Research On Model Predictive Control For Bidirectional DC-DC Converter Of Electric Vehicles

In recent years,in order to reduce the emission of automobiles,protect the environment and save resources,electric vehicles have developed rapidly.Hybrid energy can effectively solve the shortcomings of traditional electric vehicles.Therefore,it has begun to attract the attention of many researchers.Bidirectional DC-DC converter,as a key component of electric vehicles,can achieve two-way flow of energy.So,it is one of the research hotspots in this field.This paper,combined with scientific research,has researched model predictive control for bidirectional DC-DC converter of electric vehicle.The main contents are as follows:The design requirements and topology of the bidirectional DC-DC converter were analyzed,and the prediction model was established.The demand analysis of the bidirectional DC-DC converter applied to electric vehicles was carried out,and reasonable design requirements were put forward.The staggered parallel topology of the bidirectional DC-DC converter was analyzed in detail.The working states of Buck mode and Boost mode were analyzed,and the corresponding small-signal model was established to obtain the system transfer function.The system transfer function in both modes were discretized,combined with the predictive control principle,the prediction model under two working modes were established.An unconstrained model predictive controller was designed.According to the system control requirements,that is,the output current error was as small as possible and the process of tracking target value is smooth,a suitable cost function was defined.Under the unconstrained condition,the minimization problem of the cost function was transformed into solving the analytical solution with the control variable as the independent variable,so that the optimal duty cycle of turning tube was obtained.In order to improve the performance of the controller,the appropriate controller parameters were designed by simulation.Finally,the model prediction controller and PI controller were simulated,and the simulation results were compared and analyzed.A constrained model predictive controller based on particle swarm optimization was designed.Based on the model predictive control cost function,a hard constraint was added to the control variable,and the cost function of the constraint model predictive control algorithm was constructed.The particle swarm optimization was introduced to optimize the cost function.The optimization principle and calculation formula of the particle swarm optimization were introduced in detail,and the parameters of particle swarm optimization were designed through simulation,then,constrained model predictive controller was designed completely.The constraint model predictive control algorithm was verified by simulation,and the simulation results were compared with other two controller simulation results.A bidirectional DC-DC converter experimental platform was designed.Based on DSP plus FPGA dual-chip control system,an effective experimental research overall scheme was proposed,and the design of system software and hardware was completed.The hardware mainly included power component selection,the design of energy storage component parameters,sampling circuit and drive circuit.The software mainly included the design of main program and interrupt program for DSP,and the design of communication module and predictive control module for FPGA.Based on the simulation results,the experimental research on model predictive control algorithm and PI control algorithm was carried out.The experimental results showed that: the inverter adopted the model predictive control algorithm based on particle swarm optimization had better dynamic performance and steady-state performance,which was consistent with the simulation results.The feasibility and superiority of the proposed algorithm were verified.