crushing flowsheet simulation: increased productivity and ... · flowsheet design. with the...
TRANSCRIPT
O'BRYAN, K. Crushing flowsheet simulation: Increased productivity and improved flowsheet design. APCOM 87. Proceedings of the Twentieth International Symposium on the Application of Computers and Mathematics
in the Mineral Industries. Volume 2: Metallurgy. Johannesburg, SAIMM, 1987. pp. 167 - 178.
Crushing Flowsheet Simulation: Increased Productivity and Improved Flowsheet Design
K. Q'BRYAN
Process Machinery Division, Nordberg Inc., Milwaukee, WI, USA
A lO-year program to integrate crushing and screening models into daily flowsheet design practice has proven successful. A Circuit Analysis Program (CAP) is used worldwide to design and evaluate new and existing crushing flowsheets. To gain user acceptance of CAP and maximize effectiveness, model development, simulation flexibility, ease of use, and developing international access are key issues. These issues have been resolved effectively resulting in flowsheet simulations to assist in justifying increased capital investment. CAP is presently utilized in Canada, United States, and South Africa with intentions
to expand to Europe and Asia.
Introduction
A considerable amount of academic research
has been conducted to develop accurate
models of unit processes in the mineral
industries. However, the industrial appli
cation of these models to simulation of
plant flowsheets has been limited. In
1983, Nordberg Machinery, a manufacturer of
crushing, milling and screening equipment,
successfully implemented a computer program
for the simulation of crushing and
flowsheets. This Circuit screening
Analysis
worldwide
Program (CAP) is utilized
and/or for the development
analysis of crushing plant flowheets.
CAP has enabled Nordberg to improve both
worker productivity and the quality of
flowsheet design. With the information
generated by CAP, customers can be more
confident of a given flowsheet's predicted
performance. This information can be
utilized by the customer to financially
justify capital expenditures based on CAP's
simulation predictions.
cations include:
Typical appli-
1. New Flowsheet design
2. Study and modification of existing flow
sheets
3. Plant studies for evaluation of automa-
tion potential.
The development of CAP encompassed
several years, and its successful implemen
tation was a result of resolving problems
related to model development, simulation
flexibility, ease of use, and international
access. The context of this paper will
focus on these areas and conclude with a
case study of a recent flowsheet design and
plant study.
Model development
The model development is a result of data
collected from laboratory tests and field
data collected on crushing and screening
equipment. A generalized screening model
(figure 1) similar to one presented by
Karra (1979) has been developed. As shown
in figure 1, a screening oversize partition
curve is developed from the algorithm:
ESTIMATING GRADE DIFFERENTIALS FROM DRILLING RESULTS 167
100
90 OJ
.!:: 80 '" ... OJ .... 0 70
1:1 OJ OJ ... 60 y
00 Q ..- 50 ;..tIJ!lIII'.4I1lII'.4IIi!iY.4Ii!IIT.JIIIIl!r"".4!IIIT,4IJIJII"JiII1IT.4Ii!IITJIfIII!II'..JI!IIlIII":.
eJI 1:1 .• .... 40 ... Q
=-OJ ~ 30 .... == OJ Y 20 ... OJ ~
10
0
.4 .6 .8 1.0 1.2 1.4
FIGURE 1. Normalized oversize position curve for screen model
where: %C i is the oversize partition
coefficient,
Di is the geometric mean particle
size
d50 is the mean particle size
where 50% passes
Kl is a statistically determined
constant related to the designed
screen dynamics.
Each size gradation is separated via the
composition of this partition curve. The
screen model allows CAP to account for
screen inefficiency which varies depending
on the amount of near size material, fine
ness of separation, and surface character-
istics of the material. Screens are sized
relative to internationally accepted screen
sizing practice.
Field data has been collected to
determine the performance of Gyratory,
Cone, Gyradisc, Jaw and Impact crushers.
CAP's crushing models enable simulation of
168
Feed Data
~ Crusher Model
PARAMETERS
Type of Machine
Gap/Setting
Cham ber Design
Crushing Characteristics of Material
Production Data
FIGURE 2. Generalized concept of crushing model used by CAP
all these crushing devices. _.As shown in
figure 2, the crushing performance of a
given machine is determined by the user
selected liner, crusher setting, throw, and
type of crusher. Connected horsepower,
speed, and throw are generally assumed
constant for a given size and type of
machine. Provisions have been made in the
program to
agglomeration
provided the
include
device
any
in
crushing or
the circuit
gradat ion and- production
characteristics of that device are known.
The crushing and screening models within
CAP provide satisfactory estimates of
crushing and screening performance. The
most confident model predictions are based
on actual field performance of existing
equipment for a given material. Model
development for CAP will continue as addi
tional data is collected.
Simulation flexibility
Lack of simulation flexibility is a common
problem of many commercially available
simulation packages. Customers, particu-
METALLURGY: MODELLING
1ar1y those customers producing aggregate
products, utilize a wide variety of flow-
sheet configurations. In addition, two
plants producing identical products can be
designed with widely varying f10wsheets.
The speed with which a f10wsheet can be
simulated enables a user to examine several
f10wsheet options. The final design is
determined by economics, company practice,
individual preference, and technical per-
formance. For successful f10wsheet simu-
1ation of crushing and screening circuits,
CAP's simulation flexibility is an absolute
requi rement.
Simulation flexibility is the principal
technical strength of CAP. With this flow
sheet simulator users, in a matter of
minutes for each simulation, can conduct a
performance analysis of changing operating
conditions.
flowsheet
CAP is used for 99% of all
analysis by North American
users. Typical examples include:
1. The increased production from higher
performance crushers.
2. The effect of modifying screen perfor
mance on crusher loading.
3. Reduced circulating loads from tighter
crusher settings.
4. Finer products from tighter crusher
settings.
S. Creating different products by modifying
material flows.
6. Identification of f10wsheet bottlenecks.
7. Effect of material crushing character
istics on circuit performance.
A crushing/screening f10wsheet can be
simulated by dividing the f10wsheet into
simulation stages. Each stage can contain
1-100 identical crushers and 1-100
identical screens. A maximum of eight
s imu1at ion s toages are allowed. Each stage
has the ability to simulate splitters for
splitting material flows into multiple
streams. Material flows can be transferred
directly from stage to stage, deposited in
one of eight available product bins or nine
available surge bins, or recircu1ated to a
previously simulated component. By
combining surge and product bins, a total
of 17 different final products can be
created.
Ease of use
The potential users of CAP are application
and sales engineers. Many of these users
have minimal experience with computers or
computer models. Successful development of
CAP requires user-friendly input and error
prevention procedures. Early efforts to
implement these models at Nordberg failed
mainly due to difficult execution proc
edures.
Data processing experts were consulted to
develop easier to use input and output
Create Composite Data COBOL Software File
Simulate Flowsheet Performance
FORTRAN Models
Edit Final Output
FIGURE 3. Final structure of CAP for USA users. COBOL programs handle input/output and FORTRAN program models flowsheet
ESTIMA TING GRADE DIFFERENTIALS FROM DRILLING RESULTS 169
formats. CAP was transferred from a time-
sharing CDC mainframe to Nordberg's
Milwaukee IBM mainframe. The Milwaukee
mainframe's VMSP system provided full
screen edit capabilities for creating data
files. The data processing staff developed
COBOL software for data handling and error
checking.
With the use of full-screen editing and
the development of error checking COBOL
software, the final structure (figure 3) of
CAP was completed. These enhancements
transformed CAP from TTY batch operation to
a full-screen interactive program.
result, user acceptance was achieved.
International access
As a
After CAP had been successfully implemented
in Milwaukee, a study was conducted
regarding international access to CAP. Two
locations, Guelph, Canada and Johannesburg,
South Africa agreed to test and implement
CAP.
A review of problems associated with
international access identified key areas
of concern, specifically:
1. Program security.
2. Establishing communication links.
3. Economic costs.
4. Ease of use and troubleshooting.
As international access became a
priority, maintaining the security of CAP
became the key issue. Recent developments
in personal computer tachnology made
available the possibility of converting the
entire software for execution on a PC.
However, maintaining program security would
be almost impossible with the current
technology. The PC option was quickly
rejected.
The final choice for maintaining the
maximum program security became establish-
170
ing telecommunications links to a central
ized time-sharing mainframe. International
users would be provided execution~only
acess through these telecommunication
links. International telecommunication
links were established through TYMNET in
North America, SAPONET in South Africa and
DATAPAC in Canada.
Full screen access of CAP was established
in Canada and South Africa, however,
several deficiencies were identified:
1. High cost,
2. Hardcopy transmittal was of poor
quality,
3. Inability to edit CAP output.
The high transmission costs resulted from
transferring the peripheral data associated
with the full-screen mode. This problem
was especially critical during data input
when 90% of the data transmitted was
in nature. Transmission of descriptive
only data
simulator
required for the flowsheet
would reduce both data trans-
mission and execution costs.
Downloading of data to a remote location
via full-screen mode was unsatisfactory.
CAP outputs were difficult to read and
impossible to edit.
discussions with
achievable.
Use of CAP outputs in
customers was not
A second study determined that the above
problems experienced by international users
could be solved by converting the communi
cations link from full-screen mode to TTY
mode. In the TTY mode, a personal computer
with a full-screen editor 1s utilized for
executing the COBOL input programs and for
editing the simulation output. The
simulation/data transfer environment for
international users is depicted in figure
4.
With the transition of using a personal
METALLURGY: MODELLING
FIGURE 4. International users' simulation/data transfer environment. Personal computer acts as an intelligent terminal for input data development and output editing
computer as an intelligent terminal,
communications and execution costs were
reduced an average of 75%. Subsequent
enhancements include 1) on-line help
facility, 2) File handling utility, and 3)
Electronic mail
problems.
system for reporting
Case study: 450 tph aggregate plant
A North American consulting company is
responsible for the design and construction
of a basalt aggregate plant. Production
specifications and requirements forwarded
to Nordberg Application Engineers included:
375 Metric Tons per hour (MTPH) of
32mm by 0 blend.
40 Metric Tons per hour (MTPH) of
17mm by 0 blend.
35 Metric Tons per hour (MTPH) of
17mm by 10 mesh blend.
The initial request from the consultant
was for quotation of a Jaw/Cone crushing
plant to produce the desired products.
Earlier manual estimates by competitive
manufacturers indicated that the plant
production could be met with one 1109 Jaw
Crusher and two 1.3 meter cone crushers.
To fully determine the equipment required
to meet product specifications, CAP was
utilized to determine plant performance
under two options, specifically:
Option 1. One 1109 VB Jaw Crusher
Two 1560 Omnicone Cone Crushers
Base. design on 90 to 85%
screening efficenicy.
(see flowsheet in figure 5)
Option 2. One 1109 VB Jaw Crusher
Two 1560 Omnicone Cone Crushers
'One 48 inch (1.2 meter) Nordberg
Gyradisc
Base design on 90 to 85%
screening efficiency. (see
flowsheet in figure 6)
ESTIMATING GRADE DIFFERENTIALS FROM DRILLING RESULTS 171
......
-:J
N ~
tTJ ~ t""
" t""
" C
:;d o ~ ~ o tJ
tTJ
t""" t"""
....... Z o
\ h
o
-=::J
1.(
. 110
9 J
AW
42
5
TP
H
3S
S
TP
H
32
mm
x
Om
m
15
60
O
MN
ICO
NE
S
TA
ND
AR
D
31
3
TP
H
17
mm
X
1
0
me
sh
o 1
0.7
TP
H
17
mm
x
10
m
esh
FIG
UR
E 5
. T
erti
ary
crus
hing
cir
cuit
wit
h pr
oduc
tion
cap
abil
ity
of
425
tph
of
basa
lt a
ggre
gate
s
-r
15
60
O
MN
ICO
NE
S
HO
RT
HE
AD
23
8
TP
H
rn if!
>-3 ~ >-3 Z
o o ~ tJ
rn tJ
......
"Ij
"Ij rn § Z
>-3 ;; t""
" if
! "I
j :::0
o ~ tJ
:::0 P t""
" ......
Z o :::0 rn if! C
t"""
>-3
if! - --.l w
): I
45
0
TP
H
~ 0
10
0m
m..
....
. .
.
y \'1
63
mm
~~
I ~~W"
09 32mm~
. ..I.
~---..
37
4
TP
H
17mm~-
••
9.5
mm
__
__
_ --..
....
...:
.:
4. 7
mm
__
~
JIB
X
~
'1
"34
T
PH
L 2
20
T
PH
17
mm
x
10
m
esh
48
" G
YR
AD
ISC
3
78
T
PH
3
8
TP
H
6 3
2m
m
x O
mm
F
ine
s
Q
22
T
PH
27
T
PH
FIG
UR
E 6
. Q
uart
enar
y cr
ushi
ng c
ircu
it w
ith
prod
ucti
on c
apab
ilit
y ex
ceed
ing
450
tph
of
basa
lt a
ggre
gate
s
Option 1 was simulated at the request of
the consulting contractor. Option 2 was
simulated as the Nordberg recommended
flowsheet design to guarantee meeting the
customer's requirements. The simulated
performance comparison of the two options
is outlined in Tables I and I!. The CAP
simulation of the quaternary stage for
option 2 is presented in the attached
Appendix.
The data in Table I indicates total plant
productivity and compares individual
crusher performance for options 1 and 2.
This data summary clearly indicates that
option 2, the quaternary crushing circuit,
is required to meet the customer's
production requirements. Option 1 could
not produce enough fine aggregate to meet
product specifications and achieved a
maximum capacity of 425 TPH. Option 2,
with the additional quaternary crushing
stage, met the customers's production
TABLE I
174
Production Comparison on Flowsheet Simulation Options. Option 1 is a threestage crushing circuit with a flowsheet as shown in figure 5. Option 2 is a four-stage crushing circuit with a flowsheet as shown in figure 6.
Run of Quarry Feedrate:
Primary Scalper Performance Scalper Oversize (635mm x 100mm): Scalper Undersize (-200mm x Omm):
VB 1109 Jaw Crusher Performance Closed Side Setting: Total Feedrate: 80% passing of product: % of Full Load Capacity:
Option 1
425 TPH
277 TPH 148 TPH
155 mm 277 TPH 174 mm
99 %
1560 Omnicone Cone Crusher Performance (Secondary) Closed Side Setting: New Feedrate: Total Feedrate: 80% passing of Product: % of Full Load Capacity:
1560 Omnicone Cone Crusher Performance (Tertiary) Closed Side Setting: New Feedrate: Total Feedrate: 80% passing of Product: % of Full Load Capacity:
48 INCH GYRADISC Performance (Quaternary) Closed Side Setting: New Feedrate: Total Feedrate: 80% passing of Product: % of Full Load Capacity:
Plant Production Specified Products
32mm by 0 blend: 17mm by 0 blend: 17mm by 10 mesh blend:
30 mm 296 TPH 313 TPH
37 mm 88 %
13 mm 166 TPH 238 TPH
17 mm 97 %
387 TPH o TPH
38 TPH
Total 425 TPH
Option 2
450 TPH
294 TPH 156 TPH
165 mm 294 TPH 186 mm
98%
38 mm 317 TPH 374 TPH
48 mm 98 %
13 mm 177 TPH 220 TPH 17 mm
95 %
13 mm 48 TPH 93 TPH
8.4 mm 70 %
378 TPH 34 TPH 38 TPH
450 TPH
METALLURGY: MODELLING
fi
requirement with extra capacity to produce
additional cubical fine aggregates.
In this case study, the use of a Circuit
Analysis Program had distinct advantages.
First, CAP simulations enabled Nordberg to
identify the absolute requirement for 1.5
meter cone crushers rather than 1.3 meter
cone crushers originally proposed. As a
result, the customer could be assured of
meeting his plant production objectives.
Second, the entire quotation process
including examination of several flowsheet
options was completed in less than eight
hours. A similar analysis without a
Circuit Analysis Program would have
required several days of work with a lower
confidence level in final plant perform-
ance.
TABLE II
Screening Performance Comparison of Flowsheet Simulation Options. Option 1 is a three-stage crushing circuit with a flowsheet as shown in figure 5. Option 2 is a four-stage crushing circuit with a flowsheet as shown in figure 6.
Run of Quarry Feedrate
Primary Scalper Performance: Scalper Oversize (635mm x 100mm): Scalper Undersize (-200mm x Omm):
Secondary Screening Performance: Deck One
Mesh Opening Screening Efficiency Area Required
Deck Two Mesh Opening Screening Efficiency Area Required
Deck Three Mesh Opening Screening Efficiency Area Required
Tertiary Screening Performance: Deck One
Mesh Opening Screening Efficiency Area Required
Deck Two Mesh Opening Screening Efficiency Area Required
Deck Three 88 % Mesh Opening Screening Efficiency Area Required
Quaternary Screening Performance: Deck One
Mesh Opening Screening Efficiency Area Required
Deck Two Mesh Opening Screening Efficiency Area Required
Option 1 425 TPH
277 TPH 148 TPH
100 mm 95 %
5.7 m2
63 mm 92 %
6.9 m2
32 mm 89 %
11.8 m2
17 mm 84 %
10.8 m2
9.5 mm 83 %
6.5 m2
4.7 mm 88 %
6.5 m2
ESTIMATING GRADE DIFFERENTIALS FROM DRILLING RESULTS
Option 2 450 TPH
294 TPH 156 TPH
100 mm 95 %
6.8 m2
63 mm 89 %
8.8 m2
32 mm 90 %
12.6 m2
17 mm 86 %
11.4 m2
9.5 mm 83 %
8.0 m2
4.7 mm 88 %
6.7 m2
6mm 94 %
4.1 m2
3.3 mm 91 %
4.5 m2
175
Conclusions The successful implementation of computer
models in the industrial environment was
achieved through addressing user needs.
The program's simulation flexibility
coupled with its ease of use has helped CAP
gain a wide acceptance. As a result, this
flowsheet simulation technology is being
utilized on a daily basis for better
flowsheet design.
CAP's worldwide implementation was
accomplished while meeting security and
cost-effectiveness considerations. The key
roles of international telecommunications
networks <;lnd personal computers acting as
intelligent terminals enabled preservation
of key features and allowed additional
enhancements.
CAP is presently being utilized in the
United States, Canada and South Africa.
Applications include plant studies,
flowsheet design, and process control
evaluation. A future objective is to
extend CAP's use to Brazil, Europe and
Australia.
Note CAP (Circuit Analysis Program) is
protected
federal
by United
copyright
States
law.
Reproduction and/or use of CAP
without the express written
permission of Rexnord Inc. is
prohibited.
Reference
Karra, V .K. , "Development of a Model for
Predicting the Screening Performance of a
Vibrating Screen", CIM Bulletin, April,
1979.
Appendix: ACAP simulation output of quaternary crushing circuit
DATE - 06/25/87 TIME - 09:06 P AGE 10
176
STAGE #4 CRUSHING STAGE CONTAINS ( 1) 48 INCH GYRADISC CRUSHER
C R U SHE R SET U P
SETTING = 1.30 CM. LINER TYPE=FINE CIRCUITRY =CLOSED
PRODUCT ANALYSIS
(CM. )
1. 30 .95 .70 .60 .47
6M 10M 14M
P80 =
MATERIAL PASSING (PERCT)
100.0 89.0 74.5 68.0 59.0 46.1 36.2 28.0
.84 CENTIMETERS.
I N D I V I D U A L C R U S H E R C R U S H E R C A P A C I T Y I N D I C E S ..................... . .................. NEW FEED = 48 TPH CAPACITY INDEX= 70% TOTAL FEED= 93 TPH MAX. FEED = 132
PRIMARY DESTN: STAGE #4 (S) AMOUNT: 93 ( TPH)
.C R U SHE R DES TIN A T ION S
METALLURGY: MODELLING
DATE - 06/25/87 TIME - 09:06 P AGE 11
SCREEN SIZING AND PERFORMANCE ANALYSIS --- STAGE #4 ***************************************************
THEORETICAL AREA REQUIRED.
DECK
1 2
BELOW IS A TABLE GIVING THE SCREEN SIZING FACTORS USED TO DETERMINE THE THEORETICAL AREA OF THIS SCREEN. DEFINITION OF TERMS:
OPENING CM.
-------.600 .333
"OPENING" - EFFECTIVE OPENING OF SCREEN MESH. "A" - STPH PASSING A SQ. FT. OF SCREEN SURFACE. "B" - PERCENT OF OVERSIZE IN FEED TO DECK FACTOR. "c" - PERCENT HALF SIZE IN FEED TO DECK FACTOR. "D" - DECK LOCATION FACTOR. "E" - WET SCREENING FACTOR. "F" - MATERIAL WEIGHT FACTOR. "G" - SCREEN SURFACE OPEN AREA FACTOR. "H" - SHAPE OF SURFACE OPENING FACTOR. "I" - SCREEN SLOPE FACTOR.
"TUST" - THEORETICAL UNDERSIZE TONNAGE. "AREA" - THEORETICAL AREA REQUIRED.
----------SCREEN SIZING FACTORS--------A B C D E F G H I
1. 03 1. 22 1. 27 1. 00 1. 00 1. 00 1. 00 1. 00 1. 00 .72 1. 21 1. 25 0.90 1. 00 1. 00 1. 00 1. 00 1. 00
TUST MTPH
-------63.0 42.7
AREA SQ. M. ------
4.1 4.5
A 1.2 X 3.7 METER 2-DECK SCREEN SELECTED FOR THE APPLICATION
DECK#
1 2
FEED
93 58
OVERSZ
35 19
UNDRSZ
58 39
EFFICY(%)
92 91
DBF
2.4 1.6
ESTIMATING GRADE DIFFERENTIALS FROM DRILLING RESULTS
DBD
.9
.5
MAX-DBD
2.4 1.3
177
DATE - 06/25/87 TIME - 09:06 P AGE 12
STAGE #4 SCREENING STAGE CONTAINS 1 SCREEN(S) WITH 2 DECK(S).
NEW FEED 48 TOTAL FEED 93 -------------- --------- --------- ---------SCREENING DECK #1 DECK #2 DECK #2 PARAMETERS OVERSIZE OVERSIZE UNDERSIZE -------------- --------- --------- ---------OPENING (CM. ) .600 .333 TONNAGE ( TPH) 35 19 39 PROBABILITY(%) 90 90 EFFICIENCY (% ) 91 90
FEED SCREEN SCREEN SCREEN ANALYSIS PRODUCT PRODUCT PRODUCT -------------- --------- --------- ---------
PRCT PRCT PRCT PRCT INCH MM PASS PASS PASS PASS
.512 13 100.0 100
.374 9 89.0 71
.276 7.0 74.5 32
.236 6.0 68.0 15 100
.185 4.7 59.0 4 76 6M 3.3 46.1 0 21 100 8M 2.4 36.2 3 85
10M 1.7 28.0 0 67 14M 1.2 21. 8 52 20M .8 17 .1 41 28M .6 14.2 34 35M .4 11. 8 28 48M .3 10.2 24 65M .2 8.3 20
-------------- --------- --------- ---------
.. S P LIT S C R E E N ANALySIS ....... .
D E C K # 1 D E C K # 2 D E C K # 2 o V E R S Z o V E R S Z U N D R S Z ------------ ------------ ------------
PRIMARY DESTN: STAGE #4 (C) STAGE #4 ( C) STOCK BIN(3) AMOUNT: 35 ( TPH) 10 ( TPH) 12 ( TPH)
SPLIT 1 DESTN: STOCK BIN(3) STOCK BIN(l) AMOUNT: 10 ( TPH) 27 ( TPH)
------------ ------------ ------------
178 METALLURGY: MODELLING