Research On Intelligent Control Method Of Interlaced Parallel Boost Converter

Traditional fuel vehicles consume a lot of energy,and exhaust emissions will cause serious pollution to the ecological environment.New energy vehicles using fuel cells as the power source have the characteristics of pollution-free and high efficiency.However,when fuel cells are used in new energy vehicles,there are problems such as soft volt-ampere characteristics and low single-cell battery voltage,so they need to be matched by Boost converters.Bus voltage level.In addition,the service life of the fuel cell will be affected by the size of the ripple,and the ripple can be reduced through the interleaved parallel technology.This paper takes the four-phase interleaved parallel Boost converter as the object for research.In the main circuit topology,the advantages of the interleaved parallel structure are explained,and the working waveform of the converter working in CCM mode is given,and the working mode of the converter is analyzed with the duty cycle of 0.5 D 0.75 as an example.The function relationship between current ripple and duty cycle is deduced and summarized.Theoretical analysis shows that the use of the interleaved parallel structure can effectively reduce the ripple.In order to facilitate the subsequent research and analysis,the main circuit parameters of the converter are then designed.The Boost converter is unstable in topological structure.The commonly used control method is PID control.However,in practical applications,PID control is difficult to achieve the ideal control effect for complex control systems,and the immutability of parameters often makes its performance poor.In this paper,RBF neural network PID control algorithm is used to design the controller for the converter to improve the performance of the converter.The state space averaging method is used for modeling,and a suitable compensation network is designed according to the frequency response curve.Using Matlab/Simulink simulation platform to build a simulation model for comparative analysis,the simulation results show that when RBF neural network PID control is used,the steady-state and dynamic characteristics of the converter are better than PID control.Finally,a prototype of a four-phase interleaved parallel Boost converter is built for experimental verification.The experimental results show that when the RBF neural network PI control is used,the converter can realize the current sharing of the four inductors under the closed-loop control,and the input current ripple is less than 5%,which can effectively improve the service life of fuel cells and is suitable for the field of fuel cell vehicles.