Optimization Of Weighting Factors For Finite Set Model Predictive Control Of Three-phase Inverters

Power electronic converters are widely used in new energy power generation,electric vehicles,household electronic products,electric drives,aerospace and other fields.With the rapid development of power electronic technology and computer control technology,higher requirements have been placed on the efficiency,stability and scalability of power electronic converters.In order to meet these demands,in addition to continuously optimizing the topology of the converter,the use of advanced control technology has become one of the research focuses in the field of power electronic converters.In recent years,finite set model predictive control(FCS-MPC)has been widely used in the control of power electronic converters due to its advantages of simple implementation,fast response speed,and the ability to realize cooperative control of multiple variables.However,there are still some unsolved problems in finite set model predictive control.One of the problems is how to determine the weighting factor in the cost function to realize the cooperative control among various control variables.To solve this problem,this paper takes the three-phase voltage source inverter(VSI)as the research object,and studies the optimization design problem of the weighting factor in the predictive control cost function of the finite set model,The main research work and contributions are as follows:(1)The influence of the weighting factor in the cost function of the finite set model predictive control of the three-phase voltage source off-grid inverter on the system performance indexes is studied.Firstly,the basic principle of FCS-MPC algorithm is introduced,the circuit structure and working principle of three-phase VSI are analyzed,and the mathematical model for voltage prediction of three-phase VSI is established.Then,the selection of the cost function in the power electronic system and the design requirements of the weighting factor are introduced to guide the rational design of the cost function.Finally,in order to verify the influence of the design of the weighting factor in the cost function on the system performance indexes,a simulation model using FCS-MPC to regulate three-phase VSI is built on the MATLAB/Simulink platform,and the establishment and simulation results of the cost function under different control performance requirements are discussed.The results show that the reasonable selection of the weighting factor plays an important role in improving the performance indexes of the system.(2)Aiming at the design problem of the weighting factor in the finite set model predictive control of power electronic converters,a new method based on support vector regression(SVR)algorithm and multi-objective particle swarm optimization(MOPSO)algorithm is proposed to realize the automatic selection of weighting factors.Firstly,the working principle and implementation process of SVR algorithm and MOPSO algorithm are introduced;secondly,the whole design process of data acquisition,SVR model training and MOPSO optimization of SVR model is introduced in detail.Finally,based on the programming environment of Python 3.7,the functional relationship between the weighting factor and the performance indexes is established,which solves the problem that the traditional method is difficult to establish the functional relationship between the weighting factor and the performance indexes.The Pareto optimal weighting factor solution set under the optimal performance indexes is obtained through MOPSO algorithm optimization,which realizes the automatic selection of weighting factors.And a set of general design criteria is established for the optimal design of weighting factors in finite set model predictive control of arbitrary converter topology.(3)An experimental platform of three-phase voltage source off-grid inverter based on finite set model predictive control is built to verify the effectiveness of the method proposed in this paper.The design process of the hardware circuit and algorithm program in the experimental platform is introduced,and the simulation and experimental results under different weighting coefficients are discussed.The results show that the method confirms a good match between the Pareto optimal weighting factor and the performance of the predictive controller when applied to the simulated model(error is less than 2 %)and the experimental model(error is less than 10 %)with different loads.The dynamic performance test of the three-phase voltage source off-grid inverter is carried out,and the output voltage and current of the system smoothly transition to the steady-state value without overshoot,which confirms that the optimization results of the proposed method have negligible influence on the system transient performance.There are 45 figures,8 tables,and 80 references...