Research On Three-level Inverter Based On Model Predictive Control

Three-level active neutral point clamped inverter(ANPC)has been paid more and more attention and applied because of its low output harmonic,low device loss and high power density,which reduces cost and improves efficiency.In recent years,with the development of technology,the research of integrated optimal control methods for the key parameters of ANPC multilevel converter,such as neutral point voltage control,loss distribution balance,common-mode voltage suppression and device loss reduction,has gradually become the focus of scholars’attention.The most representative one is model predictive control method.The main work of this paper is divided into the following points:1.First of all,the research background and significance of multi-level are expounded.The working principle of several common three-level inverters are discussed,and the model predictive control algorithm and its research status in multi-level are described in this paper.2.In order to verify the validity and feasibility of the related control algorithm,the experimental platform of 3L-ANPC inverter is built,and the hardware circuit and software algorithm of the platform are introduced.3.In order to solve the problem of neutral-point voltage offset of three-level ANPC inverters,a finite switching sequence model predictive control strategy is proposed.The strategy not only keeps the neutral-point voltage balanced,but also solves the problem of uncertain switching frequency predicted by traditional models.In this strategy,the control set is defined as 10 sets of switching sequence,and the cost function including neutral point voltage deviation is constructed.The optimal switching sequence is determined by rolling optimization,so as to control neutral point voltage and reduce switching losses.Finally,the effectiveness of the control strategy is verified by simulation and experiment.4.In the three-level ANPC inverters,the influence of neutral point voltage caused by neutral vector can not be eliminated by redundant vector.In this paper,a modulation model predictive control strategy for three-level ANPC inverters based on virtual space vector is proposed.The multi-objective optimization of three-level ANPC inverters is achieved.By switching the switching vector sequence to control the neutral point voltage balance,the strategy can effectively reduce the loss of switching devices by reducing the number of total switching or the number of switching actions of the maximum current phase while achieving the neutral point voltage balance.The simulation and experimental results verify the effectiveness and correctness of the proposed strategy.