Research On Regenerative Brake Energy Management Strategy For A Pure Electric Vehicle Based On Driving Style

Optimizing the braking control performance based on driving style is one of the important technologies to reduce vehicle energy consumption.During urban driving situations,nearly the third of the driving energy is consumed in deceleration processes.The braking energy regenerative management system promote the energy efficiency.Several control algorithm of regenerative safety brake of electric passenger vehicles including optimization strategy of braking energy regenerative management based on driving style is investigated to improve braking driving experience and energy economy in this paper.Firstly,the simulation model of a pure electric passenger vehicle is constructed with the forward modeling method in the MATLAB/Simulink,which consists of the electric motor,battery pack,mechanical transmission and driver model.Based on one hundred and seventy thousand kilometers road testing results of sixteen types parameters in China new energy vehicle test cycle,and a mechanism analysis of the vehicle energy consumption influenced by driving style and the author proposes six kinds of energy-saving characteristic parameters to reduce energy consumption,including the average brake pedal depth,the average accelerator pedal depth,the average steering wheel angle,the change rate of pedal depth and steering wheel angle.Considering the vehicle operating data,those key driving style sensitive parameters affecting the energy consumption under the influence of the combination of the driver,the vehicle and the road are analyzed,which can provide the theoretical guidance in designing the model.Secondly,considering the complexity and high precision of the energy consumption level prediction model by driving style,a bidirectional long short term memory method based on deep learning is applied to optimize the characteristics and realize real time and identification of the driving style.Moreover,it can effectively improve the gradient disappearance and has the ability to predict the future vehicle energy consumption level.The simulation results show that the model has good identification ability and can forecast the energy consumption level by driving style.Thirdly,the regenerative brake energy management strategy of the electric vehicle is transformed into a class optimal control problem with some constraints by Berman optimality principle.At the same time,the objective function,variables and constraints of the problem are determined.Additionally,improved iterative dynamic programming algorithm is applied to acquire the optimal control law in China light-duty vehicle test cycle for passenger car,which reduces the complexity of algorithm.Driving condition is closely related to driving style.The driving style energy level prediction model can predict the vehicle energy consumption level at the next moment.Considering the difficulty of the optimal law being adopted in real vehicle,an improved iterative dynamic programming-bidirectional long short term memory(IIDP-BLSTM)algorithm is developed to handle the conflict between solution and variables of driving style.The IIDP-BLSTM controller is constructed to enhance the driving experience and improve the braking energy economy,which is set up to build the regenerative brake energy management strategy.The braking force coordination control strategy based on the energy consumption level prediction theory by driving style is adopted to adjust the motor torque.As the frictional brake coordinates the regenerative brake in braking process,the demanding brake torque of the vehicle can be satisfied.Moreover,a hardware in the loop simulation platform of the electric vehicle is constructed,and the developed strategy is investigated in this platform.The simulation results verified the adaptability and effectiveness of the proposed strategy.Finally,a control system of braking mode transition and the regenerative brake energy management for the electric vehicle is built based on the developed control strategy,which consists of the microprocessor MPC5748 G,power module,serial communication module,location module,control programming,etc.On this basis,the road experiment equipment of electric vehicle is built to carry out the test on the prescribed test route.Additionally,an improved road test scheme with the six drivers of two different driving styles is designed to verify the proposed strategy.And then,the detailed test results show that the developed strategy is feasible and effective for reducing the vehicle jerk and torque fluctuation during deceleration processes.Moreover,the algorithm of IIDP-BLSTM algorithm is better than particle swarm optimization and support vector regression specifically for the aggressive driving style.The energy economy and driving experience performance of the proposed strategy are improved significantly for the electric vehicle.The research fruits accumulate valuable experience in broadening the implementation of the energy management device and pushing the engineering realization of the regenerative brake energy management for electric vehicle.