Research Of PHEV Energy Online And Realtime Optimization Control With Self-Adaptive To Driving Cycle

Fuel economy is an important indicator to measure the performance of Hybrid Electric Vehicle (HEV). Compared with traditional vehicles, HEV has the function of regenerative braking. Its driving motor is in generator mode during braking, recycling the braking energy, storing the energy in the energy storage unit, so as to offer energy for the vehicle driver or the vehicle electronic equipment in the future, which can improve the fuel economy of vehicles. In addition, HEV is driven by at least two sources of power (usually engine and electric motor/battery), and through appropriate control strategy, the two or multiple power sources are working in a range of high efficiency, significantly improving the vehicle fuel economy. Derived from HEV, Plug-in Hybrid Electric Vehicle (PHEV) is a transitional product from the traditional gasoline vehicles to a pure electric vehicle (Battery-powered Electric Vehicle BEV). It will play an important role and have a very large potential in a long time before the battery reaches the BEV driving requirements.The braking system of HEV is the hydraulic brake, usually allocating the braking force between the front and rear axles according to a certain percentage, the consequence of which is that the front and rear wheels can not be stopped at the same time, and this is harmful to the vehicle stability and security. Moreover, the regenerative braking energy can not be made to get the optimal use because the braking force can not be precisely controlled. The HEV universally adopts experience-based logic threshold control on the front axle regenerative braking and friction braking torque distribution. The choice of threshold is difficult because it depends on the engineering experience. As to the driving energy control, HEV battery capacity is small and has no external charging device, so it can only work in CS (Charge-Sustaining) Mode. However, in addition to inheriting the advantages of the HEV, PHEV, equipped with a larger capacity battery and an external charging device, can work in the CS mode as well as the CD (Charge-Depleting) Mode, which means, it is a pure electric mode of battery-powered-alone vehicles. The energy control strategy used in PHEV is the CSCD control strategy which is mainly based on rules (Rule-Based. RB); it means to use the battery to drive PHEV first, then when the battery state of charge (SOC) drops to a certain level, the engine can be started, working together with the battery. At this point, the energy control is the same as the one of HEV. RB control strategy can not make the battery and the engine work in the optimal area, especially in the CD mode. When the battery current is large, the battery internal losses will be relatively large, thus affecting the efficiency of the battery, resulting in decreased fuel economy.In order to solve these problems to improve the PHEV fuel economy, the research of this paper focus on the following aspects:Because the braking torque in the hydraulic brake system can not be precisely controlled, causing the vehicles not obtaining the largest regenerative braking energy, this paper puts forward a new structure of PHEV----the structure of using the system of Electronic Mechanical Brake (EMB) to provide braking torque to each wheel, precisely controlling the braking torque. The paper analyzes the principles of the new PHEV structure, establishing the vehicle dynamics model. This paper builds the corresponding experimental platform, getting the characteristic curve of the engine, electric/generator and lithium-ion battery by experimental measurements; based on experimental data, the computer simulation model of the engine, electric/generator and lithium-ion battery is also established in this model. This model contributes a lot to the follow-up research.To solve the problem that the permanent magnet motor of the EMB system produces the phenomenon of the demagnetization under high temperature, thus affecting the performance of system, this paper proposes a new EMB structure to provide braking torque to the PHEV. Furthermore, this paper also studies the function of the anti-lock braking of the EMB system. There is "trembling" in the sliding mode control algorithm commonly used in the ABS system, therefore on the basis of the road surface recognition, this paper brings forward the ABS of sliding mode control based on fuzzy switching gain adjustment. A new design of vehicle body speed observer is also proposed to solve the problem that the vehicle body speed can not be accurately measured when the wheels are skidding or locking.Because the EMB braking torque can be precisely controlled, this paper uses ideal braking force distribution method (I-curve) to distribute braking force to front and rear axle, and fuzzy logic control strategy is used to replace the experience-based logic threshold control strategy, the greatest degree of regenerative braking torque can be achieved by the precise allocation between the front axle regenerative braking torque and friction braking torque. The simulation results show that this method can recycle more regenerative braking energy.By using dynamic programming (DP) and quadratic programming (QP), the energy offline of the PHEV whose operating conditions are already known (UDDSn HFEDS、US06、SC03and cold UDDS) is optimized; the energy globally optimal allocation scheme is found out; the advantages and disadvantages of the DP and QP optimization methods are comparative analyzed; and a basis is provided for the choice of the global optimization method.Using the driving cycles of the19types of passenger cars of the U.S. Environmental Protection Agency as the classification criteria, this paper proposes driving cycle prediction algorithm based on Radial Basis Function Neural Network (RBFNN). On the basis of the driving cycle forecast, two PHEV energy online and realtime control strategies are proposed—one is the RBFNN resulting from using the QP to optimize the results of training; the other is RBFNN resulting from using the QP to optimize the results of training. By using the driving cycle prediction algorithm and energy optimization of the integrated control of the driving cycles, the PHEV energy can be controlled. Compared with other energy control method, the simulation results show that this method can significantly improve the fuel economy.In order to verify the feasibility and effectiveness of the energy online and realtime optimization of the control strategy proposed in this paper, pure electric vehicle is converted into a series PHEV on the basis of the parameter optimization; the online and realtime optimal energy control strategy based on the results of the DP optimization is downloaded to the dSPACE as a energy control strategy of a real car. The real vehicle test results show that the online and realtime optimal energy control strategy proposed in this paper in high speed, medium speed, low speed and mixed speed driving clcye, PHEV fuel economy have dramatically increased----11.6%,13.1%,13.5%and10.4%respectively. The real vehicle test results show that the energy online and realtime optimal control strategy proposed in this paper is feasible and effective, providing a new approach of online and realtime optimal energy control strategy for commercial PHEV.