Model predictive direct power control using optimal section selection for PWM rectifier with reduced calculation burden

Abstract

The conventional model predictive direct power control (CMPDPC) selects two adjacent vectors and one zero vector during one control period. The cost function with minimum power ripple is to calculate the duration. However, it could occur the negative duration problems, mainly at section switching. This unexpected negative duration may result in power notch and current harmonic increase. Although, it could be improved by reselecting vector when negative duration occurs. But, the power variation rate and duration for reselected vector must be calculated again. In order to reduce the computational burden, an optimal section selection model predictive direct power control (OMPDPC) is proposed in this paper. The relationship of power variation rate between adjacent voltage vectors is developed. Then, the new vectors and duration based the table is introduced to replace with the original calculated results. Meanwhile, low switching losses vector sequence and one step delay compensation are investigated. The computational burden of this new method is reduced and it is quite easy to implement. The power notch could be greatly alleviated. The effectiveness of the proposed method is demonstrated by comparing with other two methods under various conditions. Simulation and experiment results are given to validate the proposed OMPDPC method.