Research On Compensation Control Method Of Boost PFC Based On Data-Driven

Interleaved parallel Boost converter is a typical power topology for high-power active power factor correction,which plays an important role in improving power grid harmonic pollution and increasing energy utilization.This paper discusses the influence mechanism of the soft saturation characteristics of magnetic powder core inductors on the stability margin of the converter,and studies the data-driven control method for the converter to improve the stability margin of the system and improve the dynamic and static response ability of the system on the basis of simplifying the controller loop pressure.This paper selects a two-stage interleaved Boost power factor correction(PFC)converter with magnetic powder core inductors as the research object.Firstly,the mathematical model of the system is established,and the relationship between various parameters when adopting double-loop control method is analyzed,and the boundary conditions of loop parameter design are obtained.Secondly,the soft saturation characteristics of magnetic powder core inductors are discussed,and the nonlinear change curve of inductance value decreasing with increasing current is fitted,and the influence of the soft magnetic characteristics of magnetic powder core inductors on the stability margin of the system is analyzed.To solve this problem,proportional resonance(PR)controller is used to compensate for the gain of current control loop,and the problem of generating new anti-interference and stability is studied.In order to reduce the control loop operation time and improve the power quality on the grid side without affecting the system stability,a data-driven compensation control scheme is introduced.The compensation control depends on the online control training compensation data,regression compensation model using offline data,and the use of the compensation model to optimize the power quality.First of all,the working principle and design steps of data-driven control are analyzed,and the data preprocessing method and the fitting process of nonlinear regression model are discussed.Secondly,the improvement effect of data-driven control method based on regression model on system performance is analyzed,and its adaptability is studied.In addition,the effect of proportion control,PR control and data-driven control as current loop control schemes of the converter on system performance is analyzed and evaluated respectively,demonstrating the feasibility and efficiency of the data-driven control scheme.Finally,a simulation model was built in MATLAB/Simulink,and an experimental platform was built with TMS320F28335 as the controller.The correctness of theoretical analysis and the validity of the studied control method are verified by simulation and experiment.The results indicate that using data-driven control method in Boost PFC converter can effectively improve the quality of grid side current waveform,reduce current harmonic distortion rate,and increase system power factor.