c3 feedstock optimization for multiproduct polypropylene...
TRANSCRIPT
![Page 1: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/1.jpg)
C3 Feedstock Optimization
for Multiproduct Polypropylene
Production
Pablo A. Marchetti, Ignacio E. Grossmann
Department of Chemical Engineering
Carnegie Mellon University
Wiley A. Bucey, Rita A. Majewski
Braskem America
Center for Advanced Process Decision-making
Enterprise-Wide Optimization (EWO) Meeting – March 12-13, 2013
![Page 2: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/2.jpg)
Project Overview
Polypropylene production facility
Chemical and refinery grade feedstocks with different prices and propylene purities.
Best operation will balance production rate with costs of feedstocks, maximizing plant throughput.
2
![Page 3: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/3.jpg)
Project Overview
Polypropylene production facility
Chemical and refinery grade feedstocks with different prices and propylene purities.
Best operation will balance production rate with costs of feedstocks, maximizing plant throughput.
Objectives: ◦ Development of a Non-linear Programming (NLP) model
to maximize benefits by obtaining a better balance of RG and CG feedstocks for single or multiple production orders.
◦ Determine operation rates for a schedule of multiple production orders within a 3-month timeframe.
◦ Implement user-friendly interface (GAMS model / MS-Excel)
2
![Page 4: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/4.jpg)
Process and Problem Description
Chemical
Grade (CG)
Refinery
Grade (RG)
Catalyst
Polypropylene
Propane return
Reactor
effluent Distillation
Polymerization
Feed Tank
Propylene (91%)
~79% propylene
~95% propylene
3
![Page 5: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/5.jpg)
Process and Problem Description
Chemical
Grade (CG)
Refinery
Grade (RG)
Catalyst
Polypropylene
Propane return
Reactor
effluent Distillation
Polymerization
Feed Tank
Propylene (91%)
~79% propylene
~95% propylene
3
Expensive
Cheaper
![Page 6: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/6.jpg)
Process and Problem Description
Chemical
Grade (CG)
Refinery
Grade (RG)
Catalyst
Polypropylene
Propane return
Reactor
effluent Distillation
Polymerization
Feed Tank
Propylene (91%)
Goal: Select optimal mix of chemical and refinery grade propylene
~79% propylene
~95% propylene
3
Expensive
Cheaper
![Page 7: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/7.jpg)
Mathematical Model (NLP)
Maximize Profit
Constraints on each time interval:
◦ Material balances
◦ Min/Max flow rates
◦ Constraints on composition of Propane Return, Distillation Overhead & Reactor Feed
◦ Limits on catalyst yield and flow
◦ Availability of Chemical Grade
◦ Specifications on splitter feed and recycle rate
Decision variables:
◦ Production rate of polypropylene
◦ RG and CG feedrates
◦ Distillation overhead flow and composition
◦ Reactor feed and catalyst flow
4
![Page 8: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/8.jpg)
Single/Multiple Product Models
5
![Page 9: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/9.jpg)
Single/Multiple Product Models
Single Product Model (one time interval)
◦ Maximize profit in terms of $/hr
◦ Best production rate with minimum cost of feedstocks.
Model size: 149 variables, 146 constraints
Solved with CONOPT and BARON in less than 1 CPU s.
5
![Page 10: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/10.jpg)
Single/Multiple Product Models
Single Product Model (one time interval)
◦ Maximize profit in terms of $/hr
◦ Best production rate with minimum cost of feedstocks.
Multiple Product Model ◦ Multiple orders of different products
◦ Production sequence given beforehand
◦ Profit ($) = selling prices – feedstock costs
+ propane return – others
◦ Solution gives best production rates with minimum costs for each
product
◦ Products of the same family feature same kinetic properties.
◦ Aggregation/disaggregation allows to handle large scale test cases.
Model size: 149 variables, 146 constraints
Solved with CONOPT and BARON in less than 1 CPU s.
Models implemented with GAMS 5
![Page 11: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/11.jpg)
Distillation Model
Objective: • Approximate procedure that provides overall treatment of the
distillation (no details about flows, composition, temperatures, etc. for each individual tray)
• The number of variables and constraints must remain small
6
![Page 12: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/12.jpg)
Distillation Model
Objective: • Approximate procedure that provides overall treatment of the
distillation (no details about flows, composition, temperatures, etc. for each individual tray)
• The number of variables and constraints must remain small
Aggregated group-method of Kamath et al. (2010)
L0
LN
V1
VN+1
L0 V1
LN VN+1
Tray-by-Tray Method
(Rigorous)
Group-Method
(Approximate)
Kamath, Grossmann and Biegler (2010)
Comp. and Chem. Eng. 34, pp. 1312-1319
6
![Page 13: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/13.jpg)
Distillation Model
Objective: • Approximate procedure that provides overall treatment of the
distillation (no details about flows, composition, temperatures, etc. for each individual tray)
• The number of variables and constraints must remain small
Aggregated group-method of Kamath et al. (2010)
L0
LN
V1
VN+1
L0 V1
LN VN+1
Tray-by-Tray Method
(Rigorous)
Group-Method
(Approximate)
Distillation Overhead
Bottoms
Feed
G1
G2
53% total trays
1 tray
47% total trays
Kamath, Grossmann and Biegler (2010)
Comp. and Chem. Eng. 34, pp. 1312-1319
C3 Splitter modeled
with Group-Method
6
![Page 14: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/14.jpg)
Initial linear correlation based on plant data
Distillation Model
Parameterization and Validation
7
![Page 15: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/15.jpg)
Initial linear correlation based on plant data
Distillation Model
Degrees of freedom:
• Reflux rate
• Bottoms composition
Additional Assumptions
• Fixed pressure for the whole
column = 9.778 atm
• Total condenser (top)
• Total reboiler (bottom)
• Single feed
Parameterization and Validation
7
![Page 16: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/16.jpg)
Initial linear correlation based on plant data
Distillation Model
Degrees of freedom:
• Reflux rate
• Bottoms composition
Additional Assumptions
• Fixed pressure for the whole
column = 9.778 atm
• Total condenser (top)
• Total reboiler (bottom)
• Single feed
Parameterization and Validation
Group-method
HySys simulations
Comparison of different column efficiencies against rigorous tray-by-tray simulations (Aspen HySys)
![Page 17: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/17.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 18: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/18.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 19: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/19.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 20: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/20.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
Schedule
requirements
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 21: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/21.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
Schedule
requirements
Aggregated
schedule
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 22: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/22.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
Schedule
requirements
Aggregated
schedule Initial solution
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 23: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/23.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
Schedule
requirements
Aggregated
schedule Initial solution Aggregated
multiple-product
solution
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 24: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/24.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
Schedule
requirements
Aggregated
schedule Initial solution Aggregated
multiple-product
solution
Detailed
schedule
results
• General results
• Detailed results
Aggregation /
disaggregation
procedure
8
![Page 25: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/25.jpg)
Multiple Product Model • Parameters
• Product and
product family data
• Schedule
Aggregate
products by
family
Disaggregate
results
Solve single-
product model
for each family
Solve multiple-
product model
Schedule
requirements
Aggregated
schedule Initial solution Aggregated
multiple-product
solution
Detailed
schedule
results
• General results
• Detailed results
Mid-size example (20 products, 5 families) Model size: 750 variables, 736 constraints
Solved by CONOPT in ~9 seconds.
Preliminary results show realistic tradeoff on feedstocks
costs vs production rates (depending on available time).
Aggregation /
disaggregation
procedure
8
![Page 26: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/26.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
9
![Page 27: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/27.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
9
![Page 28: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/28.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
Ref
9
![Page 29: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/29.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
Ref
9
20 products
31
![Page 30: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/30.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
Ref
9
![Page 31: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/31.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
31 % o
f M
ax P
rod
ucti
on
Rate
Ref
9
![Page 32: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/32.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
31 % o
f M
ax P
rod
ucti
on
Rate
Ref
9
28
![Page 33: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/33.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
28 31 % o
f M
ax P
rod
ucti
on
Rate
Ref
9
28
- 3days
![Page 34: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/34.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
28 31 % o
f M
ax P
rod
ucti
on
Rate
Ref
9
28
- 3days
34
![Page 35: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/35.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
34 % o
f M
ax P
rod
ucti
on
Rate
Ref
9
28
- 3days
34
+ 3days
31 28
![Page 36: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/36.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
31
Idle time
% o
f M
ax P
rod
ucti
on
Rate
Ref
9
28
- 3days
34
+ 3days
34 31 28
![Page 37: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/37.jpg)
Multiple Product Model - Example Different time horizons tested to measure the tradeoff of
feedstock costs versus production rates.
100.0 MU 103.2 MU 88.1 MU
Propylene Source
Optimal Solution
(PROFIT)
9
31
Idle time
Ref
28
- 3days
34
+ 3days
![Page 38: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/38.jpg)
Multiple Product Model - Example Added “slack” product to assess the benefits of extra production
when schedule finishes early gain/loss opportunity
10
![Page 39: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/39.jpg)
Multiple Product Model - Example Added “slack” product to assess the benefits of extra production
when schedule finishes early gain/loss opportunity
31
31
10
Ref
% o
f M
ax P
rod
ucti
on
Rate
![Page 40: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/40.jpg)
Multiple Product Model - Example Added “slack” product to assess the benefits of extra production
when schedule finishes early gain/loss opportunity
31
31
10
Ref
+ slack
prod
% o
f M
ax P
rod
ucti
on
Rate
![Page 41: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/41.jpg)
Multiple Product Model - Example Added “slack” product to assess the benefits of extra production
when schedule finishes early gain/loss opportunity
100.0 MU 109.1 MU
Propylene Source
Optimal Solution
(PROFIT)
10
31
Ref
+ slack
prod
![Page 42: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/42.jpg)
User Interface via Excel Worksheet User interface for GAMS multiple-product model developed
in MS Excel
Flexibility to easily test different production schedules with
alternative parameters.
11
![Page 43: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/43.jpg)
User Interface via Excel Worksheet User interface for GAMS multiple-product model developed
in MS Excel
Flexibility to easily test different production schedules with
alternative parameters.
11
![Page 44: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/44.jpg)
User Interface via Excel Worksheet User interface for GAMS multiple-product model developed
in MS Excel
Flexibility to easily test different production schedules with
alternative parameters.
11
![Page 45: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/45.jpg)
User Interface via Excel Worksheet User interface for GAMS multiple-product model developed
in MS Excel
Flexibility to easily test different production schedules with
alternative parameters.
NLP Model
GAMS
GDX Files
Excel - VBA
11
![Page 46: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/46.jpg)
Conclusions and Future Work
CONCLUSIONS
Multiple-product feedstock optimization nonlinear programming model developed. Process models include distillation and polymerization units.
Proposed method handles gain/loss scenarios and large schedules (through aggregation/disaggregation).
Distillation model formulated using aggregated group-method based on work of Kamath et al. 2010.
Deployment of computational tool to assess monthly feedstock purchase decisions.
Initial tests show large potential for savings in feedstock cost.
12
![Page 47: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/47.jpg)
Conclusions and Future Work
CONCLUSIONS
Multiple-product feedstock optimization nonlinear programming model developed. Process models include distillation and polymerization units.
Proposed method handles gain/loss scenarios and large schedules (through aggregation/disaggregation).
Distillation model formulated using aggregated group-method based on work of Kamath et al. 2010.
Deployment of computational tool to assess monthly feedstock purchase decisions.
Initial tests show large potential for savings in feedstock cost.
FUTURE WORK
Improvements on distillation model parameters.
12
![Page 48: C3 Feedstock Optimization for Multiproduct Polypropylene ...egon.cheme.cmu.edu/ewocp/docs/Braskem_Grossmann_3_2013.pdf · C3 Feedstock Optimization for Multiproduct Polypropylene](https://reader035.vdocuments.mx/reader035/viewer/2022070613/5b9a795309d3f21b2f8b745d/html5/thumbnails/48.jpg)
C3 Feedstock Optimization
for Multiproduct Polypropylene
Production Pablo A. Marchetti, Ignacio E. Grossmann
Department of Chemical Engineering
Carnegie Mellon University
Wiley A. Bucey, Rita A. Majewski
Braskem America
Thanks for your attention!
Questions?