Study On Energy Management For Plug- In Hybrid Electric Vehicles Based On DA-ECMS

At present,plug-in hybrid electric vehicle(PHEV)is of great significance to alleviate the energy crisis and environmental pressure,and is conducive to the realization of the transition from the traditional vehicle with internal combustion engine to the pure electric vehicle.The fuel saving and emission reduction potential of plug-in hybrid electric vehicle mainly depends on its structural advantages and energy management strategy.The energy management strategy as the core technology of hybrid electric vehicle has been a research hotspot.In the past ten years,the energy management strategy has been widely and deeply researched,from the rule-based strategy to the optimization-based strategy,the control effect has made great progress.However,in all optimization-based strategy,even the instantaneous optimization strategy,such as equivalent consumption minimization strategy(ECMS),the calculation burden is still too high for real vehicle applications,which is the biggest obstacle to the optimize algorithm applied to real vehicles.In order to improve the computational efficiency of ECMS,based on the concept of algorithm discretization,a discretization method of ECMS combined with experimental and mathematical methods is proposed.First,based on the ECMS simulation results,a mode switching torque MAP was extracted,then the driving mode was divided into pure electric driving mode and hybrid driving mode,the decoupling of the mode control and torque distribution of the ECMS strategy was realized.Then,for the hybrid driving mode,the approximate analytical equation of ECMS was deduced,and the constraint condition of the optimal torque was obtained,so that the numerical search and calculation of the whole feasible domain can be converted to the numerical calculation method that only needs to calculate 5 discrete points.Finally,the discrete equivalent consumption minimization strategy was proposed.The simulation results showed that compared with the ECMS strategy,the computation time of D-ECMS was reduced by more than 90%,and the fuel consumption deviation was less than 2%,which proved that D-ECMS can greatly improve the computation efficiency and maintain good control accuracy.In order to make full use of battery power,the optimal SOC descent trajectory was analyzed under different travel mileage and battery SOC initial.For known travel mileage,proposed a reference SOC trajectory planning method that decreases linearly with mileage,and according to the SOC deviation adjusted equivalent factor online,to realize the actual SOC following the reference.For unknown travel mileage,according to the idea of mileage utilization factor,we propose the interval mean value based on the maximum probability of historical trip mileage distribution as the reference mileage of SOC planning.In order to select a more suitable initial equivalent factor,secant method is applied to find the optimal equivalent factor of different mileage and SOC initial value,and the equivalent factor reference value MAP was obtained.Furthermore,the influence of equivalent factor reference value on control strategy was analyzed,and the necessity of online updating of equivalent factor reference value was proved,and an updating method of the reference equivalent factor was put forward.Finally,the Discrete adaptive equivalent consumption minimization strategy(DA-ECMS),which can adapt to different working conditions and travel characteristics,was proposed.Finally,the simulation and hardware-in-loop(HIL)experiment of DA-ECMS were implemented.The simulation results show that,compared with the CD-CS(Charge Depleting/Charge Sustaining)strategy,the DA-ECMS strategy can allocate the torque more reasonably and achieve lower fuel consumption.The HIL experiment proved that the real-time performance of the DA-ECMS strategy meets the needs of the embedded system.