Model Predictive Control Of Dual Active Bridge DC-DC Converter

With the "Dual carbon" target proposed,renewable energy has been rapidly developed.Due to its high conversion efficiency,bidirectional power flow,high power density and electrical isolation,the Dual Active Bridge(DAB)DC-DC converter is extensively utilized in the fields of renewable energy and electric vehicle.At the same time,with the emergence of various complex application scenarios,higher performance requirements are demanded for DAB DC-DC converter.For the DAB DC-DC converter with the uncertain system parameters,this paper investigates the control performance improvement,backflow power reduction and parameter robustness.Research work includes:Firstly,the working characteristics and principle of DAB DC-DC converter under single phase shift(SPS)and extended phase shift(EPS)modulations are compared.The mathematical models are established,and the expressions of transmission power and backflow power are derived respectively.The analysis shows that EPS modulation improves control flexibility,increases power transmission range,as compared with SPS modulation.Secondly,based on the analysis of SPS modulation,the problems including poor dynamic performance and weak parameter robustness of DAB DC-DC converter are discussed.In this paper,the prediction model of DAB DC-DC converter under SPS modulation is established,and the cost function of prediction error is constructed.The parameter sensitivity analysis of the performance with respect to circuit parameters(equivalent inductance,output capacitance,equivalent load)in DAB DC-DC converter is carried out.To mitigate the impact of inaccurate circuit parameters on performance,the robust model predictive control(RMPC)method is proposed by introducing a robust term into the traditional prediction model.The feasibility of this method is verified by simulation results.Third,the backflow power under SPS and EPS modulations are compared,and the conditions for optimizing the backflow power of DAB DC-DC converter are analyzed in order to eliminate backflow power in DAB DC-DC converter.A prediction model is established using multi-objective cost function including output voltage tracking and backflow power reduction as control objectives.In order to address the influence of uncertain parameters,a robust item is added to the prediction model,the multi-objective robust model predictive control(MRMPC)is proposed.The validity of this method is verified by simulation.Finally,the experimental platform of DAB DC-DC converter is built using TMS320F28335 as the core controller.All above theoretical analysis and control method are verified by experiments.