A machine learning-based methodology for multi-parametric solution of chemical processes operation optimization under uncertainty

Abstract Chemical process operation optimization aims at obtaining the optimal operating set-points by real-time solution of an optimization problem that embeds a steady-state model of the process. This task is challenged by unavoidable Uncertain Parameters (UPs) variations. MultiParametric Programming (MPP) is an approach for solving this challenge, where the optimal set-points must be updated online, reacting to sudden changes in the UPs. MPP provides algebraic functions describing the optimal solution as a function of the UPs, which allows alleviating large computational cost required for solving the optimization problem each time the UPs values vary. However, MPP applicability requires a well-constructed mathematical model of the process, which is not suited for process operation optimization, where complex, highly nonlinear and/or black-box models are usually used. To tackle this issue, this paper proposes a machine learning-based methodology for multiparametric solution of continuous optimization problems. The methodology relies on the offline development of data-driven models that accurately approximate the multiparametric behavior of the optimal solution over the UPs space. The models are developed using data generated by running the optimization using the original complex process model under different UPs values. The models are, then, used online to, quickly, predict the optimal solutions in response to UPs variation. The methodology is applied to benchmark examples and two case studies of process operation optimization. The results demonstrate the methodology effectiveness in terms of high prediction accuracy (less than 1% of NRMSE, in most cases), robustness to deal with problems of different natures (linear, bilinear, quadratic, nonlinear and/or black boxes) and significant reduction in the complexity of the solution procedure compared to traditional approaches (a minimum of 67% reduction in the optimization time).

[1]  Megan Jobson,et al.  Optimization of Heat-Integrated Crude Oil Distillation Systems. Part I: The Distillation Model , 2015 .

[2]  Hongye Su,et al.  Optimization of refinery hydrogen network based on chance constrained programming , 2012 .

[3]  Efstratios N. Pistikopoulos,et al.  Integrating deep learning models and multiparametric programming , 2020, Comput. Chem. Eng..

[4]  Dale D Slaback,et al.  A surrogate model approach to refinery-wide optimization , 2004 .

[5]  Peter Denno,et al.  Standards-based integration of advanced process control and optimization , 2019, J. Ind. Inf. Integr..

[6]  M. Ierapetritou,et al.  A kriging method for the solution of nonlinear programs with black‐box functions , 2007 .

[7]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[8]  Alberto Bemporad,et al.  The explicit linear quadratic regulator for constrained systems , 2003, Autom..

[9]  Alberto Bellini,et al.  Design for Reliability: The Case of Fractional-Slot Surface Permanent-Magnet Machines , 2019 .

[10]  M. Shuhaimi,et al.  A chance constrained approach for a gas processing plant with uncertain feed conditions , 2010, Comput. Chem. Eng..

[11]  Antonio Espuña,et al.  Application of the Meta-Multiparametric methodology to the control of emissions in the industry under continuous and discrete uncertain parameters , 2016 .

[12]  José A. Caballero,et al.  Hybrid simulation-equation based synthesis of chemical processes , 2018 .

[13]  Efstratios N. Pistikopoulos,et al.  On multi-parametric programming and its applications in process systems engineering , 2016 .

[14]  Christos T. Maravelias,et al.  Surrogate‐based superstructure optimization framework , 2011 .

[15]  Antonio Espuña,et al.  USING SURROGATE MODELS FOR PROCESS DESIGN AND OPTIMIZATION , 2012 .

[16]  José Antonio Caballero,et al.  Large scale optimization of a sour water stripping plant using surrogate models , 2016, Comput. Chem. Eng..

[17]  Lorenz T. Biegler,et al.  Nonlinear Waves in Integrable and Nonintegrable Systems , 2018 .

[18]  Vladimir Jovan,et al.  An approach to process production reactive scheduling. , 2004, ISA transactions.

[19]  Ignacio E. Grossmann,et al.  Recent advances in mathematical programming techniques for the optimization of process systems under uncertainty , 2015, Comput. Chem. Eng..

[20]  Andy J. Keane,et al.  Recent advances in surrogate-based optimization , 2009 .

[21]  Antonio Espuña,et al.  Ordinary Kriging: A machine learning tool applied to mixed-integer multiparametric approach , 2018 .

[22]  Gabriele Pannocchia,et al.  A Modifier-Adaptation Strategy towards Offset-Free Economic MPC , 2016 .

[23]  Zukui Li,et al.  Process operations with uncertainty and integration considerations , 2010 .

[24]  Efstratios N. Pistikopoulos,et al.  Combined model approximation techniques and multiparametric programming for explicit nonlinear model predictive control , 2012, Comput. Chem. Eng..

[25]  Antonio Ferramosca,et al.  Steady-state target optimization designs for integrating real-time optimization and model predictive control , 2014 .

[26]  Lluvia M. Ochoa-Estopier,et al.  Industrial application of surrogate models to optimize crude oil distillation units , 2018 .

[27]  I. Grossmann,et al.  An algorithm for the use of surrogate models in modular flowsheet optimization , 2008 .

[28]  Matthew J. Realff,et al.  Optimization and Validation of Steady-State Flowsheet Simulation Metamodels , 2002 .

[29]  Tao Chen,et al.  Meta-modelling in chemical process system engineering , 2017 .

[30]  Efstratios N. Pistikopoulos,et al.  Explicit/multi-parametric model predictive control (MPC) of linear discrete-time systems by dynamic and multi-parametric programming , 2011, Autom..

[31]  M. Morari,et al.  On-line optimization via off-line parametric optimization tools , 2000 .

[32]  Jeffrey D. Kelly,et al.  Unit-operation nonlinear modeling for planning and scheduling applications , 2017 .

[33]  Efstratios N. Pistikopoulos,et al.  Recent advances in multiparametric nonlinear programming , 2010, Comput. Chem. Eng..

[34]  Antonio Espuña,et al.  Dynamic Optimization of Batch Processes under Uncertainty via Meta-MultiParametric Approach , 2017 .

[35]  Sébastien Gros,et al.  An Analysis of the Directional-Modifier Adaptation Algorithm Based on Optimal Experimental Design , 2016 .

[36]  Pu Li,et al.  Set-Point Optimization for Closed-Loop Control Systems under Uncertainty , 2007 .

[37]  E. Pistikopoulos,et al.  A multiparametric programming approach for mixed-integer quadratic engineering problems , 2002 .

[38]  Efstratios N. Pistikopoulos,et al.  Perspectives in Multiparametric Programming and Explicit Model Predictive Control , 2009 .

[39]  E. Pistikopoulos,et al.  Multiparametric Programming in Process Systems Engineering: Recent Developments and Path Forward , 2021, Frontiers in Chemical Engineering.

[40]  Benoit Forget,et al.  Kriging-based algorithm for nuclear reactor neutronic design optimization , 2012 .

[41]  Efstratios N. Pistikopoulos,et al.  Multi-Parametric Programming: Volume 1: Theory, Algorithms, and Applications , 2007 .

[42]  B. Roffel,et al.  Advanced Practical Process Control , 2008 .

[43]  Iftekhar A. Karimi,et al.  Design of computer experiments: A review , 2017, Comput. Chem. Eng..

[44]  Mohamed Nabil Fathy Ibrahim,et al.  Impact of Sampling Technique on the Performance of Surrogate Models Generated with Artificial Neural Network (ANN): A Case Study for a Natural Gas Stabilization Unit , 2019, Energies.

[45]  E. Pistikopoulos,et al.  Algorithms for the Solution of Multiparametric Mixed-Integer Nonlinear Optimization Problems , 1999 .

[46]  Antonio Espuña,et al.  Optimal management of bio-based energy supply chains under parametric uncertainty through a data-driven decision-support framework , 2020, Comput. Ind. Eng..

[47]  Muhammad Nadeem Rafique Chaudhary,et al.  Real time optimization of chemical processes , 2009 .

[48]  Efstratios N. Pistikopoulos,et al.  Control of a dual mode separation process via multi-parametric Model Predictive Control , 2019, IFAC-PapersOnLine.

[49]  Kai Dadhe,et al.  Real‐Time Optimization in the Chemical Processing Industry , 2017 .

[50]  Michael Bortz,et al.  Simulation and Multi‐criteria Optimization under Uncertain Model Parameters of a Cumene Process , 2017 .

[51]  Antonio Espuña,et al.  Multiparametric metamodels for model predictive control of chemical processes , 2016 .

[52]  Elaine T. Hale,et al.  Multi-Parametric Nonlinear Programming and the Evaluation of Implicit Optimization Model Adequacy , 2004, IFAC Proceedings Volumes.

[53]  Georgios M. Kopanos,et al.  Mixed-integer multiparametric Metamodeling: A machine learning tool applied to reactive scheduling , 2018 .

[54]  A. Shokry,et al.  Mixed-Integer MultiParametric Approach based on Machine Learning Techniques , 2017 .

[55]  Antonio Espuña,et al.  Applying Metamodels and Sequential Sampling for Constrained Optimization of Process Operations , 2014, ICAISC.

[56]  Hirokazu Anai,et al.  A symbolic-numeric approach to multi-parametric programming for control design , 2009, 2009 ICCAS-SICE.

[57]  Efstratios N. Pistikopoulos,et al.  Process design and control optimization: A simultaneous approach by multi‐parametric programming , 2017 .

[58]  Andy J. Keane,et al.  Engineering Design via Surrogate Modelling - A Practical Guide , 2008 .

[59]  Efstratios N. Pistikopoulos,et al.  Integrating Deep Learning and Explicit MPC for Advanced Process Control , 2020, 2020 American Control Conference (ACC).

[60]  Tao Chen,et al.  Response surface methodology with prediction uncertainty: A multi-objective optimisation approach , 2012 .

[61]  Efstratios N. Pistikopoulos,et al.  Uncertainty in process design and operations , 1995 .