Abstract:

Hysteresis model predictive control (HMPC) which is based on the finite control set model predictive control (FCS-MPC) has the advantages of fast dynamic response and multi-objective optimization in the three-level inverter. But its switching operation isn’t regular and the switching frequency fluctuates over a large range, resulting in that the frequency spectrum of the output current is scattered and it is difficult to design the filter. In order to handle these problems, a adaptive hysteresis band hysteresis model predictive control (AHB-HMPC) method is proposed in the paper, which selects the fluctuation range of the switching frequency as a control target. By introducing the idea of hysteresis control and adjusting the current’s hysteresis size online, the average switching frequency of the system is within the center of hysteresis band, and the frequency of the output current relatively concentrates around the average switching frequency which is convenient for filter design and remains the HMPC’s advantages known as fast dynamic response and multi-objective optimization. Finally, the simulation and experimental results show that the control method is feasible and effective.

Key words: Three-level inverter, hysteresis model predictive control, switching frequency