The Research On Improved Modulated Model Predictive Control For Matrix Converter

Matrix converter is an AC-AC converter,which has high power density,can realize bidirectional energy flow,can generate sinusoidal output current with adjustable frequency and amplitude,and can achieve unit power factor on the input side.The transformation field has broad application prospects.Scholars from various countries have successively proposed control strategies for matrix converters.Among them,the modulation model predictive control has attracted wide attention,which has the characteristics of fast dynamic response and high quality of output current waveform.However,the concept of this method is complex,the calculation amount is large,and when the system parameters change,the system control effect is significantly reduced.Therefore,this paper takes the matrix converter as the research object to simplify the traditional modulation model predictive control.At the same time,the corresponding control strategy is proposed when the system is in the over-modulation state and the system load parameters change.The main research contents are as follows:A simplified modulation model predictive control is proposed.This article first explains the basic principles of traditional modulation model predictive control,including the generation of basic voltage vectors,the selection of vector combinations,and the calculation of duty cycle.Aiming at the problems of complex control and large amount of calculation in traditional methods,this paper proposes a simplified modulation model predictive control.This simplified method screens the basic voltage vectors according to the sectors where the two vectors of the input voltage vector and the desired voltage vector are located,and reduces the number of combinations of the basic voltage vector and the vector,thereby reducing the amount of calculation of the system.Finally,the correctness of the simplified method is verified in the RT-LAB semi-physical simulation platform,and the effectiveness of the simplified method is verified by comparing the resource occupancy of the traditional method and the simplified method in the control board.A space vector overmodulation strategy is proposed.This article first theoretically analyzes the modulation area boundary of the simplified modulation model predictive control,and conducts simulation verification in MATLAB/Simulink,and finally uses the regular hexagon enclosed by the basic voltage vector as the modulation area boundary.This paper proposes a space vector overmodulation strategy for the overmodulation state of the system.The core idea of this strategy is to take the vector whose vertex falls on a regular hexagon and is closest to the original control target as the new control target.Simulations and experiments show that when the system is in the over-modulation state,the space vector modulation strategy is added,the output waveform of the system is significantly improved,and the control accuracy of the system is significantly improved.A load parameter identification strategy is proposed.This article first analyzes the robustness of the simplified modulation model predictive control load parameters.Because the load resistance in the circuit is different from the resistance parameter of the load model in the controller,there is an obvious error between the output current and the reference current.This paper proposes a load parameter identification strategy,which derives the true value of the load resistance in the circuit through the mathematical relationship between the output voltage and the output current in the load,and corrects the resistance parameters of the load model in the controller in time to ensure the control performance of the system.Simulation and experimental results show that when the load resistance in the circuit changes,the load parameter identification strategy is added,the amplitude of the output current does not change suddenly,and the reference current can be effectively tracked,and the control accuracy of the system is guaranteed.