| New energy vehicles have become the inevitable trend of future automotive industry development, hybrid electric vehicles (HEV) are one of most promising new energy vehicles with the advantages of good performance, long range for traditional fuel cars and the advantages of low emission, high energy efficiency and regenerative braking energy recovery for future electric vehicles. At present, the dynamic performance of HEV still can not catch up with the traditional fuel cars, the main reason to this problem is that the power density of battery energy storage system is low which can not meet the high power density requirements when the vehicles at startup, accelerating or climbing a hill state. In addition, battery at regenerative braking state can’t rapidly recover the regenerative braking energy which causes the waste of energy. Therefore, it is necessary to add an auxiliary energy storage system to meet the demand of high power density of HEV, at the same time, recover the regenerative braking energy quickly and fully. Ultracapacitor (UC) reling on large power density, short charging time and long cycle life is undoubtedly one of the best choices. UC combines with battery constituting the hybrid energy storage system (HESS) which can complement the advantages of battery and UC, meanwhile, maximum limit the insufficient of battery or UC single power, the performance of energy storage system of HEV is greatly improved. This paper studies the energy management system (EMS) of HESS including power splitting between battery and UC, regenerative braking energy recovery, energy efficiency optimization and state of charge (SOC) prediction, the main researches and the corresponding results are as follows:(1) The topological structures of HESS in HEV are given in this paper, and the adaptive filter power splitting (AFPS) control strategy is put forward on the basis of analyzing the topology principle and working state. Control objectives of AFPS are that UC bears the peak power of load power demand, and battery mainly undertakes the average power, so as to achieve the UC "cutting the peaking power and filling the valley power" to battery. At the same time, in order to make the HESS adapting to the different driving cycles, it is necessary to design and optimize the control parameters of AFPS. The simulation results under advanced vehicle simulator (ADVISOR) verify the effectiveness of proposed AFPS control strategy and parameters optimization method.(2) To improve the efficiency of regenerative braking energy recovery in HEV with HESS, a braking force distribution strategy has been introduced. In ADVISOR, the braking force distribution strategy is based on speed of HEV. While in this paper, the vehicles body force, road environment and structure of HEV are used to distribute the braking force between regenerative braking force and friction force according to the ECE R13 braking regulations. Meanwhile, the safe charging conditions of battery and UC are taken into consideration. The modeling and simulation verification of the proposed braking force distribution strategy have been conducted under ADVISOR, and the results indicate that braking force distribution strategy can achieve a good distribution between regenerative braking force and friction force, the efficiency of regenerative braking energy recovery is enhanced and the HEV mileage prolonged.(3) By the way of enhancing the energy utilization efficiency and reduce the losses of HESS, a maximum energy efficiency optimization algorithm has been proposed. The mathematical models of battery, UC and DC/DC converter in HESS are established so as to solve the power losses of each function module. Optimization goal is to reduce the losses of HESS with energy efficiency being improved. Due to optimization belongs to nonlinear programming problem, the approximate programming method is adopted to solve it. Optimization algorithm has been programmed under ADVISOR, the simulation results show that the power losses after optimization of HESS are reduced greatly, and the energy efficiency are enhanced. So does driving motor, the working points in high efficiency area are increased.(4) For the purpose of predicting the SOC of battery in HESS rapidly and accurately, the Bayesian Extreme Learning Machine (BELM) is introduced in this paper. The basic principle of extreme learning machine and bayesian extreme learning machine are presented in detail. In order to enhance the abilities of fitting and generalization, bayesian method is used to optimize the weight of output layer of extreme learning machine. Under the condition of driving cycles, the working voltage, current, surface temperature, historical value of SOC and internal resistance are chose to predict the real value of SOC, and the energy feedback process of regenerative braking condition is also taken in consideration. The simulation results under MATLAB and example results demonstrate that the proposed prediction model has higher predicted accuracy, and achieve real-time and accurate SOC prediction.(5) The overall experiment structure of EMS in HESS for HEV is given in this paper, and software and hardware are designed minutely, so the small power HESS prototype has been developed. The experiment study of EMS is carried out, and the experimental results can prove the practical feasibility of the proposed EMS in this paper. |
PDF Full Text Request