Multistep Model Predictive Control of Induction Motors for Reducing Switching Frequency

Abstract

A multistep model predictive current control strategy for reducing the switching frequency of inverters is proposed to address the challenges of low energy efficiency of small power induction motors. Based on the discrete mathematical model of the motor drive system, the strategy extends the current trajectory corresponding to each allowable switching sequence employing iterative prediction with quadratic interpolation, and uses average switching frequency within the predicted range as a cost function. In order to acquire the ideal switching vector in real-time, the cost function conducts online rolling optimization. Compared to model predictive direct current control, this method can reduce the average switching frequency of the two-level inverter while enhancing current harmonic performance. The simulation results verify the efficiency of the proposed method.