Improving Transient Response of Model Reference Neuro-Controller via Constrained Optimization

A robust adaptation algorithm based on error normalization is introduced to update the weights of model reference neural network controller. Tracking error is normalized by a variable normalizing gain specified by solving a constrained optimization problem. The so-called piecewise quadratic cost function is proposed as the performance index to improve the transient response specifications. The conditions for robust convergence, saturation limit of actuators and maximum possible speed of response form the constraints of the problem in terms of the variable normalizing gain. Simulation results provided, demonstrate the improvements in transient behavior of control signal and output response obtained by the method, even in the presence of disturbances and parameter variations.

[1]  S. Tsujii,et al.  Improvement in stability and convergence speed on normalized LMS algorithm , 1995, Proceedings of ISCAS'95 - International Symposium on Circuits and Systems.

[2]  H. Koofigar,et al.  Robust Adaptive Tracking Using Mixed Normalized/Unnormalized Estimation Errors , 2007, 2007 Information, Decision and Control.

[3]  T. C. Chen,et al.  Model reference neural network controller for induction motor speed control , 2002 .

[4]  Petros A. Ioannou,et al.  On the stability proof of adaptive schemes with static normalizing signals and parameter projection , 1993, IEEE Trans. Autom. Control..

[5]  P. Sicard,et al.  Neural network based model reference adaptive control structure for a flexible joint with hard nonlinearities , 2004, 2004 IEEE International Symposium on Industrial Electronics.

[6]  János D. Pintér,et al.  Global optimization in action , 1995 .

[7]  Anuradha M. Annaswamy,et al.  Robust Adaptive Control , 1984, 1984 American Control Conference.

[8]  Maurizio Cirrincione,et al.  An MRAS-based sensorless high-performance induction motor drive with a predictive adaptive model , 2005, IEEE Transactions on Industrial Electronics.

[9]  Hai-Bin Wang,et al.  Model reference neural network control for boiler combustion system , 2005, 2005 International Conference on Machine Learning and Cybernetics.

[10]  S. Sastry,et al.  Adaptive Control: Stability, Convergence and Robustness , 1989 .

[11]  Alessandro Astolfi,et al.  Is normalization necessary for stable model reference adaptive control? , 2005, IEEE Transactions on Automatic Control.

[12]  Kwang Y. Lee,et al.  An optimal tracking neuro-controller for nonlinear dynamic systems , 1996, IEEE Trans. Neural Networks.