Support vector machine based nonlinear model multi-step-ahead optimizing predictive control

Abstract

A support vector machine with guadratic polynomial kernel function based nonlinear model multi-step-a-head optimizing predictive controller was presented. A support vector machine based predictive model was established by black-box identification. And a quadratic objective function with receding horizon was selected to obtain the controller output. By solving a nonlinear optimization problem with equality constraint of model output and boundary constraint of controller output using Nelder-Mead simplex direct search method, a sub-optimal control law was achieved in feature space. The effect of the controller was demonstrated on a recognized benchmark problem and a continuous-stirred tank reactor. The simulation results show that the multi-step-ahead predictive controller can be well applied to nonlinear system, with better performance in following reference trajectory and disturbance-rejection.

Keywords

nonlinear model predictive control / support vector machine / nonlinear system identification / kernel function / nonlinear optimization

References

[1] ShuDi-qianPredictive Control System and its Application [M], 1996, Beijing, China Machine Press(in Chinese)

[2] MorariM, LeeJ H. Model predictive control: past, present and future[J]. Computer and Chemical Engineering, 1999, 23(4–5): 667-682

[3] LiuB, ShenQ, SuH Y, et al.. A nonlinear predictive control algorithm based on fuzzy online modeling and discrete optimization systems[J]. Proc IEEE of Man and Cybernetics, 2003, 1: 816-821

[4] KambhampatiC, MasonJ D, WarwickK. A stable one-step-ahead predictive control of non-linear systems [J]. Automatica, 2000, 36: 485-495

[5] ZhuJingIntelligent Predictive Control and Application [M], 2002, Hangzhou, Zhejiang University Press(in Chinese)

[6] VapnikV NThe Nature of Statistical Learning Theory[M], 1995, New York, Springer-Verlag

[7] CortesCPrediction of Generalization Ability in Learning Machines[D], 1995, Rochester, Univ New York, USA

[8] SmolaA J, SchölkopfBA tutorial on support vector regression [R], 1998, London, Royal Holloway College, London Univ

[9] Müller K R, Smola A J. Predicting time series with support vector machines [A]. Proc of ICANN[C]. Springer LNCS 1327, 1997.

[10] Miao Q, Wang S F. Nonlinear model predictive control based on support vector regression[A]. Proceedings of the 1st International Conference on Machine Learning and Cybernetics[C]. Beijing, 2002.

[11] SuykensJ A K. Nonlinear modelling and support vector machines[J]. IEEE Transactions on Instrumentation and Measurement Technology, 2001, 1: 287-294

[12] SuykensJ A K, VandewalleJ, de MoorB. Optimal control by least squares support vector machines[J]. Neural Networks, 2001, 14(1): 23-35

[13] NarendraK S, ParthasarathyK. Identification and control of dynamic systems using neural networks [J]. IEEE Transaction on Neural Networks, 2000, 1: 4-27

[14] LehmanB, BentsmanJ, LunelS V, et al.. Vibrational control of nonlinear time lag systems with bounded delay: averaging theory, stabilizability, and transient behavior[J]. IEEE Transactions on Automatic Control, 1994, 39: 898-912

[15] CaoY Y, FrankP M. Analysis and synthesis of nonlinear time-delay systems via fuzzy control approach [J]. IEEE Transactions on Fuzzy Systems, 2000, 8(2): 200-211