An Improved Model Predictive Control Method for Induction Motor Drives Fed by Indirect Matrix Converter

Abstract

In the model predictive control (MPC) method for the indirect matrix converter (IMC), due to the limited switching state and the fixed control frequency, the output voltage of converter is limited. As a result, if the valid output voltages and the reference value are mismatched, there will be rich harmonics in the output and input currents. In order to solve this problem, an improved model predictive control method is proposed for the induction motor drives fed by an indirect matrix converter, in which the grid current and the motor flux vector are used as the control objectives. A zero vector switching state of the inverter circuit is inserted in each sampling period, and the optimal duty cycle and the corresponding switching states are obtained by using the monotonicity of the cost function with the time constraint of the switching state. The simulation and experimental results show that good grid current quality and dynamic/steady state performances for the induction motor are achieved by using the proposed method, and the zero current commutation for bi-directional switches is realized. Compared with the conventional MPC, the harmonics of the input and output currents for IMC are both reduced, especially under the middle and low speed conditions.