Neuro-fuzzy-based dynamic quadratic criterion-iterative learning control for batch process

Abstract

Considering the potentials of iterative learning control as a framework for industrial batch process control and optimization, a novel dynamic parameters-based quadratic criterion-iterative learning control (Q-ILC) is proposed in this paper. Firstly, Q-ILC with dynamic parameter is used to improve the performance of ILC. As a result, the proposed method can avoid the problem of initialization of the optimization controller parameters, in which a trial and error procedure is usually resorted to in the existing iterative algorithms used for the optimization of the batch process. Next, we make the first attempt to provide a rigorous description and proof to verify that the changes of the ILC policy converges with respect to the batch index number, which are normally validated only on the basis of the simulation results in the previous works. Lastly, an example is used to illustrate the performance and applicability of the proposed method.

Keywords

Batch processes iterative learning control quadratic criterion neuro-fuzzy model

Get full access to this article

View all access options for this article.

References

Boggs

Tolle

(1995) Sequential quadratic programming. Acta Numerica 4: 1–51.

Bonvin

(1998) Optimal operation of batch reactors: a personal view. Journal of Process Control 8: 355–68.

Cybenko

(1989) Approximation by superposition of a sigmoidal function. Mathematics of Control, Signals, and Systems 2: 303–14.

Fletcher

(1987) Practical Methods of Optimization. 2nd ed. New York: Wiley.

Girosi

Poggio

(1990) Networks and the best approximation property. Biological Cybernetics 63: 169–79.

Jia

Jiping

Shi

Min-sen

Chiu

(2009) Research on fuzzy neural model with global convergence for batch process. Information and Control (in Chinese) 38(6): 685–91.

Lee

Kim

(2000) Model-based iterative learning control with a quadratic criterion for time varying linear systems. Automatica 36(5): 641–57.

Nocedal

Wright

(1999) Numerical Optimization, chapters 12, 18. New York: Springer.

Park

Sandberg

(1991) Universal approximation using radial-basis-function networks. Neural Computation 3(2): 246–57.

10.

McAvoy

Werbos

(1992) Long-term prediction of chemical process using recurrent neural networks: a parallel training approach. Industrial & Engineering Chemistry Research 31: 1338–52.

11.

Terwiesch

Argarwal

Rippin

DWT

(1994) Batch unit optimization with imperfect modelling: a survey. Journal of Process Control 4: 238–58.

12.

Tian

Zhang

Morris

(2001) Modeling and optimal control of a batch polymerization reactor using a hybrid stacked recurrent neural network model. Industrial & Engineering Chemistry Research 40: 4525–35.

13.

Tian

Zhang

Morris

(2002) Optimal control of a batch emulsion copolymerisation reactor based on recurrent neural network models. Chemical Engineering and Processing 41: 531–8.

14.

Xiong

Zhang

(2003) Product quality trajectory tracking of batch processes using iterative learning control based on time-varying perturbation model. Industrial & Engineering Chemistry Research 42: 6802–14.

15.

Xiong

Zhang

(2004) Modelling and optimal control of fed-batch processes using a novel control affine feedforward neural network. Neurocomputing 61: 317–37.

16.

Xiong

Zhang

(2005a) A batch-to-batch iterative optimal control strategy based on recurrent neural network models. Journal of Process Control 15: 11–21.

17.

Xiong

Zhang

(2005b) Neural network model-based on-line re-optimisation control of fed-batch processes using a modified iterative dynamic programming algorithm. Chemical Engineering and Processing 44: 477–84.

18.

Xiong

Zhang

Wang

(2005) Tracking control for batch processes by integrating batch-to-batch iterative learning control and within-batch on-line control. Industrial & Engineering Chemistry Research 44: 983–92 (2005).

19.

Zafiriou

Adomaitis

Gattu

(1995) An approach to run-to-run control for rapid thermal processing. Proceedings of the American Control Conference, 1286–8.

20.

Zhang

(2008) Batch-to-batch optimal control of a batch polymerisation process based on stacked neural network models. Chemical Engineering Science 63: 1273–81.