Least Squares Support Vector Machines Combined Inverse Control of Two-motor Variable Frequency Speed-regulating System

Abstract

A Least Squares Support Vector Machines (LSSVM) combined inverse control strategy is proposed for two-motor variable frequency speed-regulating system. First, a right inverse control method is adopted to decouple speed and tension of the two-motor system. Then, a left inverse soft-sensing model is applied to estimate the motor rotor flux, which is substituted into the right inverse system in order to form a whole combined inverse controller. To overcome the difficulties in implementing the combined inverse controller by an analytical method, a LSSVM is used to approximate the static nonlinear mapping. Designed two additional controllers, the LSSVM combined inverse control strategy is achieved. Finally, the effectiveness and capability of this proposed control strategy is verified by computer simulations.