Research On Model Predictive Control Of Electric Vehicle Charging And Discharging Based On Three-Phase Converter

In order to alleviate the energy crisis and environmental pollution,transportation electrification with electric vehicles(EV)as the core is developing rapidly,and emerging vehicle-to-grid(V2G)and grid-to-vehicle(G2V)relationships for EVs have emerged.An effective charging and discharging control strategy for EVs is critical to ensuring the normal operation of EVs.Based on the research background of EV charging and discharging,this thesis studies the model predictive control(MPC)strategy for EV charging and discharging.By investigating the charging and discharging standards,topology,and problems with MPC for EVs,the three-phase bridge fully controlled rectifier is taken as the main research object in order to achieve the research goals of smooth switching between charging and discharging modes of EVs,reduced MPC algorithm computational burden,and enhanced robustness of MPC to parameters.In order to achieve smooth switching between the charging and discharging modes of EVs,the reasons why traditional strategies cannot achieve smooth switching are analyzed.Then the traditional switching strategy is improved by using a shared integrator for the charging and discharging outer control loops,which can avoid the problem of large differences between the feedback current and the reference current at the instant of charging mode(constant current/voltage charging)switching.In addition,the control runaway problem that may be caused in switching from the discharging to charging mode can be avoided by resetting the integrator.The feasibility of the improved handover strategy to achieve a smooth handover is verified through simulation.In order to enhance the robustness of MPC to parameter changes,the extended state observer(ESO)is used to study a disturbance-rejection MPC.The predictive current model based on ESO is derived.The proposed method does not need accurate system parameters,and it only requires the input and output data of the system to obtain the predicted current.The simulation results verify that disturbance-rejection MPC based on ESO has strong robustness to model parameter changes and mismatches.In order to address the issues of low control accuracy in single-vector MPC and high computational burden in multi-vector MPC,an adjacent vector-based MPC(AVB-MPC)is adopted.A threshold is used in AVB-MPC to control the application frequency of multivector MPC to improve control accuracy.At the same time,the voltage vector application table based on the principle of using adjacent vectors reduces the number of MPC iterations and computational burden.The simulation results verify that AVB-MPC has a similar computational burden with single-vector MPC while the control accuracy is improved.An AVB-MPC based on ESO(ESO-AVB-MPC)control strategy is proposed by integrating ESO into AVB-MPC to enhance robustness.The simulation results verify the feasibility and effectiveness of the ESO-AVB-MPC strategy.Finally,an experimental platform for charging and discharging control of EV batteries is built using a battery simulator and an AC/DC converter to further validate the simulation results.The experimental results are consistent with the simulation results,further verifying the feasibility and effectiveness of the proposed control strategy.