Research On Adaptive Energy Management Control Strategy Of Extended Range Electric Bus

The "electrification,networking,intelligence,and sharing" has become a recognized future development trend in the automotive industry.The automobile will evolve from traditional traveling tools to clean and intelligent mobile terminals.Extended range electric vehicles are the transition from traditional fuel-fueled vehicles to pure electric vehicles.It has the advantages of long driving range,relatively mature manufacturing and electronic control technology,and has been accepted by the market.The extended range electric bus energy management(driving control)strategy will directly affect the overall vehicle performance,and then affect the market promotion of the model.Therefore,research on energy management strategies has an important role in promoting the application of new energy vehicles.This article relies on the school-enterprise cooperation project to study the energy management strategy of an extended range electric bus.The main work contains the establishment of a global optimization model based on the Pontryagin’s minimum principle,target driving cycle generation,and global optimal control sequence research,the formulation of adaptive energy management strategy based on the SOC reference trajectory,the simulation analysis of the entire vehicle and the control strategy,etc.The specific research contents are as follows:(1)Based on the experimental data of components,models of key components such as auxiliary power unit(APU),power battery pack,and drive motor are established.A search algorithm for the optimal working curve of APU is proposed,which reduces the calculation amount of energy management strategy.The correspondence between the APU’s allocated power and the optimal operating point was obtained;(2)Based on the Pontryagin’s minimum principle,combined with component characteristics and system performance requirements,control quantities,state quantities,and system terminal constraints are established,and a global optimization model for extended range electric buses is constructed.(3)The intelligent optimization algorithm requires a large amount of driving cycle data.Inorder to solve the problem of a large number of driving cycles required for training and verification,a Markov Chain target driving cycles generation method is proposed.Use standard driving cycles and collected data to establish urban roads and high-speed probability transition matrices to generate a series of target operating conditions;(4)The architecture and implementation method of adaptive control strategy are proposed.By constructing an adaptive co-state function,the co-state value is divided into two parts: a fixed term and a dynamic term.According to the actual driving cycle and the initial state of the vehicle,the three-dimensional map of the fixed term is interpolated to obtain the initial solution of the co-state value.With the goal of following the SOC reference trajectory,the dynamic term is adjusted in real time.In order to adapt to different driving cycles,driving distance and initial state of SOC;(5)Using the Shooting Method gradient descent method,the initial value of the optimal co-state of the global optimization model under certain driving cycle is solved.A series of target driving cycles are used to solve the global optimization model multiple times under different operating conditions and different SOC initial values to establish a three-dimensional map of the initial terms of SOC-operating conditions-optimal co-state values;(6)In order to obtain the SOC reference trajectory required for the dynamic term,SOC reference trajectories with SOC threshold switching are proposed by analyzing a series of SOC global optimal change trajectories at different distances and different initial SOCs;(7)In order to obtain a co-state initial solution by interpolating a fixed-term 3D plot,real-time driving cycle information needs to be obtained.Establish an offline map for route search,and access the average speed information of each road by accessing the Amap platform,and get the real-time driving cycle of the driving route through statistics;(8)In order to verify the designed adaptive control strategy,a vehicle model,as well as models of PMP,ECMS,and adaptive control strategies were established in Matlab /Simulink.Based on the constructed target driving cycle and standard cycle,a comparison experiment is performed under different driving mileage and initial SOC conditions.Simulation results show that the adaptive control strategy designed in this paper can reduce the energy consumption of the vehicle on the premise that the control quantity,state quantity constraints and system SOC terminal constraints are met.And it has goodadaptability to changes in operating conditions,mileage and SOC initial value.