Building Cogeneration Planning Scheduling Systems using IBM ILOG ODME, CPLEX and impress
Building Cogeneration Planning
Applications using IBM ODME
DecisionBrain & Industrial Algorithms LLC.
Copyright, DB & IAL
• What is ODME?
• What are Industrial Modeling Frameworks?
• What is iMPress?
• ODME-iMPress Implementation
• Proof of Concept
Based on IBM ILOG Optimization
Engines and Tools
High-performance mathematical and constraint programming solvers, modeling
language, and development environment
Build and deploy analytical decision support applications based on optimization
Oil&Gas Production Scheduling
ILOG ODM Enterprise
Embeds all CPLEX Optimization Studio
ODM Enterprise IDE
Client & Planner
Application UI Configuration
Application UI Customization
Industrial Modeling Frameworks
• Process industry business problems are
complex hence an iMF provides a pre-project
or pre-solution advantage (head-start).
• An iMF embeds intellectual-property and
know-how related to the process’s flowsheet
modeling as well as its problem-solving
• iMPress stands for “Industrial Modeling &
PRE-Solving System” and is our proprietary
platform for discrete and nonlinear modeling.
• iMPress can “interface”, “interact”, “model”
and “solve” any production-chain, supply-
chain, demand-chain and/or value-chain
• Off-Line Environments:
– Usually « dynamic » optimization with discrete
(logic) & linear variables using Mixed Integer
Linear Programming not including feedback
• On-Line Environments:
– Usually « steady-state » optimization with
continuous & nonlinear variables using NLP
including feedback (and feedforward).
– Usually includes steady-state detection, data
reconciliation and regression (« moving horizon
estimation ») with diagnostics for monitoring.
• Sometimes called « load shedding, shifting &
– Determines steam and power production subject to
supply availability and demand requirements.
– Respects transition (sequence-dependent)
management of producing units such as boilers and
turbogenerators i.e., understands resting (standby),
ramping (startup/shutdown) and running (setup)
which IAL calls « Phasing ».
– Similar to a « product wheel » found in specialty batch
& fast moving consumer goods industries.
Off-Line Optimization –
« Phasing »
• « Phasing » forces a predictable operational
sequence or order for selected units.
• Typically assumes discrete/logic variables are
fixed – IAL calls this « phenomenological
• If plant is at « steady-state* » then optimize
process or operating conditions using NLP
(IPOPT, KNITRO, XPRESS-SLP, IAL-
• Apply nonlinear data reconciliation &
parameter estimation to provide gross-
error/outlier detection & calibrate model.
* Kelly & Hedengren, « A steady-state detection algorithm to detect non-stationary drifts in processes », Journal of Process Control,
• An important aspect is to callout/callback to
physical/thermodynamic properties such as
– STEAM67.DLL is « wrapped » in
STEAM67_H.DLL to compute saturated enthalpy
and its first-order derivatives using its saturated
FBAL,HOTF + COLDF - WARMF
HFBAL,HOTH*HOTF + COLDH*COLDF - WARMH*WARMF
Cogeneration (Steam/Power) iMf
Copyright, Industrial Algorithms LLC
• Time Horizon: 168 time-periods w/ hour
• Continuous Variables = 5,000
• Binary Variables = 1,000
• Constraints = 7,500
• Time to First Good Solution = 5 to 30-
• Time to Provably Optimal = 5 to 15-minutes
• A domain-specific data model was created in
ODME using the usual master-data and
• A mapping between iMPress’ data model and
ODME’s data model was established.
• Java code was written to export iMPress’ IML
file (Industrial Modeling Language).
• SWIG Java was used to create a Java Native
Inerface (JNI) to iMPress.
• Java code was written to call iMPress-CPLEX
using its API’s.
• Java code was written to access the solution(s)
from iMPress-CPLEX using its API’s and to
populate the ODME solution-data partition.
Demand Variability Scenario Data w/
Reference in ()
Trend Plots for Demand Variability
Scenario w/ Reference
• Perfectly fit your business model and decision processes
• Sophisticated optimization capabilities able to tackle complex,
non-linear and large-scale problems
• A solution that can be quickly adapted to new production
• A user-friendly GUI to help planners driving refinery operational
excellence and analyzing refinery behavior
• What-if scenario analysis for confident decision-making
• See all your data and options in one place with drill-downs and
• Collaborate with other planners
• Powered by IBM ILOG CPLEX Optimizers
• Select plant type, size and complexity.
• Determine if off-line or on-line
• Configure plant model.
• Integrate data sources.
• Solve plant model with plant data.
• Tune plant model (for accuracy &