the thesis introduction in the

90


form of interaction scheme which generally presents material impacts of jointed carbonate rock massif, the state of karsticifaction and the characteristics of monolithic rock pieces on some aspects of the execution of an engineering object (Figure 1.1.). Practical aspects of problem analysis are met during work execution on any engineering strucure in calcareous areas and to illustrate the concept, in Figure 5.1., 5.2. and 5.3. some examples of the conceptual matrix for the underground excavation, foundation and slopes are given. Only three main parameters in the diagonal of matrix are shown and each of them in fact is to be treated as a group of parameters.

Figure 5.1. Conceptual martix of interaction with three main elements in diagonal for the case of underground excavation (partially modified according to J.A.Hudson, 1993).

91


Figure 5.2. Conceptual martix of interaction with three main elements in diagonal

for the case of foundation on the rock base (partially modified according to J.A.Hudson, 1993).

.

92


Figure 5.3. Conceptual martix of interaction with three main elements in diagonal

for the case of the slope (partially modified according to J.A.Hudson, 1993).

Number of elements that can be included in the main diagonal of the matrix can be increased. For example, the analysis commonly includes the impact of groundwater. It is essential to include the main influences that arise from the condition and properties of geological environment on the design solutions from one side and to identify potential impact that structure has on the environment on the other. This is a separate problem with any engineering building, where the conceptual matrix is of great use to set the problem in a proper manner, and as such, the issue is further analyzed by analytical or numerical procedures. Further analysis in the dissertation provides illustrative examples are given in order to apply certain concept to solving the theoretical-practical problems.

93


5.2. Formation of the numerical model and interaction matrix for

carbonate rock massifs with Examine2D software package

5.2.1. Introductory remarks Application of numerical models in mechanics is common in recent time, for solving stress-strain behavior around the underground openings or under the foundation of large surface structures. Analysis conducted in this chapter has two aims. First objective was to show numerical models can be used as a useful tool for designers. With the software package Examine2D, the state of induced stress around the hydrotechnical tunnel in carbonate rock complexes has been modeled and the results were compared to the site investigation results conducted aruound the same tunnel. Numerical model can be used in this way to offer a picture of rock masiff stress behavior. This is particularly useful in the initial stages of design since coslty and extensive site investigation can be avoided. Second objective was to form the interaction matrix between carbonate rock complexes as geological medium and engineering activity by variation of properties and condition parameters of carbonate complex, radius of hydrotechnical tunnel as construction parameter, i.e. engineerng activity as well as variation of stress states with software package Examine2D in order to examine the rock mass reaction, its stress-strain behaviour and in accordance with the obtained information define the best technical solution of a given task. Results of such analysis may not be the only ones that need to be relied on when making practical decisions, but should serve to clarify issues which cause doubt and uncertainty of designer. Assumptions about homogeneity, isotropy and linear elasticity of the material this software package is based on, make designers take results of such analysis with caution and it also means that the decision-making during the design must use the experience, knowledge, and engineering evaluation. However, when these models are formed on the basis of reliable input parameters and when mathematical formula that realistically reflect the behavior of rock mass in real conditions are used, they can be very useful in the design process in optimizing the geometry of the excavation, the preliminary costs for supporting and ensuring the excavation, achieving the required strength factors, or the simulation of extensive and costly site testing in the initial stages of design.

5.2.2. Applicated field testing methods of stress state around hydrotecnical tunnel in carbonate rock massifs

Starting from the accepted assumption that the propagation velocities of elastic waves are the function of stress in rock massif the experiment was condiced by Langof (Langof, 1977; Langof and Bijeljanin, 1985) in order to determine the correlative relation between „in situ“ and induced stress states

94


and wave propagation velocity and therefore to find the „in situ“ stress value by the combination of static and dynamic method. Tests were conducted for hydrotehnical tunnel Fatnica, Bileća-accumulation in the hydropower system on Trebišnjica, Bosnia and Herzegovina (Geotechnical Department of Civil Engineering Faculty in Sarajevo). Tunnel diameter is D = 7.10 m is found in the compact banked limestones and dolomites at a depth of 50 m. This study used a combined static Tincelin-Mayer method for determining the induced stress with two mutually perpendicular hydraulic flat jacks and geophysical methods of „ventilation“ between the two pairs of parallel and mutually perpendicular drill holes at a distance of 0,80 m deepth of 6,0 m. This static method obtains a induced stress in two directions (σ sV i σ sH) while the method of „ventilation“ between the drill holes the wave propagation velocities i.e. „in situ“ in two directions (vertical and horizontal) before creating the cut in the deeper zones (v pV i v pH) and after creating the cut (v sV i v sH), during the action load on the hydraulic flat jacks (Figure 5.4. (a) and (b)).

(a) (b)

Figure 5.4. Velocity diagram along the horizontal (a) and vertical (b) hydraulic flat

jack in limestone, where the average velocity: (1) before cut formation, and (2) during the load σ = σ sV i σ sH.

Correlation is estabilished between iduced stress in two directions (σ sV i σ sH) and „in situ“ stress in two directions (σ pV i σ pH) throught the relation of „in situ“ (vpV i v pH) and induced (v sV i v sH) velocities i.e.:

pvp s V 1,3MPaV V svVσ = σ = (5.1.)

pvp s H 2,0MPaH H svHσ = σ = (5.2.)

95


From the obtained “in situ” stress results according to expressions 5.1. i 5.2. the coffeicient of lateral pressure is obtained K = 1,54 ( 54,13,1/0,2/ ==P

VPH σσ )

for particular case of the tunnel in compact limestones and dolomites. Along with the use of data from the static – dynamic model and theoretical relations for the stress distribution according to the theory of elasticity, distribution of radial σr and tangential σt stress with zones around the tunnel is obtained (Figure 5.5.).

Figure 5.5. Lines of stress distribution and characteristic zones: (1) released, (2)

increased and (3) “in situ” stress, obtained by static-geophysical methods in the limestone – dolomite massif.

5.2.3. Numerical model of stress-strain behaviour around hydrotechnical tunnel in carbonate rock massive formed with aid of software package Examine 2D

Based on the example shown aboveand, and by the appplication of the principle of conceptual interaction matrix from figure 5.6., example of interaction matrix is given with main tree components: elasticity modulus which depends on geological environment characteristics, tunnel diameter as

96


a construction component i.e. engineering activity and stress state as interaction between rock massif and structure.

Figure 5.6. Matrix of interactions with three elements in the diagonal for the case

of hydrotechnical tunnel in compact limestone and dolomite. Further text provides series of analysis using software package Examine2D.

5.2.3.1. Software package Examine2D Examine2D is software package designed for solving two dimensional straignt stress states based on the boundary elements method and is used in stress and strain elastic analysis in underground or surface excavations (www.rocscience.com). Straight stress state means that the modelled excavation had infinite length perpendicular to the plane of the section being examined. Boundary elements method is a method which only the boundary of excavation is divided on elements while the inner part of the rock massive is mathematically defined as the infinite continuum. Analysis in this sofware assume that material is:

• homogeneous;

Modulus of elasticity Ex i Ey (MPa)

1 1

The value of the elasticity modulus affects the distribution of induced stress states 1 2

The value of the elasticity modulus affects the possible tunnel openings to be build without supporting system or it affects the choice of the support 1 3

Size of the excavation affects the radius of plasticisation zone and the depth of the relived stresses zone 3 1

Diameter of tunnel

3 3

The size of the excavation affects the way of forming the induced stress state 3 2

The value of the induced stress affects the deformability 2 1

Stress state

2 2

High stress level affects possible failures around the excavation 2 3

97

http://www.rocscience.com/


• isotropic or transverse isotropic; • linearly elastic.

Radial σr, tangential σt and shear τrt stress and displacements in radial ur and tangential ut direction around a circular hole (Figure 5.7.) calculated with Examine2D coincide with the analytical solution of Kirsch's (Brady and Brown, 1993; Poulos and Davis, 1974):

+−−−−+= θ

σσ 2cos)341)(1()1)(1(

2 4

4

2

2

2

2

ra

raK

raKv

r

+−+++= θ

σσ 2cos)31)(1()1)(1(

2 4

4

2

2

raK

raKv

t

−+−= θ

στ 2sin)321)(1(

2 4

4

2

2

ra

raKv

rt (5.3.)

−−−−+−= θυσ

2cos)1(4)1()1(4 2

22

raKK

Gra

u vr

+−−−= θυσ

2sin)21(2)1(4 2

22

raK

Gra

u vt

where are:

)1(2 '

'

ν+=

EG - shear modulus for straignt stress state and isotrophy

condition;

'2

EE1

=− ν

, ν

ν−

=1

'v ,

E - Young elasticity modulus,

ν - Poisson's coefficient, a - circular hole diameter,

r i θ - radius vector and angle in at which define the position of the point of interest for stress calculation,

K - the coefficient of lateral pressure.

98


Figure 5.7. Geometric model.

5.2.3.2. Results and comments Figures 5.8. and 5.9. show diagrams of radial and tangential stress in the sides of hydrotechnical tunnel of diameter D = 7,1 m, depth d = 50 m in compact limestones and dolomites formed with Examine 2D software package for the coefficient of later pressure 1,54. From radial and tangential stress diagram it can be see that numerical analysis and site test values overlap and the zone of „in situ“ stress appers to be present at the distance of approximately 8 m from the edge of excavation i.e. undesturebed elastic zone where radial stress starts to asimptoticialy approach the value of horizontal stress σH = 2,0 MPa and tangential stress approaches values of vertical stress σV = 1,3 MPa. Since the calculations are made on the basis of elasticity theory, tangential stress diagram does not strat at zero, i.e. coordinate begining which and the case according to the theory of plasticity where fact that after the excavations stress desintegration occurs and as a consequence of this zone of released stress or disturbed zone is formed between the excavation contour and plastic zone - zone of increased stress is considered. This zone usually cannot take over the stress and tangential stress values are reduced to zero.

99


Figure 5.8. Diagram of radial stress on side of (θ = 0°) hydrotechnical tunnel

(diameter D = 7.10), at a depth of 50 m in the compact banked limestones and dolomites for the lateral pressure coefficient of 1.54, obtained with the software package Examine2D.

100


Figure 5.9. Diagram of tangential stress on side of (θ = 0°) hydrotechnical tunnel (diameter D = 7.10), at a depth of 50 m in the compact banked limestones and dolomites for the lateral pressure coefficient of 1,54, obtained with the software package Examine2D.

Figure 5.10. shows a diagram of the total displacements of rock massifs in the sides (θ = 0°) and crown (θ = 90°) of the tunnel at a distance of up to 10 m from the edge of the observed hydrotechnical tunnel. It should be noted that the displacements calculated using Examine2D are elastic displacements which in reality represents only a very small component of the actual displacements that occur around the excavation. However, elastic displacements can be a useful indicator of the general trend of deformation.

101


Figure 5.10. Total displacement in the sides (θ = 0°) and crown (θ = 90°) tunnel at

a distance of up to 10 m from the edge of the observed hydrotechnical tunnel.

(a) (b)

Figure 5.11. Deformation vectors and deformed contour (a) and stress trajectories

(b) hydrotechnical tunnel D = 7.10 m in compact banked limestones and dolomites, the lateral pressure coefficient of 1.54, Young modulus of elasticity in the direction of x and y axes Ex = 14.500 MPa, Ey = 10.000 MPa, Poisson's coefficient µ = 0,25 uniaxial compressive strength of intact part of rock massif σci = 75 MPa, GSI = 50 and mi = 12, which are the data obtained for the rock massif at the location of hydrotechnical tunnel.

Figure 5.11. (a) shows the deformation vectors representing the directions and relative magnitude of elastic deformation. In the figure one can also see

102


deformed shape of the given hydrotechnical tunnel. In Figure 5.11. (b) one can see trajectory of the stress in the model, which represents the directions and relative magnitude of the principal minimum and maximum stress. Diagrams were created for the lateral pressure coefficient of 1.54, Young modulus of elasticity in the direction of x and y axes Ex = 14.500 MPa, Ey = 10.000 MPa, Poisson coefficient µ = 0,25, uniaxial compressive strength of intact part of rock massif σci = 75 MPa, GSI = 50 i mi = 12, which is the data obtained for rock massif on the location of hydrotechnical tunnel.

5.2.3.3. Analysis of influence in the matrix of interaction Forming the matrix of interaction as show in Figure 5.6. between carbonate rock complex as geological medium and engineering activity and therefore the assesment of rock mass reaction i.e its stress-strain behaviour is conducted in three steps: Step I First of all the variation of Young modulus as rock massif parameter is done for following values:

(I) option: Ex = 10.000 MPa, Ey = 10.000 MPa, GSI = 30, σci = 50 MPa;

(II) option: Ex = 14.500 MPa, Ey = 10.000 MPa, GSI = 50, σci = 75 MPa;

(III) option: Ex = 35.000 MPa, Ey = 24.000 MPa, GSI = 80, σci = 100 MPa, in order to asses the influence of this geological environment parameter on rock massif stress behavior and possible tunnel excavation with or without supporting i.e. its infuence on technical procedure which will be applied during excavations. It was done with hydrotechnical tunnel diameter D = 7,10 m, „in situ“ stress state σ pV = 1,3 MPa, σ pH = 2,0 MPa i.e. lateral coefficient pressure K = 1,54. From the chart in Figure 5.12. (a) and (b) one can see how the change in elasticity modulus affects distribution of induced radial and tangencijal stress, where by the induced stress increase with the decrease of the module i.e. increase of deformability of rock massif. This change will influence the possible tunnel openings which can be constructed without support or with a choice of support. This can be seen on the Figure 5.13. (a), (b) i (c) which shows strength factor diagrams and failure trajecotry of tunnel side (θ = 0°) up to the 10 m depth around the considered hydrotechnical tunnel for different options of elasticity modulus. Strength factor is a quantitative measure of relationship between strength of rock massif and induced stress, which allows user to define failure criteria for rock massif. If it is less than 1, then the failure in the rock massif under given stress will occur. Figure shows that the failure zone is higher as rock massif has smaller module, and (I) option should be eliminated due to the large failure zone, (II) option will require adequate

103


support to be provided, and in (III) option the tunnel can be executed without support.

(a)

(b)

Figure 5.12. Distribution of radial (a) and tangentijal (b) stress in a hydrotechnical

tunnel sides (θ = 0°), diameter D = 7.10, at a depth of 50 m, the coefficient of lateral pressure of 1.54 and different values of modulus of elasticity obtained with the software package Examine2D.

104


(a) (b)

(c) Figure 5.13. Diagrams of strength factors and failure trajectories in the sides

hydrotechnical tunnel (θ = 0°), diameter D = 7.10, at a depth of 50 m, the coefficient of lateral pressure of 1.54, and the option (I) (a), option (II) (b) and option (III) (c) the values of elasticity modulus were obtained with a software package Examine2D.

Step II In the second step hydrotechnical tunnel diameter change was performed for the following values: (I) option D = 5,0 m; (II) option D = 7,1 m; (III) option D = 10,0 m, where the modulus of elasticity was Ex = 14.500 MPa, Ey = 10.000 MPa, the value of GSI = 50, uniaxial compressive strength σci = 75 MPa, „in situ“ stress conditions σ pV = 1.3 MPa, σ pH = 2.0 MPa , i.e., coefficient of lateral pressure K = 1.54.

105


(a)

(b)

Figure 5.14. Distribution of radial (a) and tangentijal (b) stress in a hydrotechnical

tunnel sides (θ = 0°), for the elastic modulus Ex = 14.500 MPa, Ey = 10.000 MPa, GSI = 50, uniaxial compressive strength σci = 75 MPa, coefficient lateral pressure coefficient of 1.54 and different values of the tunnel diameter D obtained with software package Examine2D.

The aim was to consider the impact of the excavation size on the way of forming the induced stress and deformability of rock massif through plastification zone radius and the depth of the released stress zone. The figure 5.14. (a) and (b) shows that with increasing diameter of the tunnel leads to the decrease of radial and the increase in tangential stress and increase of the radius of the plastification zone and depth of the released stress zone.

106


Step III In the third step, the induced stress conditions around the tunnel (Figure 5.15. (a) and (b)) over the three options of „in situ“ stress conditions, i.e. lateral pressure coefficient K: (I) option K = 0,0; (II) option K = 1,5; (III) option K = 2,0, for the Young elacticity modulus Ex = 14.500 MPa, Ey = 10.000 MPa, value GSI = 50, uniaxial compressive strength σci = 75 MPa and hydrotechnical tunnel diameter D = 7,10 m.

(a)

(b)

Figure 5.15. Distribution of radial (a) and tangetial (b) stress in the sides of hydrotechnical tunnel (θ = 0°), for Young module Ex = 14.500 MPa,

107


Ey = 10.000 MPa, value of GSI = 50, uniaxial compressive strength σci = 75 MPa, the tunnel diameter D = 7.1 m, and different values of the lateral pressure coefficient, obtained with a software package Examine2D.

Figure 5.16. Displacements in the sides of hydrotechnical tunnel (θ = 0°), the

elastic modulus Ex = 14.500 MPa, Ey = 10.000 MPa, value of GSI = 50, uniaxial compressive strength σci = 75 MPa, tunnel diameter D = 7.1 m and different values of the coefficient of lateral pressure obtained with a software package Examine2D.

The impact of changes of induced stress on rock massif deformability and possible failure in excavations is analised. The figure 5.15. (a) and (b) shows the growth of radial and tangential stress with increase in lateral pressure coefficient K which leads to increased displacements, figure 5.16. i.e. deformation, and consequently to increase in failures during tunnel excavations. Analyzed variations clearly indicate complexity of the model formation and large number of possible influences between the parameters in the matrix, but at the same time it also points out the great benefits of such approaches because from the potential impacts we can obtain knowledge on the expected behavior of a complex system of rock-structure. It is a great skill of researchers to choose the most important parameters for analysis, to set a good concept and to make a realistic analysis.

108

6. CONCLUSIONS AND RECOMMENDATIONS

6. CONCLUSIONS AND RECOMMENDATIONS Having in mind modern tendencies in geotechnics, the aim of this thesis was to contribute in the field of geotechnics modelling trough proposal of modelling methodology of carbonate rock massif, as a very wide foundation for construction of complex engineering structures such as dams, tunnels, underground excavations, and the like. During the elaboration of this methodology, we have used past researches of other authors in the field of engineering-geological and geotechnics modelling, extrapolation of parameters, scale effect, classification of rock massif, method of testing the conditions and qualities of rock massif and the like, in one hand. In the other hand, however, as a basis for methodology development were used results of series of geological, engineering-geological and laboratory and field geophysical and geomechanics researches on the locations of dams “Salakovac” and “Grabovica” on the Neretva River, Bosnia and Herzegovina, and “Sveta Petka” on the River Treska in Republic of Macedonia. Some of the proposed procedures of modelling can be considered as modification of already existing methods, while the others were given for the first time. When it comes to scientific contribution of thesis and subject originality, following should be emphasized: Analysis, statistic processing and comparison of researcher’s own and

gathered laboratory tests of the most important engineering-geological and geomechanics characteristics of carbonate rock massifs from the Balkan Peninsula.

Formation of physical models for carbonate rock complexes, which are defined through “Geological strength index“ (GSI) classification, inspired by similar models for some other rock massifs formed by Hoek, Marinos and Benissi, 1998; Marinos P. and Hoek, 2000; Hoek, E., Marinos, P. and Marinos, V., 2005; Marinos, P; Hoek, E. and Marinos, V., 2006.

These physical models were formed on the basis of geological and engineering-geological researches of carbonate rock massifs on the locations of dams “Salakovac” and “Grabovica” on the Neretva River, Bosnia and Herzegovina, and “Sveta Petka” on the River Treska in Republic of Macedonia. However, they can be applied on other carbonate complexes, primarily on carbonate rocks of Dinara and Pelagonija massifs, having in mind local geological circumstances.

Formed physical models mainly include following cases: 1. quasi homogenous and isotropic environment what is the case with

extremely jointed carbonate complexes and massive carbonate complexes in the other hand;

2. quasi homogenous anisotropy environment, which refers to mainly bedded limestone;

3. combination of isotropic and anisotropic environment with the karstification condition, and for weathered rock mass with clay filling of joints.

109


GSI values determined through here suggested physical methods together with unaxial compressive strength intact parts of rock massif σci, and material constant mi represent input parameter with determining shearing strength through Hoek-Brown failure criteria.

In order to predict total mechanical and hydraulic behaviour of carbonate massifs and particularly in the cases of rock’s karstification, besides the quality of rocks, it is definitely important to know also the karstification condition. Therefore, the thesis proposes combination of GSI classification with the karstification condition, which quantitative measure is total porosity of rock massif.

Forming of analytic models for prognosis of possible intervals of deformation module D, cohesion c and angle of friction ϕ depending from the unaxial compressive strength of intact parts of rock massif σci for physical models of carbonate rock complexes defined through GSI through programme packet RockLab.

Based on these, according to my knowledge, for the first time formed transformation functions, we can obtain input data for numerical analysis, which demand knowing of material characteristics in order to carry out the stability analysis or stress and deformations. It is particularly visible in preliminary phases of structures design when field investigations for determination of rock massifs’ characteristics demand time, expenses, or the reliability of results of these tests will be in doubtful.

Defining of original dependences between quality of rock massif (RMR and GSI), dynamic (vl and Edyn) and static characteristics (D and E) carbonate rock massifs for locations of dams “Salakovac” and “Grabovica” on Neretva River, that could be valid for carbonate area of entire Bosnia and Herzegovina.

Comparison of these dependences with forming models for carbonate complexes by other authors and for other locations for their verification.

Forming of numerical model and interaction matrix for the purpose of analysing stress-strain behaviour around hydraulic tunnel in carbonate rock complexes, with the help of programme packet Examine 2D.

Comparison of numerical model with results of field investigation of induced stress condition around the same tunnel.

Because of its verification, suggested methodology of carbonate rock massif modelling with time must experience critical re-examination in terms of possibilities to apply it on carbonate complexes on other locations, other facilities, and in specific phases of designing and the like. All this will create preconditions for its modification, further development. However, it will open doors and possibilities for further researches, considering that it is practically impossible to exhaust this scientific theme with only one paper. Regarding the critical review of proposed methodology and directions of further researches, we should emphasize the following: Due to different way of displaying deformability and shearing strength in

different direction when analysing anisotropic masses, physical models that are defined through „Geological strength index“ (GSI) classification

110


should be applied with caution and certain amount of critics, where the real solution can be found if the classification, that is assessment of GSI values is carried out separately for each cases and parallel and transversally for foliation that implies one of the possible ways of further researches in this area.

Analytical models for prognosis of possible intervals of deformation modules D, cohesion c, and the angle of friction ϕ formed through programme packet RockLab are useful as input data for numerical analysis with smaller objects. However, its application with big and important facilities, susceptibility constructions where it is necessary to undertake adequate stability checks such as high arch and gravity dams, limited to preliminary calculating and initial phase of designing, while for the phase of preliminary design and final design, when the data reliability is very important, we must carry out field investigations. The directions of further researches would primarily refer to gathering as many data on parameter D, c and ϕ values for carbonate complexes obtained trough field and laboratory measurements and their comparison to outcomes obtained via these analytical models, as it is done for deformation modulus from the location of „Sveta Petka“dam. In addition, we can conclude that the process of modelling must be available and harmonized with research and design phases. It is normal that in initial phases we use more simple approaches, which meet current quality and quantity of available data. Results of such kind of initial models, with complex facilities can indicate to need for new data and they enable re-interpretation of existing data, what in the other hand influences the improvement of models or leads to new ideas for new model types.

Original correlation between quality of rock massif, static and dynamic qualities of carbonate complexes on localities of “Salakovac” and “Grabovica” dams on Neretva River as it is already mentioned, can valid for carbonate complexes of the entire Bosnia and Herzegovina, but they cannot be routinely applied on a new specific location. The differences which are appeared during the comparison of models obtained on different locations are results of different lithology and structural and tectonic conditions on specific locations, effect of anisotropy, level of load that is used during the test, way of testing, volume of rock massif that was a subject to the test, and the like. The choice of the regressive model can also influence the outcome, where we should emphasize that the regressive coefficients change with the change of number of results in the next research phase. Forming of similar correlations for carbonate complexes from other location in Bosnia and Republic of Macedonia can be on of the directions in further analyses, however the ways to modify and develop these models.

In this work, we have given the example of formation of numerical model and conceptual interaction matrix for one hydropower tunnel in carbonate complexes, where we should emphasize that formation of such matrixes a separate problem with every structure. Analyses carried out in the dissertation show that formation of conceptual matrixes is a complex process were due to numerous effects between parameters in matrix. However, when these effects are properly examined and problem defined

111


in a right way, we can expect that through further analytical or numerical analyses, the behaviour of complex system of rock – structure would be properly examined.

Based on aforementioned, we can conclude that there are many unlimited possibilities for further researches in this area. The purpose is to improve and confirm methodologies suggested in this thesis and not only when it comes to carbonate complexes but also for other types of rock massifs which for there are no similar models suggested so far.

112

7. APPENDIX

7. APPENDIX:

7.1. Table of disturbance factor Df wich is necessary in term for Hoek-Brown failure criterion

7.2. Table for MR-relation between deformation modulus of rock mass

and modulus of intact part of rock mass irm EE / (term Hoek and Diederichs, 2006)

113

7. APPENDIX

Tabela 7.1. Table for estimating Df (disturbance factor), www.rocksciennce.com.

114

http://www.rocksciennce.com/

7. APPENDIX

Tabela 7.2. Guideline for estimate MR value–based on papaers Deerea, 1968; and Palmstrom and Singha, 2001.

115

8. LITERATURE

8. LITERATURE Literature:

1. Božičević, S., (1983): Otkrivanje i proučavanje šupljine pod

građevinama u kršu. Mehanika stijena, temeljenje, podzemni radovi, Knjiga 1, Društvo GIT Zagreb, Društvo MSPR Hrvatske, Zagreb, str. 19-53.

2. Brady, B. H. G., Brown, E. T., (1993): Rock Mechanics: for underground mining, 2

nd Ed., Chapman & Hall, London.

3. Hawkins, A. B., (1998): Aspect of rock strength. Bull Eng Geol Env 57:17-30, pp 35-48.

4. Herak, M., (1973): Geologija, struktura, dinamika i razvoj zemlje. Školska knjiga, Zagreb, str. 26-145.

5. Hoek, E., Brown E.T., (1980): Empirical strength criterion for rock masses. J. Geotech. Engng Div., ASCE 106 (GT9), pp 1013-1035.

6. Hoek, E., (1983): Strength of jointed rock masses. 23rd. Rankine Lecture. Geotechnique, 33(3), pp 187-223.

7. Hoek, E., Brown E.T., (1997): Practical estimates of rock mass strength. Intnl. J. Rock Mech&Mining Sci.&Geomechanics Abstracts. 34(8), pp 1165-1186.

8. Hoek, E., Marinos, P., Benissi, M., (1998): Applicability of the geological strength index (GSI) classification for very weak and sheared rock masses. The case of the Athens Schist Formation. Bull Eng Geol Environ (1998) 57, pp 151-160.

9. Hoek, E., (2000): Rock engineering, Course notes by Evert Hoek. http: //www.rocksciennce.com.

10. Hoek, E., Carranza-Torres, C., Corkum, B, (2002): Hoek-Brown failure criterion-2002 edition. http: //www.rocksciennce.com.

11. Hoek, E., Marinos, P i Marinos, V., (2005): Characterisation and engineering properties of tectonically undisturbed but lithologically varied sedimetray rock masses. International Journal of Rock Mechanics & Mining Sciences 42 (2005), pp 277-285.

12. Hudson, J.A., (1993): Rock Properties. Comprehensive Rock Engineering.

13. Hudson, J.A., Harrison, J.P., (1997): Engineering Rock Mechanics. Pergamnon Press.

14. Ilijovski, Z., Jovanovski, M., Velevski, A., (2004): Metodologija na inženerskogeološko modeliranje na pregradnoto mesto za brana „Sveta Petka“. Prvi nacionalen kongres za brane, Ohrid.

116

8. LITERATURE

15. Jašarević, I. (1999): Prikaz novog klasifikacionog postupka „JAK“ za stijenske mase. Zbornik radova, Simpozij Mehanika stijena i tuneli, Zagreb.

16. Jašarević, I., Kovačević, M.S., Miščević, P., (1995): Modeliranje geotehničkih problema u stijenskim masivima, kompjuterski program FLAC. Građevinski godišnjak '95, Hrvatsko društvo građevinskih inženjera, 503-540, Zagreb.

17. Jovanovski, M., (2001): Prilog kon metodologija na istražuvanje na karpestite masi kako rabotna sredina, doktorska disertacija. Univerzitet „Sv. Kiril i Metodije“, Gradežen fakultet – Skopje, Skopje, 2001.

18. Jovanovski, M., Gapkovski, N., (1995): Analiza na možnostite na empiriski metodi pri proektiranje na podzemnite objekti. 6-ti Simpozium na društvo na gradežni konstrukteri na Makedonija, Ohrid.

19. Jovanovski, M., Krvavac-Špago, A., Ilijovski, Z., Peševski, I., (2008): Nekoi zavisnosti za inženjerskogeološkite karakteristiki na karbonatni karpesti masi od balkanski poluostrov. Prv kongres na geolozite na Republika Makedonija. Makedonsko geološko društvo i Univerzitet „Goce Delčev “ Štip.

20. Jovanovski, M., Gapkovski, N., (1998): Geostrukturni i matematički modeli za analiza na stabilnost na škrilesti karpesti masi. 6-ti Simpozium za teoretska i primeneta mehanika, Ohrid.

21. Jovanovski, M., Gapkovski, N., Ilijovski, Z., (2002): Correlation between Rock Mass Rating and deformability on a profile for arch dam Sveta Petka. 10-th International Conference of the DGKM, Ohrid.

22. Jovanovski, M., Gapkovski, N., Petrevski, Lj., (1996): Primer ekstrapolacije rezultata istraživanja kod specifičnih mekih stenskih masa. The International Conference: Trends in the Development of Geotechnics, Beograd.

23. Knill, J., (2003): Core values: the first Hans-Cloos lecture, Landslides. Bulletin of Eng Geol Environ (2003) 62:1-34, pp 11-13.

24. Krvavac, A., Jovanovski, M., Gapovski, N., Ilijovski, Z., (2006): Fizički i analitički modeli za karbonatne stijenske masive. II Simpozijum makedonskog udruženja za geotehniku, Ohrid.

25. Kujundžić, B., (1973): Sadržina i metodika izrade inženjersko-geoloških preseka i inženjersko-geoloških i geotehničkih modela. Saopštenja IX kongresa Jugoslovenskog komiteta za visoke brane, Zlatibor.

26. Kujundžić, B., (1977): Osnovi mehanike stena (I). Građevinski kalendar, SGIJT, Beograd.

27. Kujundžić, B.; (1979): Osnovi mehanike stena (II). Građevinski kalendar, SGIJT, Beograd, str 207-370.

28. Kujundžić, B., Petrović, Lj., (1980): Korelacija statičkih i dinamičkih karakteristika deformabilnosti krečnjačkih stenskih masa. V simpozij JDMSPR, 1, 5 -12, Split.

117

8. LITERATURE

29. Kujundžić, B., (1983): Opšta fizičko-strukturna svojstva stenskih masa. Mehanika stijena, temeljenje, podzemni radovi, Knjiga 1, Društvo GIT Zagreb, Društvo MSPR Hrvatske, Zagreb, str. 183-209.

30. Kujundžić, B., (1983): Terenska ispitivanja deformabilnosti, čvrstoća i naponskih stanja u stenskim masama. Mehanika stijena, temeljenje, podzemni radovi, Knjiga 1, Društvo GIT Zagreb, Društvo MSPR Hrvatske, Zagreb, str. 253-280.

31. Langof, Z., (1977): Određivanje prirodnih napona u nekim slučajevima degradiranih sedimentnih struktura. IV jugoslavenski simpozij o mehanici stena i podzemnih radova, Kosovska Mitrovica, Zvečani, str. 199-205.

32. Langof, Z., Bijeljanin, Lj., (1985): Sekundarna naponska stanja i deformacije oko podzemnog otvora u jednom krečnjačkom masivu. 6 simpozijum Jugoslovenskog društva za mehaniko hribin in podzemna dela, 1, Titovo Velenje, str. 205-210.

33. Lapčević, R., (1994): Stabilnost padina i kosina u karbonatnom kompleksu unutrašnjih Dinarida Srbije. Doktorska disertacija, Rudarsko-geološki fakultet, Beograd.

34. Lapčević, R., (1996): Efekat razmere u heterogenim stenskim masama. Međunarodni naučni skup Pravci razvoja geotehnike, Beograd.

35. Lapčević, R., (2005): Čvrstoća, deformabilnost i prirodna napregnutost čvrstih stenskih masa. Posebna izdanja Rudarsko-geološkog fakulteta, Beograd.

36. Lokin, P., Lapčević, R., Petričević, M., (1989): Principi i kriterijumi zoniranja, izbora uzoraka i ekstrapolacije rezultata ispitivanja na stenski masiv kod podzemnih objekata. VII JDMSPR, Beograd.

37. Lolino, P., Parise, M., Reina, A., (2004): Numerical analysis of the behavior of a karst cavern at Castellana-Grotte, Italy. Numerical Modelling of Discrete Materials – Konietzky (ed.) © 2004 Taylor&Francis Group, London, ISBN 90 5809 6351, pp 49-55.

38. Magdalenić, A., Raljević, B., Jurak, V., (1976): Inženjersko-geološke i hidrološke osnove za potrebe konsolidacionih radova. I Zbornik referata JUSIK, Zagreb, str. 57-65.

39. Marinos, P., Hoek, E., (2000): GSI: A geologically friendly tool for rock mass strength estimation. Proceedings of the International Conference on Geotechical & Geological Engineering (GeoEng 2000), Technomic Publishing Co. Inc., Melbourne, Australia, pp 1422-1440.

40. Marinos, P., Hoek, E., (2001): Estimating the geotehnical propreties of heterogeneous rock masses such as flysch. Bull Eng Geol Environ (2001) 60, pp 85-92.

41. Marinos, V., Marinos, P., Hoek, E., (2005): The geological strength index: applications and limitations. Bull Eng Geol Environ (2005) 64, pp 55-65.

118

8. LITERATURE

42. Marinos, P; Hoek, E. i Marinos, V., (2006): Variability of the engineering properties of rock masses quantified by the geological strength index. the case of ophiolites with special emphasis on tunnelling. Bull Eng Geol Environ (2006) 65, pp 129-142.

43. Milanović, P.T., (1999): Geološko inženjerstvo u karstu. Energoprojekat.

44. Mitrinović, M., Simić, M., Salihović, E., (1981): Istraživanje i projektovanje za hidroelektranu Grabovica. Energoinvest, Tehnika, nauka, inženjering, 18-1, Sarajevo, str. 107-116.

45. Müller, L., (1963): Der Felsbau, B.I. Ferdinand Enke Verlag,623, Stuttgart.

46. Müller, L., (1969): Gesteins und Gebirg, Seigenschaften in Abhängigkeit vo Betrachteten Grössenbereich. Z. deutsch. geol. Ges., Jahrgang 1967, 119, Hannover.

47. Olujić, M., (1977): Daljinska istraživanja u inženjreskoj geologiji. Mehanika stijena, temeljenje, podzemni radovi, Knjiga 1, Društvo GIT Zagreb, Društvo MSPR Hrvatske, Zagreb, str. 19-53.

48. Papić, M., (2005): Primjenjena statistika u MS Excelu za ekonomiste, znanstvenike i neznalice. ZORO d.o.o., Zagreb, 2005.

49. Pavlović, N., (1995): Neka razmatranja o geotehničkom modeliranju. Istraživanje i sanacija klizišta, Drugi simpozij, Donji Milanovac, str. 171-176.

50. Pavlović, N., (1996): Interakcija geološke sredine i inženjerske djelatnosti. The International Conference: Trends in the Development of Geotechnics, Beograd, str. 136-145.

51. Pavlović, N., (1996): O metodologiji geotehničkog modeliranja. The International Conference: Trends in the Development of Geotechnics, Beograd, str. 239-248.

52. Poulos, H.G., Davis, E. H., (1974): Elastic Solutions for Soil and Rock Mechanics, John Wiley & Sons, New York.

53. Selimović, M., (1974): Statical Analysis of Percentage of Failed Rock for the Purpose of Selecting the Location on the Salakovac Diversion Tunnel. Proceed, Third Congr., ISRM, II/B, Denver, pp.1326-1331.

54. Selimović, M., (2003): Mehanika stijena, Prvi dio, Teoretske osnove. Građevinski fakultet Univerziteta „Džemal Bijedić“ u Mostaru, Mostar.

55. Selimović, M., (2004): Mehanika stijena, Drugi dio, Istraživanja stijenskih masa. Građevinski fakultet Univerziteta „Džemal Bijedić“ u Mostaru, Mostar.

56. Selimović, M., Hlebar, V., Ivanković, T., Milinić, O. (1981) Osnovni rezultati istražnih radova na Hidroelektarni Salakovac. Simpozij o izgradnji hidroelektrana na srednjoj Neretvi, 1, Mostar, str. 55-66.

57. Simić, R., (1983): Projektiranje hidrotehničkog tunela (građevinsko-tehnički) elementi. Mehanika stijena, temeljenje, podzemni radovi,

119

8. LITERATURE

Knjiga 2, Društvo GIT Zagreb, Društvo MSPR Hrvatske, Zagreb, str. 815-842.

Design documention:

58. Geološki izvještaj, Knjiga 1, Glavni projekat, Brana HE Salakovac.

Energoinvest Sarajevo, Sarajevo, 1970. 59. Geološke podloge, HE Grabovica-Idejni projekat. Energoinvest

Sarajevo, Sarajevo, 1970. 60. Ispitivanje deformabilnosti stijenske mase metodom hidrauličkog

jastuka, Glavni projekat, Brana HE Salakovac. Institut za vodoprivredu, „Jaroslav Černi“, Beograd i „Energoinvest“ Sarajevo, 1970.

61. Ispitivanje deformabilnosti stijene ogledima pritiska „in situ“ u velikoj razmjeri. HE Grabovica, Mehanička ispitivanja, Knjiga 1. Institut za vodoprivredu, „Jaroslav Černi“, Odjeljenje za konstrukcije, Beograd, 1976.

62. Ispitivanja čvrstoća na pritisak okomito na slojeve za kamenolom HE Salakovac. Institut za ispitivanje materijala i kostrukcija, Građevinski fakultet u Sarajevu, 1970.

63. Phare Project, Upgrading of the E75 road from Demir Kapija to Gevgelija, Report from the geotechnical investigations for the tunne 1-5. Jovanovski, M., Gjorgevski, S. et all., Skopje, 2000.

64. Rezultati seizmičkih ispitivanja u profilu brane HE Salakovac. Zavod za geološka i geofizička istraživanja – Beograd. Odjeljenje za inženjersku geofiziku. Beograd, 1968-1972.

65. Seizmička ispitivanja u području pregradnog mjesta HE “Grabovica”. Zavod za geološka, hidrogeološka, geofizička i geotehnička istraživanja, Geofizički institut“, Beograd, 1975.

66. Svoden elaborat za izvršenite istražuvanja i ispituvanja za betonska lačna brana na pregadnoto mesto „Sveta Petka“ na reci Treska. Jovanovski, M., Ilijovski, Z., Prvulović, Z., Jančevski, J., Građevinski Institut „Makedonija“, a.d., Zavod za Geotehnika, Skopje, 2004.

120

LIST OF SYMBOLS

LIST OF SYMBOLS

a constant of Hoek-Brown failure criterion -

wich depends on GSI values

a diametar of circular opening in rock mass cm, m

c cohesion kN/m2

γ unit weight of rock mass kN/m3

D modulus of deformation MPa, GPa D, Df disturbance factor -

D diametar of hydrotechnical tunnel m

Ds shear modulus of deformation MPa, GPa

d size of samples (monolith) cm, m

E modulus of elasticity MPa, GPa Edyn dynamic modulus of elasticity MPa, GPa

Ei modulus of deformation of intact elements of rock mass MPa, GPa Erm modulus of deformation of rock mass MPa, GPa

ERMR Excavation Rock Mass Rating -

φ angle of internal friction °

θ angle in polar coordinate system that defined position ° of point where the stress is demande

G shear modulus for plane strain and isotropic conditions MPa, GPa

K, k coefficent of lateral pressure -

ζ coefficent of damage of rock mass %

GSI Geological Strength Index -

Is Index of strength gotten by Point load test MPa

Lu Lugeon’s unit for permeabilty l/min, m, 10bara

MR relation between deformation modulus of rock mass - and modulus of intact part of rock mass irm EE /

m Poisson’s number -

mb Hoek-Brown constant for rock mass -

mi Hoek_Brown constant that defines the frictional - characteristics of the component minerals in intact rock elements

n total porosity of rock mass %

ν Poisson’ s coefficient -

SYMBOL MEANING UNITS

121

LIST OF SYMBOLS

p pressure N/m2

REV Representative element volume m3

RMR Rock Mass Rating -

RQD Rock Quality Designation

RSR Rock Structure Rating

R determination coefficient -

r correlation coefficent -

r radius vector in polar coordinate system that defined - position of point where the stress is demande

ρ density kg/m3

s constant of Hoek-Brown failure criterion - wich depends on GSI values

sh distance of horizontal strata of rock mass cm, m

sv distance of vertical discontinuities of rock mass cm, m

σ stress of rock mass MPa

σc uniaxial compressive strength of rock mass MPa

σcd uniaxial compressive strength of rock mass sample MPa of diametar d (mm)

σc50 uniaxial compressive strength of rock mass sample MPa of diametar d = 50 mm

σci uniaxial compressive strength of intact elements of MPa rock mass

σ ph „in situ“ horizontal stress MPa

σ sh induced horizontal stresss MPa

σ’n normal effective stress MPa

σr radial stress around circular opening in rock mass MPa

σt tensile strength MPa

σt tangential stress around circular opening in rock mass MPa

σ pv „in situ“ vertical stress MPa

σ sV induced vertical stress MPa

σ1 principal vertical stress MPa

σ3 principal horizontal stress MPa

σ’1 principal effective vertical stress MPa

σ’3 principal effective horizontal stress MPa

σ’3max maximal lateral pressure at triaxial shear test MPa

122

LIST OF SYMBOLS

τ shear stress MPa

τrt shear stress around circular opening in rock mass

ur displacements in radial direction around circulr opening mm in rock mass

ut displacements in tangential direction around circulr opening mm in rock mass

v pH „in situ“ velocitiy of vawes in horizontal direction m/s, km/h

v sH induced velocitiy of vawes in horizontal direction m/s, km/h

vl velocity of longitudinal elastic vawes m/s, km/h

v pV „in situ“ velocitiy of vawes in vertical direction m/s, km/h

v sv induced velocitiy of vawes in vertical direction m/s, km/h

123

azra Špago - УКИМ spago.pdf · university “ss. cyril and methodius“ skopje . faculty of...

Documents