LSSVM Predictive Control Based on Improved Free Search Algorithm for Nonlinear Systems

In order to improve control performance of nonlinear systems, a predictive control method based on improved free search algorithm and least square support vector machine was proposed. This predictive control method utilized least square support vector machine to estimate the nonlinear system model and forecast the output value. The output error is reduced through output feedback and error correction. The rolling optimization of control values are obtained through an improved free search algorithm. This predictive control method can be used to design effective controllers for nonlinear systems with unknown mathematical models. Through the simulation experiment of single variable and multivariable nonlinear systems, the simulation results shown that the predictive control method has an excellent adaptive ability and robustness.