pes ieee usa jul2008

8
 1   Abstrac t—Designing electric installation projects, demands not only academic knowledge, but also other types of knowledge not easily acquired through traditional instructional methodologies. A lot of additional empirical knowledge is missing and so the academic instruction must be completed with different kinds of knowledge, such as real-life practical examples and simulations. On the other hand, the practical knowledge detained by the most experienced designers is not formalized in such a way that is easily transmitted. In order to overcome these difficulties present in the engineers formation, we are developing an Intelligent Tutoring System (ITS), for training and support concerning the development of electrical installation projects to be used by electrical engineers, technicians and students.  Index Terms Electric Installation Design, Intelligent Tutoring Systems, Knowledge Based Systems. I. INTRODUCTION rojects of electrical installations for private use require not only academic knowledge but also other types of knowledge not easily acquired through traditional instructional methodologies. Usually, this kind of knowledge is only acquired by the students when they start a professional life. A lot of additional empirical knowledge is missing and so the academic instruction must be completed with different kinds of knowledge, such as real-life practical examples and simulations. On the other hand, the practical knowledge detained by the most experienced designers is not formalized in such a way that is ea sily transmitted. In order to overcome these difficulties present in the engineers formation, we are developing an Intelligent Tutoring System, for training and support concerning the development of electrical installation projects to be used by electrical engineers, technicians and students. An intelligent tutoring system, broadly defined, is any computer system that provides direct customized instruction or feedback to students, i.e. without the intervention of human beings. An G. Santos, A. Gomes, L. Faria, S. Ramos and Z. Vale are with GECAD – Knowledge Engineering and Decision-Sup port Research Gr oup of the Electrical Engineering Institute of Porto – Polytechnic Institute of Porto (ISEP/IPP), Rua Dr. António Bernardino de Almeida, 431, 4200-072 Porto, Portugal (e-mail: [email protected] / [email protected] / [email protected] / [email protected] / [email protected]). ITS may employ a host of different technologies. However, usually such systems are more narrowly conceived of as artificial intelligence systems. This work is integrated in the research project TINSEL, standing for Intelligent Tutors for Electrical Installation Design. The TINSEL team was involved in previous projects in the area of ITS [1][2][3] applied to Power Systems operation [4][5][6]. The main goal of this work is twofold: Training of students in this domain know ledge; Updating of teachers and technici ans on new technical solutions and legislation. The goals stated above are being translated into: • Creation of a k nowledge base about electric installations projects; Development of a tutor for electric install ation projects training. II. ITS FRAMEWORK The Intelligent Tutoring System is being developed based on a client-server architecture where a part of its functionality is being implemented to run on the client side, and another part to run on the server side. The parts will communicate over the Internet. In the client side, the interface will be responsible to send trainee actions to the server, and also to show the feedback generated by the server. One of the goals of the project is to endow the ITS with capabilities to support collaboration between all learning entities, including different trainees and virtual agents. In order to support collaboration between all learning entities, the tutor will include a suitable environment to support cooperative work between several students and between students and the tutoring system. While permitting different types of interaction between the different agents, this framework shall be allow the tutor to apply different learning strategies, in order to tailor the tutoring process to the trainee’s characteristics. The frame-work must not only address the issues related to information flow but also be able to motivate and facilitate the discussion as well as organizing the distributed decision process. The tutor will be divided in two modules. The first one is a tutoring module intended to teach the legislation and its practical application in the context of an electrical project. An Intelligent Tutoring App roach to Support Students and Technicians in Electrical Installation Design Gustavo Santos, António Gomes, Luiz Faria, Sérgio Ramos, Zita Vale P

Upload: sergioramos

Post on 03-Jun-2018

219 views

Category:

Documents


0 download

TRANSCRIPT

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 1/7

1

Abstract —Designing electric installation projects, demands notonly academic knowledge, but also other types of knowledge noteasily acquired through traditional instructional methodologies.A lot of additional empirical knowledge is missing and so theacademic instruction must be completed with different kinds ofknowledge, such as real-life practical examples and simulations.On the other hand, the practical knowledge detained by the mostexperienced designers is not formalized in such a way that iseasily transmitted.

In order to overcome these difficulties present in the engineersformation, we are developing an Intelligent Tutoring System(ITS), for training and support concerning the development ofelectrical installation projects to be used by electrical engineers,technicians and students.

Index Terms – Electric Installation Design, IntelligentTutoring Systems, Knowledge Based Systems.

I. I NTRODUCTION

rojects of electrical installations for private use require notonly academic knowledge but also other types ofknowledge not easily acquired through traditional

instructional methodologies. Usually, this kind of knowledgeis only acquired by the students when they start a professionallife.

A lot of additional empirical knowledge is missing and sothe academic instruction must be completed with differentkinds of knowledge, such as real-life practical examples andsimulations. On the other hand, the practical knowledgedetained by the most experienced designers is not formalizedin such a way that is easily transmitted.

In order to overcome these difficulties present in theengineers formation, we are developing an IntelligentTutoring System, for training and support concerning thedevelopment of electrical installation projects to be used byelectrical engineers, technicians and students. An intelligenttutoring system, broadly defined, is any computer system thatprovides direct customized instruction or feedback tostudents, i.e. without the intervention of human beings. An

G. Santos, A. Gomes, L. Faria, S. Ramos and Z. Vale are with GECAD –Knowledge Engineering and Decision-Support Research Group of the ElectricalEngineering Institute of Porto – Polytechnic Institute of Porto (ISEP/IPP), RuaDr. António Bernardino de Almeida, 431, 4200-072 Porto, Portugal (e-mail:[email protected] / [email protected] / [email protected] / [email protected] /[email protected]).

ITS may employ a host of different technologies. However,usually such systems are more narrowly conceived of asartificial intelligence systems.

This work is integrated in the research project TINSEL,standing for Intelligent Tutors for Electrical InstallationDesign. The TINSEL team was involved in previous projectsin the area of ITS [1][2][3] applied to Power Systemsoperation [4][5][6].

The main goal of this work is twofold:• Training of students in this domain knowledge;• Updating of teachers and technicians on new technical

solutions and legislation.The goals stated above are being translated into:

• Creation of a knowledge base about electric installationsprojects;

• Development of a tutor for electric installation projectstraining.

II. ITS FRAMEWORK

The Intelligent Tutoring System is being developed based

on a client-server architecture where a part of its functionalityis being implemented to run on the client side, and anotherpart to run on the server side. The parts will communicateover the Internet. In the client side, the interface will beresponsible to send trainee actions to the server, and also toshow the feedback generated by the server.

One of the goals of the project is to endow the ITS withcapabilities to support collaboration between all learningentities, including different trainees and virtual agents. Inorder to support collaboration between all learning entities,the tutor will include a suitable environment to supportcooperative work between several students and betweenstudents and the tutoring system. While permitting differenttypes of interaction between the different agents, thisframework shall be allow the tutor to apply different learningstrategies, in order to tailor the tutoring process to thetrainee’s characteristics. The frame-work must not onlyaddress the issues related to information flow but also be ableto motivate and facilitate the discussion as well as organizingthe distributed decision process.

The tutor will be divided in two modules. The first one is atutoring module intended to teach the legislation and itspractical application in the context of an electrical project.

An Intelligent Tutoring Approach to SupportStudents and Technicians in Electrical

Installation DesignGustavo Santos, António Gomes, Luiz Faria, Sérgio Ramos, Zita Vale

P

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 2/7

2

This module will also be in charge of the transfer of theknowledge needed to accomplish each phase of an electricalproject. We are implementing this module using AdaptiveHypermedia Technologies (see chapter III.A).

The second module of the ITS is a problem solvingmodule where the learner will have the opportunity to solvepractical problems proposed by the ITS. This module will bebased on the Interactive Problem Solving technologies (see

chapter III.B).

III. ITS TECHNIQUES

This chapter describes the state-of-the-art of someIntelligent Tutoring System techniques that we considerrelevant to our work. As mentioned before, the tutor will bedivided in two modules. Each model will be responsible toteach different types of knowledge. Different kinds ofknowledge require different teaching techniques. One moduleis being implemented using Adaptive Hypermedia and theother will be based on the Interactive Problem Solvingtechniques.

A. Adaptive Hypermedia

Adaptive Hypermedia systems apply different forms ofuser models to adapt the content and the links of hypermediapages to the user [7]. We distinguish two major technologiesin adaptive hypermedia: presentation and navigation support.

Education always was one of the main application areas foradaptive hypermedia. First Web-based Adaptive andIntelligent Educational Systems (AIES) using adaptivehypermedia technologies were reported in 1996 [8][9].Adaptive navigation support (ANS) can be considered as ageneralization of curriculum sequencing technology in ahypermedia context. It shares the same goal (to help studentsto find an "optimal path" through the learning material) buthas more options than traditional sequencing. In a WWWcontext where hypermedia is a basic organizational paradigm,adaptive navigation support can be used very naturally andefficiently.

There are several known ways to adapt the links [7]. Twoexamples of ANS-based systems are ISIS-Tutor [10] withadaptive link hiding and adaptive annotation, and Hypadapter[11] with adaptive link hiding and adaptive sorting. The threemost popular Web-based AIES are direct guidance, adaptivelink annotation, and adaptive link hiding. The most popularform of ANS on the Web is annotation. It was used first inELM-ART [8] and since that applied in all its descendants.ELM-ART and InterBook also use adaptive navigationsupport by sorting.

Adaptive presentation is very important in WWW contextwhere the same "page" has to suit very different students.Only two Web-based AIES implement complete adaptivepresentation: PT [12] and AHA [13]. Medtec [14] is able togenerate adaptive summary of book chapters. MetaLinks cangenerate a special preface to a content page depending onwhere the student came from to this page. A very interestingexample of adaptive presentation is suggested in WebPersona

project [15] where presentation is performed by a life-likeagent.

More detailed information about Adaptive Hypermedia canbe found on [16] a book written as a state-of-the-art reviewand a textbook for graduate courses on adaptive Web,adaptive information systems and like.

B. Problem Solving Support

Problem solving support has been considered the mainduty of an ITS system and a main value of an ITS technology.Three problem solving support technologies can be identified:intelligent analysis of student solutions, interactive problemsolving support, and example-based problem solving support.

Intelligent analysis of student solutions deals with students’final answers to educational problems no matter how theseanswers were obtained. To be considered as intelligent, asolution analyzer has to decide whether the solution is corrector not, find out what exactly is wrong or incomplete, andpossibly identify which missing or incorrect knowledge maybe responsible for the error.

Interactive problem solving support is a more recent andpowerful technology. Instead of waiting for the final solution,this technology can provide a student with intelligent help oneach step of problem solving and it allows to watch thestudent’s actions, understand them, and use thisunderstanding to provide help and to update the student modelas in LISP-TUTOR [17]. Another example of application ofinteractive problem solving can be seen in an ITS previouslydeveloped by our team [1][2][3] and aimed to train ControlCentre operators to get diagnosis of incidents in electricalnetworks.

The example-based problem solving technology is thenewest one. This technology is helping students to solve newproblems not by articulating their errors, but by suggestingthem relevant successful problem solving cases from theirearlier experience. An example is ELM-PE [18] or, in theWeb context, in ELM-ART [19] and ELM-ART-II [20].

The Case-Based Reasoning concept seems particularly welladapted to the curricula’s structure of the particular domainwe are dealing with. The idea of case-based teachingpresented by Schank [21] derived from the field of case-basedreasoning [22][23], which is the process of solving newproblems based on the solutions of similar past problems.Schank states that students learn better from casesrepresenting real tasks to be presented at the precise point of

becoming interested in acquiring the information conveyed bythe case. It is also relevant the research carried out by VanLehn and al [24] in ways of adapting tutor behavior tostudents’ performance, by varying the scaffolding level of thesupport given. On the subject of representing, selecting andplanning tutoring strategies the works of Schank [25] on Goal-Based Scenarios, Murray [26] on Discourse ManagementNetworks and Cho and all [27] seems suitable for ourpurposes.

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 3/7

3

IV. T EACHING MODEL

Different teaching models had been proposed in the area ofIntelligent Tutoring Systems with the purpose of teaching indifferent types of domains. Teaching some of these domainsrequires exposition like approaches where the learner mustacquire knowledge presented in an expositive way -sometimes through the application of adaptive hypermediatechniques. Law and some sub-domains of medicine are

examples of these domains. However, the learning process inother domains requires the acquisition of skills that can onlybe achieved by the learner through the resolution of practicalproblems [2].

Furthermore, the design of electrical installations involvesthe use of different types of knowledge and expertise whichcan only be efficiently transmitted through the application ofdifferent teaching techniques. As stated in session II, thelegislation and its practical application in the context of anelectrical project is one of the major types of knowledgeneeded to accomplish the design of electrical installations.Due to the nature of this type of knowledge, the adaptivehypermedia technologies become the best approach totransmit such kind of knowledge. On the other hand, the useof interactive problem solving technologies reveals to be anoption to the acquisition of competences needed to performtasks such as circuit dimensioning.

Since the electrical installation design requires suchdifferent kinds of competences, which demands for differentteaching techniques, the teaching model devised for theTINSEL project includes different tutoring approaches [28].Figure 1 presents in detail the tutoring model proposed andthe tutor’s roles in different learning situations.

T r a i n e e a s s e s s m e n t

P r a c t i c a l p r o b l e m s

s o l v i n g

T r a i n e e g u i d a n c e t h r o u g h

a d a p t a t i v e h y p e r m e d i a

t e c h n i q u e s

E x a m e l l a b o r a t i o n a n d

t r a i n e e p e r f o r m a n c e

e v a l u a t i o n

P r o b l e m s e l e c t i o n a c c o r d i n g

t o t h e t r a i n e e m o d e l

T u t o r a s s i s t a n c e d u r i n g

p r o b l e m s o l v i n g

D e t e c t i o n o f

m i s c o n c e p t i o n s

T r a i n e e s h o w s t o h a v e e n o u g h

k n o w l e d g e t o s o l v e p r a c t i c a l

p r o b l e m s

T r a i n e e w a s n o t a b l e t o

a c c o m p l i s h t h e p r o p o s e d

p r a c t i c a l p r o b l e m a n d t h e h i n t s

p r o v i d e d b y t h e T u t o r w e r e n o t

e n o u g h t o o v e r c o m e h i s

d i f f i c u l t i e s

T r a i n e e s o l v e d t h e

p r o p o s e d p r a c t i c a l

p r o b l e m s u c e s s f u l l y

K n o w l e d g e

e x p l o r a t i o n / a c q u i s i t i o n

E v e n t s r e s p o n s i b l e f o r l e a d i n g t h e t r a i n e e

t o a n e w t y p e o f l e a r n i n g u n i t

T u t o r r o l e

L e a r n i n g u n i t

Fig. 1. Tutoring Model and Tutor’s Roles

The tutoring model presented in Figure 1 contains threetypes of learning units, represented by square boxes. Therounded boxes represent the tutoring roles during the activityof the trainee while in a learning unit.

A trainee starts his training by taking some lessons oftheoretical nature. The knowledge acquired during these

lessons is essential to enable the trainee to accomplish tasksof more practical nature involved in the design of an electricalinstallation. For this type of lessons the tutoring model definesthe use of learning units from the type named “knowledgeexploration” (corresponding to the top most box in Figure 1.The implementation of instances of this kind of learning unitis based on the application of adaptive hypermediatechniques. The role of the tutor here is to select the mostadequate hypermedia pages according to the trainee modeland to supply guidance while the learner navigates through theselected hypermedia pages.

After the trainee had consulted some hypermedia pages, thetutor recommends the trainee to be submitted to an

assessment process in order to evaluate the trainee’sknowledge about the domain topics addressed in the reviewedpages. These assessments are implemented through multiplechoice exams and correspond to the second type of learningunit present in the tutoring model. In this type of learningunits, the tutor’s role includes the exam elaboration and thecorresponding trainee performance evaluation. After theassessment, if the tutor finds misconceptions in the traineeknowledge then the tutor guides the trainee back to thelessons related with the misconceptions and orients the traineeto over-come these. However if the trainee had a positiveperformance and shows enough knowledge to solve a sub-setof practical problems then the tutor suggests the trainee to

accomplish such a problem.The practical problem solving activity corresponds to the

third type of learning unit and applies the well known learningtechnique named “learning by doing”. The implementation ofthe instances of this kind of learning unit is based oninteractive problem solving techniques. The mechanisms usedto detect misconceptions and to infer the reasons behind thesemisconceptions are based on the “model tracing” technique[2], [24].

In this stage of the tutoring process, the role of the tutor istwofold:

• To select the problem according to the presentlearning needs of trainee, matching the trainee’s model;

• To assist the trainee during the problem solvingactivity.

During problem solving, if the trainee reveals difficulties,the tutor will provide hints in order to overcome thesedifficulties. The hints provided by the tutor start from genericand less detailed hints and follow to more specific and moredetailed hints if the trainee still maintains his difficulties.However, if all the hints supplied by the tutor were not

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 4/7

4

enough to overcome the difficulties, then the tutor guides thetrainee back to the lessons containing the knowledge needed.

Finally, if the trainee solved the proposed practical problemsuccessfully the tutor can conclude that the trainee is able tofollow the next learning unit and guides him to a new“knowledge exploration” unit.

V. S YSTEM ARCHITECTURE

The Intelligent Tutoring System is being developed basedon a client-server architecture where a part of its functionalityis being implemented to run on the client side (Front-End),and another part to run on the server side (Back-End) [29].The parts will communicate over the Internet basically usingthe HTTP Protocol. In the client side, the interface will beresponsible to send trainee actions to the server, and also toshow the feedback generated by the Back-End. The serverside is responsible for receiving trainee’s actions, model thestudent according to those actions, and provide the necessarymechanism to generate the interface that will interact with thestudent.

To support the designed ITS architecture, a set ofcomponents had to be developed. After the implementation ofa prototype, some tests proved that the designed architectureis able to support the ITS entirely. Figure 2 represents theimplemented architecture.

Fig. 2. System Architecture

The system has two larger parts, the Front-End and theBack-End. The Front-End is the interface that interacts withthe student. It sends trainee actions to the server and alsoshows the feedback. It is basically composed of a Web-Portal,the entry point of the ITS. It makes available contents andfunctionalities to assist the learning process of the students.The Web-Portal can be decomposed in two other smallcomponents, the Interface of the Tutor Module and the WebContent Management System.

The Interface of the Tutor Module is the part of the systemthat communicates directly with the Tutor Module to providedata exchanging between them. The Web ContentManagement System is a content management system (CMS)implemented as a web application, for creating and managingHypertext content. It is used to manage and control a large,dynamic collection of web material (HTML documents,

images, animations, videos, etc.). It facilitates contentcreation, content control, editing, and many essential Webmaintenance functions.

The Back-End is responsible for receiving trainee’s actions,model the student according to those actions, and provide thenecessary mechanism to generate the interface that willinteract with the users. It is composed by three other parts thatcan be considered as layers, the Operating System, the

Relational Database, and the Tutor Module. The OperatingSystem and the Relational Database are the lower layers ofthe Back-End respectively. They are just ordinary componentsof any Information System. The Tutor Module is the mainpart of the Back-End. It is responsible for planning theinstruction, teaching, monitoring the students, and assessingthe students.

The architecture was developed to allow the interaction ofthe students with the ITS through an interface that could beaccessed from anywhere by a Web browser. This will improvethe tutoring process since the students do not have to leavetheir working place to study, or complement their studies.

VI. K NOWLEDGE ACQUISITION AND REPRESENTATION The Domain Knowledge corresponds to the subject that it

is intended to be transmitted to the students and it is expectedto be assimilated by them. In artificial intelligence the typicalknowledge classification is divided in two types: Declarativeknowledge and Procedural knowledge [30][31]. Thedeclarative knowledge describes what it is known about acertain problem, which includes statements that could be trueor false; statements that describes an object or a concept. Itcorresponds to a descriptive representation. The Proceduralknowledge describes how a problem is solved or how to act inpresence of a certain situation (“how to do it”). Strategies,schedules and proceedings are typical representations for thiskind of knowledge. Also in the psychology field, thedistinction between Declarative and Procedural knowledgeexists [32] [33] [34].

The approaches to teach Declarative and Proceduralknowledge are quite different independent of the knowledgefield. Teaching Declarative knowledge is usually based inexpositive methods. Teaching Procedural knowledge usingexpositive methods usually is not satisfactory.

The acquisition of the Procedural knowledge is a complexprocess and it is based in the progressive construction of amental model, generally achieved by solving real problems.The declarative knowledge appears, for instance, inrecognition tasks, while the Procedural knowledge isdemonstrated during the realization of a certain tasks.Sometimes, the declarative knowledge by itself, it is not veryuseful. The Procedural knowledge can use Declarativeknowledge as support to the inference of new knowledge.

Nowadays, the domain knowledge representation is basedon an approach directed to the domain to be represented. Theselection of a technique, or combination of techniques,depends of the domain that we want to represent. Thetaxonomies developed by Kyllonen e Shute [35] aboutlearning abilities, gives an example to demonstrated how

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 5/7

5

different domains can be completely distinct in terms ofcontents.

Projects of electrical installations for private use require notonly academic knowledge but also other types of knowledgenot easily acquired through traditional instructionalmethodologies. Thus, in order to create a knowledge baseabout electric installation design, it was necessary to performa knowledge acquisition process to extract and represent the

knowledge of the experts in this domain. This task isextremely complex and it can take a lot of time.Usually, the knowledge acquisition phase involves the

participation of several experts. Nevertheless, some problemsregarding the knowledge transmission can occur among theseveral entities that participate in the development of theknowledge base. The reason why this kind of problems canoccur is related to the unavailability of the experts. Normally,they are extremely busy professionals and cannot participateactively in the knowledge representation. They can also thinkthat these tools are threats, and possibly replace their jobs inthe future, thus, they will not collaborate as well. When it isverified an active and participative involvement of all entities

in the project, the problems that can happen will be certainlyeasier to solve.

The tutor module is responsible by the pedagogicaldecisions of the ITS. The tutor module uses the informationof the knowledge base to propose questions to the student, toanalyze his performance and to help him find the correctanswer. The knowledge base of our project was created to

represent the knowledge related to electric installation designprojects.

The knowledge representation consisted in the definition ofa Semantic Network [36] representing, processes, concepts,sub-concepts and instances related with the regulatedknowledge, as well as the knowledge based on the empiricalknowledge of professional experts. Figure 3, shows theknowledge representation for the power determination to use

in electrical installations.

Process P1.1 is the “Definition of the Installation UsageType”. To accomplish this task it is necessary to know theexisting types of installations. The types of installations areconcepts represented in the Figure 3. Those concepts aredefined in the Technical Rules of Low Voltage ElectricInstallations, chapter 1, section 1.1 and chapter 8, section 1(Portuguese electrical regulation).

The second process P1.2, “Physical Characterization of theInstallation Usage “, must be done after the process P1.1,since the physical characterization of the installation requiresa previous definition of the installation type, namely due to

the specific elements that vary according to the installationtype. For example, in a house, besides the architecturalcharacterization it is also necessary to define the number ofmain compartments.

After defining the usage installation type (process P1.1)and characterizing usage installation type (process 1.2), wecan perform P1.3, “Definition of the Hired Power to

Fig. 3. knowledge representation for the power determination

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 6/7

6

Installation Usage”, in order to define the power to hire.To accomplish process P1.3 it is necessary, to dominate the

Technical Rules, which is presented in chapter 8, section 1 ofthe Technical Rules of Low Voltage Electric Installations. Onthe other hand it also useful to dominate, the empiricalknowledge of the experts.

For example, for commercial usage buildings, the powerwill be defined according to the regulation specified in

chapter 8, section 126 of the Portuguese regulation. Theempirical knowledge of the experts can complement theregulation because the experts take into consideration,economical aspects, functional aspects, and the flexibility ofthe installations. Algorithm 1 represents an algorithm todefine the power to hire in commercial usage buildings.

If (Commercial Buildings Uses){

If (know commercial activity){

Determine the installed power ();

Determine the functioning coefficients ();

Determine the corrected power ();

Determine the power according to the regulation criteria ();

Definition of the normalized power to hire ();

}else{ Determine the power according to the regulation criteria ();

Definition of the normalized power to hire ();

}

}

Definition of the normalized power to hire (){

If (known commercial activity){

If(P R ≥ PCT ){

P Hired = Normalize_Power(P R);

}else{

If (PCT = normalized value){

P Hired = P TC

}else{

P Hired = Normalize_ Power(P CT );

}

}

}else{

P Hired = Normalized _Power(P R);

}

}

Determine the installed power (){

List and characterization of the equipments;

}

Determine the functioning coefficients (){

C S = Definition of the simultaneity coefficient;

C U = Definition of the usage coefficient;

}

Determine the corrected power (){

PCT = installed power x C S x C U ;

}

Determine the power according to the regulation criteria (){

AreaH =commercial building area in m 2;

P Area = 30VA x AreaH;

If (P Area <3,45 kVA){

P R = 3,45 kVA;

}Else{

P R = P Area ;

}

}

Algorithm 1. Representation of the knowledge to definethe power to hire in commercial usage buildings

VII. C ONCLUSIONS

An Intelligent Tutoring System in the field of ElectricInstallation Design will fill a gap in the learning processstructure, performing a function not fulfilled by anyone. Itwill also complement and improve the tutoring processoffered by other instructional techniques in the followingaspects:

• The experts in the field of electrical installationdesign tend to be rare and usually they are notavailable to assist novice designers;

• The training quality suffers from the lack of anevaluation mechanism and the fact that the sametraining program is applied to all the participants;

• The trainees do not have to leave their workingplace since the tutor can be used through theinternet.

The teaching model proposed is able to ensure adequatesupport of the trainees in the acquisition of the different typesof knowledge and expertise required to perform successfullythe design of electrical installations, since it defines differentteaching technologies appropriated to the acquisition of theheterogenic types of knowledge involved.

The use of these Tutors can therefore improve electricalinstallation designers' formation, which is critical due to thelack of well-trained technicians in this area.

VIII. R EFERENCES [1] Silva, A., Faria, L., Vale, Z., Ramos, C., Marques, A., "User Modelling

Concerning Control Centre Operators Training", IEEE Porto Power Tech,2001.

[2] Faria, L., “Treino e Apoio a Operadores de Centros de Controlo eCondução – Uma Abordagem Baseada em Conhecimento e TutoresInteligentes”, Ph.D. dissertation, Dept. Elect. And Computer Eng., Facultyof Engineering, University of Porto, Portugal, 2002.

[3] Faria, L., Vale, Z., Ramos, C., Marques, A., “Intelligent Training ofIncident Diagnostic Task: Adapting Curriculum to Operator Needs”,Intelligent Systems Applications to Power Systems (ISAP’2003), Lemnos,Grécia, 1-3 Septem-ber, 2003.

[4] Vale, Z., Fernandes, F., Rosado, Marques, A., Ramos, C., Faria, L.,"Better KBS for Real-time Applications in Power System Control Centers:the Experience of SPARSE Pro-ject", Computers in Industry, Elsevier, 97-111, 1998.

[5] Vale, Z., Ramos, C., "Reasoning about Time in AI applica-tions for PowerSystems Control Centers", Int. Journal of Engineering Intelligent Systemsfor Electrical Engineering and Communications, 91-96, CRL PublishingLtd, 7, 2, 1999.

[6] Vale, Z., Ramos, C., Faria, L., Malheiro, N., Marques, A., Rosado, "Real-Time Inference for Knowledge-Based Appli-cations in Power SystemControl Centers", SAMS - Journal Systems Analysis-Modelling-Simulation, Gordon & Breach Science Publishers, 2001.

[7] Brusilovsky, P.: Methods and techniques of adaptive hyper-media. UserModelling and User-Adapted Interaction 87-129, 6, 2-3, 1996.

[8] Brusilovsky, P., Schwarz, E., and Weber, G.: ELM-ART: An intelligenttutoring system on World Wide Web. In: Frasson, C., Gauthier, G. andLesgold, A. (eds.) Intelligent Tutoring Systems. Lecture Notes in ComputerScience, Vol. 1086. Springer Verlag, Berlin, 261-269, 1996.

[9] De Bra, P. M. E.: Teaching Hypertext and Hypermedia through the Web.Journal of Universal Computer Science 2, 797-804, 12, 1996.

[10] Brusilovsky, P. and Pesin, L.: An intelligent learning envi-ronment forCDS/ISIS users. In: Levonen, J. J. and Tukianinen, M. T. (eds.) Proc. ofThe interdisciplinary workshop on complex learning in computerenvironments (CLCE94), Joensuu, Finland, EIC, 29-33, 1994.

[11] Hohl, H., Böcker, H.-D., and Gunzenhäuser, R.: Hy-padapter: An adaptivehypertext system for exploratory learning and programming. UserModelling and User-Adapted Interaction 6, 2-3, 131-156, 1996.

8/11/2019 Pes Ieee Usa Jul2008

http://slidepdf.com/reader/full/pes-ieee-usa-jul2008 7/7

7

[12] Kay, J. and Kummerfeld, B.: User models for customized hypertext. In:Nicholas, C. and Mayfield, J. (eds.): Intelli-gent hypertext: Advancedtechniques for the World Wide Web. Lecture Notes in Computer Science,Vol. 1326. Springer-Verlag, Berlin, 1997.

[13] De Bra, P. and Calvi, L.: AHA! An open Adaptive Hyper-mediaArchitecture. The New Review of Hypermedia and Multimedia 4, 115-139, 1998.

[14] Eliot, C., Neiman, D., and Lamar, M.: Medtec: A Web-based intelligenttutor for basic anatomy. In: Lobodzinski, S. and Tomek, I. (eds.) Proc. ofWebNet'97, World Conference of the WWW, Internet and Intranet,Toronto, Canada, AACE, 161-165, 1997.

[15] André, E., Rist, T., and Müller, J.: WebPersona: A Life-Like PresentationAgent for Educational Applications on the World-Wide Web. In:Brusilovsky, P., Nakabayashi, K. and Ritter, S. (eds.) Proc. of Workshop"Intelligent Educational Systems on the World Wide Web" at AI-ED'97,8th World Conference on Artificial Intelligence in Education, Kobe, Japan,ISIR, 78-85, 1997.

[16] Peter Brusilovksy, Alfred Kobsa, Wolfgang Nejdl, “The Adaptive Web:Methods and Strategies of Web Personalization” Springer Lecture Notes inComputer Science, Vol. 4321/2007

[17] Anderson, J. R. and Reiser, B.: The LISP tutor. Byte 10, 4, 159-175, 1985.[18] Weber, G.: Individual selection of examples in an intelligent learning

environment. Journal of Artificial Intelligence in Education 7, 1, 3-31,1996.

[19] Brusilovsky, P., Schwarz, E., and Weber, G.: ELM-ART: An intelligenttutoring system on World Wide Web. In: Frasson, C., Gauthier, G. andLesgold, A. (eds.) Intelligent Tutoring Systems. Lecture Notes in ComputerScience, Vol. 1086. Springer Verlag, Berlin, 261-269, 1996.

[20] Weber, G. and Specht, M.: User modelling and adaptive navigationsupport in WWW-based tutoring systems. In: Jameson, A., Paris, C. andTasso, C. (eds.) User Modelling. Springer-Verlag, Wien, 289-300, 1997.

[21] Schank R, Case-Based Teaching: four experiences in educa-tional softwaredesign, Interactive Learning Environments, 1, 231-254, 1999.

[22] Kolodner J, Improving human decision making through case-baseddecision aiding, AI Magazine, 12, 2, 52-68, 1991.

[23] Kolodner J, Case-Based Reasoning, San Mateo, M Kauf-man, 1993.[24] VanLehn K et al., Fading and Deepening: The Next Steps for Andes and

other Model-Tracing Tutors, ITS'2000 Conf., Springer, 2000.[25] Schank, Roger C., Goal-Based Scenarios: Case-Based Reasoning Meets

Learning by Doing. In: David Leake (Ed) Case-Based Reasoning:Experiences, Lessons & Future Di-rections. AAAI Press/The MIT Press,295-347., 1996.

[26] Murray W, Control for ITSs: A Comparation of Blackboard Architectureand Discourse Management Networks. Tech. Rep. R-6267, FMCCorporation.

[27] Cho B, and al, A Curriculum Planning Model for an ITS, 12th Int. FloridaAI Research Soc. Conf., FLAIRS'99, USA, 197-201, 1999.

[28] Luiz Faria, António Gomes, António Silva, Zita Vale, Carlos Ramos,Gustavo Santos, Sérgio Ramos, Fernando Ferreira, “ A teaching Model forElectrical Installation De-sign Instruction based on Intelligent Tutoring”,International Conference on Knowledge Engineering and Decision Sup-port, Lisboa, Portugal, 10 e 11 de Maio, 2006.

[29] Luiz Faria, António Gomes, Gustavo Santos, António Silva, Zita Vale,Carlos Ramos, Sérgio Ramos, Fernando Ferreira, “TINSEL – A web-based Intelligent Tutoring System on Electrical Installation Design”,CEE07 - 2nd International Conference on Electrical Engineering,Coimbra, Portugal, 26-28 de Novembro, 2007.

[30] Charniak, E. e McDermott, D., “Introduction to Artificial Intelligence”,Reading, MA, Addison-Wesley, 1985.

[31] Rich, E. e Knight, K., “Artificial Intelligence”, New York, McGraw-Hill,1991

[32] Anderson, J. R., “Acquisition of Cognitive Skill”, Psychological Review,n. 89, 369-406, 1982

[33] Sleeman, D. H. e Brown, J. S., “Intelligent Tutoring Systems”, AcademicPress, New York, 1982

[34] VanLenh, K., “Student Modeling”, em Polson, M. C. e Richardson, J. J.(Ed.), Foundations of Intelligent Tutoring Systems, 57-78, LawrenceErlbaum Associates, Hillsdale, NJ, 1988

[35] Kyllonen, P. C. e Shute, V. J., “A Taxonomy of Learning Skills”, emAckerman, P., Sternberg, R. e Glaser, R., Learning and IndividualDifferences, 117-163, W. H. Freeman, San Francisco, 1988

[36] Sowa, John F., ed., “Principles of Semantic Networks: Explorations in theRepresentation of Knowledge”, Morgan Kaufmann Publishers, San Mateo,CA, 1991

IX. B IOGRAPHIES

Gustavo Santos Was born in Brazil in 1980. Hegraduated at the Polytechnic Institute of Porto in 2006,receiving three academic awards during hisgraduation.

He is currently a Ph.D. student of ComputerSciences at University of Minho (Braga-Portugal), andone of the Portuguese students taking part in the dualdegree doctoral program between Carnegie MellonUniversity (Pittsburgh-USA) and the PortugueseGovernment.António Gomes graduated in University of Porto in1996 and received his MSc degree in ElectricalEngineering from the same University in 2004.

He is currently an Assistant Professor of ElectricPower Systems in the Polytechnic Institute of Porto.His research interests include Power SystemsOperation and Control, energy efficiency, loadresearch and electrical installations.

Luiz Faria Received his Ph.D. degree in ElectricalEngineering at University of Porto in 2002. He iscurrently Vice President of the Department ofInformatics in the Institute of Engineering of Porto ofthe Polytechnic Institute of Porto.

His research interests include Intelligent Tutoringsystems, Knowledge Based Systems, AdaptiveHypermedia and Interactive Problem Solving.

Sérgio Ramos graduated in the Polytechnic Instituteof Porto in 1999 and received his MSc degree fromthe Instituto Superior Técnico (Lisbon-Portugal) in2006.

He is currently an Assistant Professor of ElectricPower Systems in the Polytechnic Institute of Porto.His research interests include competitive electricitymarkets, energy efficiency, load research andelectrical installations.

Zita Vale graduated in the University of Porto in1986 and received her Ph.D degree in ElectricalEngineering from the same University in 1993.

She is currently a Coordinator Professor in thePolytechnic Institute of Porto. Her research areasinclude Power Systems Operation and Control,Electricity Markets, Decision Support and ArtificialIntelligence.