Research On Torque Distribution Strategy Of Hybrid Electric Vehicle Based On FCMAC Neural Network

When the hybrid electric vehicle is traveling,and in the context of the power performance,Torque distribution strategy should be adjusted in real time for the different driving conditions,and the torque distribution between the power sources in the power system,and the purpose of saving energy and emission reduction.In this paper,based on the vehicle dynamics structure of parallel hybrid electric vehicle,introduced main components modeling method;Secondly,according to the fuzzy control theory and the fuzzy control system structure,in the hybrid driving mode,the influence of the driver's operation intention on torque distribution is comprehensively considered,and the difference value of the optimal torque of the engine at the optimal torque of the engine at the current speed and the moment of vehicle demand? T,SOC and the accelerator pedal change rate ? are designed as input,and the engine torque distribution coefficient P as the output of the three-dimensional fuzzy controller;Established on this basis,by considering the fuzzy rules of fuzzy controller based on expert experience and certain,cannot guarantee each algorithm in the optimal state,and in the process of vehicle driving,complex conditions change,under different working conditions require different fuzzy rules to realize the vehicle torque distribution.In this way,the fuzzy control is combined with the neural network of the Cerebellar Model Articulation Controller(CMAC),and the fuzzy rules are not only based on the expert experience,but also through the CMAC neural network to learn the fuzzy rules and optimize the rules,replace the fuzzy rule library,and design the Fuzzy Cerebellar Model Articulation Controller(FCMAC)neural network controller to ensure that each algorithm is optimal.Finally the AVL-Cruise software builds a car dynamics simulation model,and in the Matlab/Simulink environment,and the Matlab/API way to achieve the simulation,and in the meantime,the dynamic performance of the nedcs,it's a simulation of the fuzzy control strategy and the FCMAC neural network control strategy,and it proves that the two kinds of control strategies proposed in this paper can achieve the torque distribution,but the latter control is better,and it can be more efficiently.