Multi-objective Energy Management Control Strategies Considering Equivalent Fuel Consumption And Battery Life For Plug-in Hybrid Vehicles

At present,the increasingly serious energy dilemma and environmental pollution problems in the society are prompting the rapid development of the new energy automobile industry to alleviate a series of social problems caused by traditional fuel vehicles.Among the existing new energy vehicles,plug-in Hybrid Electric Vehicle(PHEV)is the best alternative to conventional fuel vehicles.PHEV equips a high-cost,high-capacity power battery that can be recharged from an external power grid.As a result,the PHEV can work more in pure electric mode on real roads to reduce fuel consumption and exhaust emissions.However,frequent charging and discharging will cause irreversible loss to the battery,which will shorten the service life of the battery.Replacing the power battery before the end of the service life of PHEV will greatly increase the service cost of PHEV.Therefore,it is important to design energy management strategies that take battery life into account to reduce the use cost of PHEV.In this paper,two kinds of multi-objective energy management strategies considering battery life are studied by taking plug-in commuter hybrid electric vehicles as the research object.The main research contents are as follows:A Recurrent Neural Network-Based Adaptive Equivalent Consumption Minimization Strategy(RNN-A-ECMS)considering battery life is designed.The strategy design includes two parts: offline design and online design.The offline design also includes two parts.The first part is to train an RNN offline with the SOC optimal reference track obtained by Dynamic Programming(DP)and historical traffic data.The other part is to use Particle Swarm Optimization(PSO)and Pontryagin's Minmum Principle(PMP)offline establishing MAPs of the initial value of the equivalent factor in ECMS and the weight coefficient used to balance the two objectives of energy consumption and battery life according to different SOC and power demand.Online design part will be offline trained RNN and two MAPs by a PI controller integrated for online RNN-A-ECMS energy management control system.Equivalent factor and weight coefficient can be adjusted according to the current real-time traffic information in order to improve the control performance of the designed energy management strategy and adaptability to the driving conditions outside the training set.An adaptive equivalent fuel consumption minimum(A-ECMS)energy management strategy considering battery life was designed by using the differential evolution(DE)algorithm to directly solve the multi-objective energy management optimization control problem.In order to reduce the online computation burden by computing the outside of the pareto solution set.First,historical traffic data are divided into different path section according to the road traffic characteristics.Then,the multi-objective energy management optimization problem is solved by DE in every path section.And the pareto solution MAPs of control variable are obtained according SOC,power demand and section number of road.Finally,the optimization range of A-ECMS control variable is determined by looking up the pareto solution set MAPs.The A-ECMS designed in this way not only increases the selection range of the optimal solution to improve the optimization effect,but also reduces the complexity of the online calculation of the energy management strategy to improve the real-time performance.In order to fully demonstrate the effectiveness and the adaptability of different working conditions of the proposed energy management control strategy,this paper uses a cosimulation platform MATLAB/Simulink and GT-SUITE to simulate,and compared with other energy management strategy.