SVM with linear kernel function based nonparametric model identification and model algorithmic control

In this work, a support vector machine (SVM) with linear kernel function based nonparametric model identification and its application in model algorithmic control (SVM_MAC) technique is presented. An impulse response model involving manipulated variables is obtained via system identification by SVM with linear kernel function according to random test data or manufacturing data, not via special impulse response test. And an explicit control law of a moving horizon quadratic objective is obtained through the predictive control mechanism. Also the characteristic of internal model control (IMC) of SVM_MAC is studied. The approach of SVM based nonparametric model identification and SVM_MAC is illustrated by a simulation of a system with dead time delay. The results show that SVM-MAC technique has good performance in keeping reference trajectory and disturbance-rejection.