Generalized predictive control model based on support vector machines

Abstract

Aiming at nonlinear decontrolled plants at large exist in industrial processes, this paper firstly introduces the support vector machine and least squares support vector machine briefly. On this basis, we propose a nonlinear generalized predictive control model based on least squares support vector machines. This method can overcome the classic quadratic programming method for solving support vector machines curse of dimensionality problem, and has a good robustness, suitable for large-scale computing. So use least squares support vector machines as nonlinear predictive model have more advantages.