High Performance Direct Power Predictive Control For Reduced Matrix Converter

Reduced matrix converter(RMC)has attracted attention from many scholars for its green grid-side performance,high power density,bidirectional energy flow,and wide voltage range.There is no DC energy storage link in the topology,and the transformer couples the front and rear stages,resulting in the immunity is poor,and the traditional modulation strategy is unable to solve this weak immunity problem.Model predictive control method has become the current research hotspot due to its simple principle,easy to add constraints,and strong anti-disturbance performance.In this paper,the reduced matrix converter and model predictive control method are combined to improve the system’s anti-disturbance performance.First of all,this paper carries on the modeling analysis to the reduced matrix converter,regards the power as the constraint target of the model predictive control,realizes the reduced matrix converter power predictive control.Secondly,in view of the conventional power predictive control error problem caused by delay and single cost function,a new cost function is constructed by adding the difference between two adjacent beat predictions in the conventional cost function as an optimization item to realize reduced matrix converter error minimization power predictive control.This method can make the prediction result more accurate and the error smaller.Finally,in view of the conventional power predictive control strategy that is unable to cope with the impact of unbalanced grid and harmonic conditions,the cascaded delayed signal cancellation(CDSC)method combined with the extended instantaneous power theory is used to achieve reduced matrix converter CDSC extended power predictive control.This method can effectively filter out harmonics and correctly extract the positive and negative sequence components of the power grid,and eliminate the influence of unbalanced power grid and low-order harmonics on the performance of the reduced matrix converter on the grid side.The control strategy is simulated and verified by experiments.The results show that the reduced matrix converter power predictive control has good input and output performance;the error minimization power predictive control has lower input and output ripple than the conventional power predictive control,and the cost function prediction error is greater small;CDSC extended power predictive control can effectively suppress the input current distortion caused by unbalanced harmonic power grid conditions to ensure grid-side performance.