Research On System Structure And Control Strategy Of Hybrid Electric Vehicle

The huge depletion of oil resources by traditional cars has caused serious energy crisis and environmental pollution.To vigorously develop new energy automobile technology has become the primary task of the global automobile industry.Hybrid cars combine the advantages of both pure electric cars and traditional internal combustion engine cars.It not only inherits the advantages of high efficiency and low emission of pure electric vehicles,but also has the strength of high specific energy and specific power of petroleum fuel,which has significantly improved the emissions and fuel economy of traditional internal combustion engine vehicles and increased the driving range of electric vehicles.The moderate price of hybrid cars makes it suitable for industrial applications and becomes an important direction for the development of new energy vehicles.Therefore,the research on key technologies of hybrid vehicles is very important.Based on a comprehensive analysis of the development profile,driving types and key technologies and conducted on the development of energy management and control systems of hybrid vehicles,this dissertation mainly focuses on the energy management strategies of hybrid electric vehicles and the design of structure and control system of power battery and super capacitor hybrid power source.The main work is as follows:With the overall consideration of the accuracy,timeliness,and readiness of the model,the author selects look-up table method and comprehensive experiment to establish the models of engine,motor,battery,drivetrain,vehicle dynamics,vehicle control system and develop a vehicle control system based on NXP's MC9S12XEP100 microcontroller,which realizes the design of interface circuit,power module clock circuit,reset circuit,data acquisition module and CAN bus and succeeds in designing a control system to satisfy high and low tem-perature tests,vibration tests,and electromagnetic compatibility tests.Optimized control of the energy management of hybrid vehicles is needed to optimize the operation of hybrid vehicles and realize good power and energy efficiency.Based on fuzzy control theory and the structural principle of fuzzy controller,this dissertation builds the fuzzy control strategy of the energy management for hybrid drive system,which takes total vehicle demand torque and battery pack as input and engine demand output torque command as output to establish the model of fuzzy control energy management system in Matlab/Simulink and then embedded this model and vehicle model to hybrid car performance simulation software advisor to perform the simulation verification.The contrast between sim-ulation results and electrically assisted energy management strategies shows that the equiva-lent fuel consumption per 100 km reduces 5.2%,HC emission per kilometer lowers 0.279 g,NOX lowers 0.116 g and the system efficiency raises 25.7%.These results indicate that the fuzzy control energy management strategy can reasonably distribute the output torque of the engine and the motor to reduce the fuel consumption of hybrid vehicles more effectively and significantly improve the fuel economy.Although the fuzzy control can achieve good control effect,its control rules are based on expert experience and cannot apply to the changes in operating conditions.To overcome this limitation,a kind of energy management strategy based on Actor-Critic reinforcement learning is designed to obtain optimized control strategies through strategy learning,which is one of the innovations of this dissertation.This method uses a neural network to estimate the state-action value function,which greatly relieves the problem of “dimensional disaster”.And the adopting of Actor network makes it possible for this algorithm to deal with continuous state and action space problems and solves the problem that the rule-based energy management strategies cannot adapt to different driving conditions.According to the simulation ex-periment,the contrast of energy management strategy based on Actor-Critic reinforcement learning and the rule-based energy management strategies suggests the improvement of fuel economy for 46.38%,engine efficiency for 7.8%,electric efficiency for 2.6% and the generat-ing efficiency for 8.8%.The results show that the Actor-Critic enhanced learning energy management strategy has a good optimization effect in lifting the fuel economy,engine effi-ciency,motor efficiency etc.significantly.The performance of electric vehicles depends on the power system structure and its control strategy.This dissertation innovatively proposes a novel topology and fuzzy logic control strategy for power battery and super capacitor hybrid energy sources.A bidirectional DC-DC converter is used to couple power battery and super capacitor to give full play to the advantages of power batteries and super capacitors and improve the transient characteristics of electric vehicles,which reduces the damage of high current charge and discharge to power battery.Compared to the single battery system,the hybrid energy sources system reduces the energy consumption for 14.67% and increases the driving range for 17.79%.The experimental results show that the structure and fuzzy logic control strategy of the hybrid power source of the power battery and super capacitor have obvious advantages in terms of energy utilization and mileage improvement.Compared to the electric vehicles only with power battery power sources,the voltage and current curves of the power battery are also significantly smoother,indicating that this structure and control strategy can effectively protect the power battery.