REAL-TIME PROCESS OPTIMIZATION BASED ON GREY-BOX NEURAL MODELS

This paper investigates the feasibility of using grey-box neural models (GNM) in Real Time Optimization (RTO). These models are based on a suitable combination of fundamental conservation laws and neural networks, being used in at least two different ways: to complement available phenomenological knowledge with empirical information, or to reduce dimensionality of complex rigorous physical models. We have observed that the benefits of using these simple adaptable models are counteracted by some difficulties associated with the solution of the optimization problem. Nonlinear Programming (NLP) algorithms failed in finding the global optimum due to the fact that neural networks can introduce multimodal objective functions. One alternative considered to solve this problem was the use of some kind of evolutionary algorithms, like Genetic Algorithms (GA). Although these algorithms produced better results in terms of finding the appropriate region, they took long periods of time to reach the global optimum. It was found that a combination of genetic and nonlinear programming algorithms can be use to fast obtain the optimum solution. The proposed approach was applied to the Williams-Otto reactor, considering three different GNM models of increasing complexity. Results demonstrated that the use of GNM models and mixed GA/NLP optimization algorithms is a promissory approach for solving dynamic RTO problems.

[1]  Thomas E. Marlin,et al.  The effect of model fidelity on real-time optimization performance , 2004, Comput. Chem. Eng..

[2]  Raghunathan Rengaswamy,et al.  A framework for integrating diagnostic knowledge with nonlinear optimization for data reconciliation and parameter estimation in dynamic systems , 2001 .

[3]  L. Biegler,et al.  Advances in simultaneous strategies for dynamic process optimization , 2002 .

[4]  John D. Perkins,et al.  Economic analysis of different structures of on-line process optimization systems , 1998 .

[5]  Rolf Isermann,et al.  Adaptive on-line steady-state optimization of slow dynamic processes , 1978, Autom..

[6]  Jose A. Romagnoli,et al.  Data Processing and Reconciliation for Chemical Process Operations , 1999 .

[7]  A. J. Morris,et al.  Soft-sensing: a solution to the problem of measurement delays , 1989 .

[8]  Arthur Jutan,et al.  Grey-box modelling and control of chemical processes , 2002 .

[9]  D. C. White,et al.  Online optimization : What have we learned? , 1998 .

[10]  Mark A. Kramer,et al.  Modeling chemical processes using prior knowledge and neural networks , 1994 .

[11]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part I: Quantitative model-based methods , 2003, Comput. Chem. Eng..

[12]  George Stephanopoulos,et al.  Studies in the synthesis of control structures for chemical processes: Part IV. Design of steady‐state optimizing control structures for chemical process units , 1980 .

[13]  Goldberg,et al.  Genetic algorithms , 1993, Robust Control Systems with Genetic Algorithms.

[14]  John D. Perkins,et al.  Structural design for on‐line process optimization: I. Dynamic economics of MPC , 1999 .

[15]  Babu Joseph,et al.  On-Line Optimization of Chemical Processes , 1982, 1982 American Control Conference.

[16]  S. Joe Qin,et al.  A survey of industrial model predictive control technology , 2003 .

[17]  Thomas E. Marlin,et al.  Design cost: a systematic approach to technology selection for model-based real-time optimization systems , 1996 .

[18]  Raghunathan Rengaswamy,et al.  A review of process fault detection and diagnosis: Part III: Process history based methods , 2003, Comput. Chem. Eng..

[19]  Lyle H. Ungar,et al.  A hybrid neural network‐first principles approach to process modeling , 1992 .

[20]  Enrique Luis Lima,et al.  Adaptive hybrid neural models for process control , 1998 .

[21]  D. Himmelblau,et al.  Optimization of Chemical Processes , 1987 .

[22]  Thomas E. Marlin,et al.  Model adequacy requirements for optimizing plant operations , 1994 .

[23]  Luis Puigjaner,et al.  On-line process optimization: parameter tuning for the real time evolution (RTE) approach , 2004, Comput. Chem. Eng..

[24]  Karel Ch. A. M. Luyben,et al.  Strategy for dynamic process modeling based on neural networks in macroscopic balances , 1996 .

[25]  Sheng Chen,et al.  Parallel recursive prediction error algorithm for training layered neural networks , 1990 .

[26]  Thomas E. Marlin,et al.  Multiple data sets for model updating in real-time operations optimization , 2002 .

[27]  Miguel J. Bagajewicz.  Process Plant Instrumentation: Design and Upgrade , 2000 .

[28]  Darci Odloak,et al.  Industrial implementation of a real-time optimization strategy for maximizing production of LPG in a FCC unit , 2000 .

[29]  Vern W. Weekman,et al.  Advanced control practice in the chemical process industry: A view from industry , 1976 .

[30]  David E. Goldberg,et al.  Genetic Algorithms in Search Optimization and Machine Learning , 1988 .

[31]  Tsutomu Mita,et al.  Parametrization of state feedback H┣D2∞┫D2 controllers. , 1990 .

[32]  Jean-Marie Flaus,et al.  Moving horizon state estimation with global convergence using interval techniques: application to biotechnological processes , 2003 .

[33]  Dale E. Seborg,et al.  Adaptive control of photolithography , 1992 .

[34]  Ravi Nath,et al.  On-line dynamic optimization of olefins plants , 2000 .