Research On The Multi-objective Parameter Optimization Of Plug-in Hybrid Electric School Bus Powertrain System

The development of Plug-in hybrid electric vehicle(PHEV)is one of approach to solve the problems that oil resources are depleted day by day and city haze problems occur frequently.The dynamic parameter optimization problem of PHEV is one of the hot research topic in recent years.The plug-in hybrid electric school bus(PHESB)has been choosen as the research object in this paper,and develops hybrid optimization algorithm based on Multi-Island genetic algorithm(MIGA)and sequential quadratic programming algorithm(NLPQL),by setting the global optimal energy management strategy based on dynamic programming(DP).The main research achievements are as follows:(1)In the environment of Simulink/MATLAB,this paper has built forward simulation models of mixed type PHESB and parallel type PHESB,and the rule-based control strategy are formulated for models.On the basis of four principles to energy control strategy,3 running modes of PHESB has been established in this paper,including electric vehicle mode(EV),charge-depletion mode(CD)and charge-sustaining mode(CS).In this energy control strategy,SOC values of power battery have been set as threshold method of mode switch,and the setting of switching between CD and CS mode could be reversible.So this paper has developed rule-based control strategies,and completed the two configurations PHESB model building.(2)The global optimal energy management strategy of two configurations of PHESB have been proposed,which is based on dynamic programming algorithm(DP).The DP problems of PHESB have been solved by using MATLAB,and the energy consumption per hundreds kilometer corresponding to the global optimal control strategy has been calculated,through the establishment of two sets of configuration PHESB simulation model of dynamic programming.China Transport Bus driving cycle has been chosen as driving cycle in this paper,and the simulation results between the rule-based control strategy and global optimal control strategy based on DP algorithm were been compared.The simulation results show that global optimal energy management strategy can make fuel economy of two configurations of PHESB increase by more than 10%.The theoretical value of fuel consumption of two configurations PHESB have been calculated,to be prepared for the optimization of power system parameter calibration.(3)A hybrid algorithm of parameter optimization has been proposed,which is based on MIGA and NLPQL.In order to ensure the fairness of each power system combination scheme performance evaluation in the process of parameter optimization,PHESB power system parameters optimization is proposed based on global optimization of energy management strategy.The comparison,between dynamic programming results before parameters optimization and after parameters optimization,shows that the mixed type PHESB's fuel consumption per 100 km after parameters optimization is 17.24L/100 km,and fuel economy has increasing 6.9% comparing to it before parameters optimization;the parallel type PHESB's fuel consumption per 100 km after parameters optimization is 21.39L/100 km,and fuel economy has increasing 12% comparing to it before parameters optimization.(4)Hardware-in-loop(HIL)experiment platform integrated with CAN simulation environment has been set up.MATLAB/Simulink models and control strategy of two configurations PHESB before and after optimization have been downloaded to the real controller,based on the automatic code generation technology,and corresponding HIL experiments have been carried out.Finally,simulation process and data have been observed and record by using CANoe software.HIL experiment results show that after combination parameters optimization algorithm,fuel consumption per 100 km of mixed type PHESB dropped from 25.7 L to 22.9 L,which fell by 10.8% than before optimization;fuel consumption per 100 km of parallel type PHESB dropped from 28.9 L to 26.4 L,which fell by 8.6% than before optimization.The hybrid parameter optimization algorithm optimization results of fuel economy for two configurations of PHESB has been proved.