Neural Network Model Predictive Control of Nonlinear Systems Using Genetic Algorithms

In this paper the synthesis of the predictive controller for control of the nonlinear object is considered. It is supposed that the object model is not known. The method is based on a digital recurrent network (DRN) model of the system to be controlled, which is used for predicting the future behavior of the output variables. The cost function which minimizes the difference between the future object outputs and the desired values of the outputs is formulated. The function ga of the Matlab’s Genetic Algorithm Optimization Toolbox is used for obtaining the optimum values of the control signals. Controller synthesis is illustrated for plants often referred to in the literature. Results of simulations show effectiveness of the proposed control system.

[1]  Yukihiro Toyoda,et al.  Exponential ARX model-based long-range predictive control strategy for power plants , 2001 .

[2]  Wen Yu,et al.  Nonlinear system identification using discrete-time recurrent neural networks with stable learning algorithms , 2004, Inf. Sci..

[3]  Ching-Chih Tsai,et al.  Generalized predictive control using recurrent fuzzy neural networks for industrial processes , 2007 .

[4]  Mekki Ksouri,et al.  Multi-criteria optimization in nonlinear predictive control , 2008, Math. Comput. Simul..

[5]  D.R. Hush,et al.  Progress in supervised neural networks , 1993, IEEE Signal Processing Magazine.

[6]  Yuichi Nakamura,et al.  Approximation of dynamical systems by continuous time recurrent neural networks , 1993, Neural Networks.

[7]  M. Gupta,et al.  Approximation of discrete-time state-space trajectories using dynamic recurrent neural networks , 1995, IEEE Trans. Autom. Control..

[8]  Leang-San Shieh,et al.  A new approach for neural control of nonlinear discrete dynamic systems , 2005, Inf. Sci..

[9]  P. S. Sastry,et al.  Memory neuron networks for identification and control of dynamical systems , 1994, IEEE Trans. Neural Networks.

[10]  Vesna Ranković,et al.  Identification of nonlinear models with feed forward neural network and digital recurrent network , 2008 .

[11]  Vesna Ranković,et al.  MODEL PREDICTIVE CONTROL BASED ON THE TAKAGI-SUGENO FUZZY MODEL , 2007 .

[12]  Yi Cao,et al.  Nonlinear system identification for predictive control using continuous time recurrent neural networks and automatic differentiation , 2008 .

[13]  Sung-Kwun Oh,et al.  Hybrid identification in fuzzy-neural networks , 2003, Fuzzy Sets Syst..

[14]  M. Hagan,et al.  TRAINING RECURRENT NETWORKS FOR FILTERING AND CONTROL , 1999 .

[15]  Frank L. Lewis,et al.  Identification of nonlinear dynamical systems using multilayered neural networks , 1996, Autom..

[16]  Panagiotis D. Christofides,et al.  Robust hybrid predictive control of nonlinear systems , 2005, Autom..

[17]  Panagiotis D. Christofides,et al.  Predictive control of transport-reaction processes , 2005, Comput. Chem. Eng..

[18]  Octavian Pastravanu,et al.  A neural predictive controller for non-linear systems , 2002, Math. Comput. Simul..

[19]  Okyay Kaynak,et al.  A comparative study of neural network structures in identification of nonlinear systems , 1999 .

[20]  Françoise Lamnabhi-Lagarrigue,et al.  Nonlinear systems parameters estimation using radial basis function network , 2006 .

[21]  Khashayar Khorasani,et al.  Adaptive time delay neural network structures for nonlinear system identification , 2002, Neurocomputing.