Research On Control Strategy Of Single-ended Flyback DC-DC Converter

As a typical switching power supply,DC-DC converters are important power supply devices in testing equipment,home electronics and smart cars.Some intelligent hardware devices have stricter requirements on the power supply quality of power supplies.Therefore,research and design of DC-DC converter control strategy with high precision and good dynamic performance has important theoretical significance and practical application value.Based on the basic technology of DC-DC converter,this paper analyzes and summarizes the circuit characteristics and working principle of the classical topology.It uses the commonly used state space averaging method to establish a mathematical model for the Flyback DC-DC converter and calculate the circuit parameters of the main power loop.Based on the frequency analysis,PI compensation and PID compensation analysis of Flyback converter,a predictive fuzzy PI control strategy is proposed for the problems of low output precision and large overshoot in linear control.The Single-valued model prediction algorithm is introduced to improve the input of the fuzzy PI controller and reduce the influence of phase lag.However,fuzzy logic control is a coarse adjustment method.To further improve the control precision,an output feedback model predictive control method is proposed.On the basis of designing the state feedback model predictive controller,the Luenberger-Type observer is added,and observer gain can ensure that the state estimation dynamic error is global exponential stability,and the outer loop PI control is designed to eliminate the voltage offser error.Finally,the closed-loop stability analysis of the global convergence of the output voltage and the global convergence of the inductor current with the input constraints is given.Based on DSP to build a hardware experimental platform,the whole experimental system includes main power loop,EMI filter circuit,isolated sampling circuit,drive circuit and DSP control circuit.The simulation results show that the steady-state error of the system controlled by predictive fuzzy PI control is about 0.55%,the dynamic adjustment time is 0.02s,and the overshoot is 1.3%.In the presence of input voltage and load disturbance,the output feedback model predictive control strategy can not only guarantee global stability,but also realize the convergence of system error in a limited time.The steady-state error is 0.22%,and there is no overshoot.The experimental results show that the system has high control accuracy,fast dynamic response time and good robustness when the input voltage and load change.