Multi-objective model-predictive control for high-power converters

This paper presents a multi-objective model-predictive control (MOMPC) strategy for controlling converters in high-power applications. The controller uses the system model to predict the system behavior in each sampling interval for each voltage vector, and the most appropriate vector is then chosen according to an optimization criterion. By changing the cost function properly, multiobjectives can be achieved. To eliminate the influences of one step delay in digital implementation, a model-based prediction scheme is introduced. For high-power applications, the converter switching frequency is normally kept low in order to reduce the switching losses; this is done by adding a nonlinear constraint in the cost function. However, to avoid system stability deterioration caused by the low switching frequency, an N-step horizontal prediction is proposed. Finally, the control algorithm is simplified using a graphical algorithm to reduce the computational burden. The proposed MOMPC strategy was verified numerically by using MATLAB/Simulink, and validated experimentally using a laboratory ac/dc converter.