Automobile torque distribution control method based on fuzzy cerebellar model articulation controller (FCMAC)

Abstract

The invention discloses an automobile torque distribution method based on a fuzzy cerebellar model articulation controller (FCMAC). Fuzzy control and a cerebellar model articulation controller (CMAC)are combined while the driving intension of a driver is considered, and a CMAC is embedded into a fuzzy controller so as to constitute the FCMAC. The FCMAC learns fuzzy rules through the CMAC, and theGauss fuzzy membership function serves as the actual mapping function so that the division mode of an input space and the activation characteristics of an association unit can be adjusted on line andin real time; the distribution coefficient of the required torque of an engine is amended through adjustment of an output weighed value; and finally, the required torque value of the engine is calculated through defuzzification, and torque distribution of a hybrid electric vehicle under the hybrid-drive condition is achieved.

Description

technical field [0001] The invention relates to the field of hybrid electric vehicle control, in particular to a method for controlling torque distribution of a complete hybrid electric vehicle based on a fuzzy cerebellar model neural network. Background technique [0002] Hybrid electric vehicles can have multiple operating modes such as pure electric mode, hybrid drive mode, engine drive mode, and energy feedback during actual operation. The torque distribution control strategy is based on the size of the required torque and the state of charge (SOC) of the power battery. The status information of the vehicle such as the remaining power realizes the free switching of various modes such as pure electric, hybrid drive and engine drive, making it have the advantages of traditional vehicles and pure electric vehicles. In the hybrid drive mode, the torque distribution of hybrid electric vehicles In the process of vehicle torque distribution, the control strategy can not only en...

AI Technical Summary

Problems solved by technology

[0008] In order to solve the fuzzy rules existing in the hybrid vehicle torque control system based on fuzzy control, it needs to be determined according to the experience of experts, but different experts have different experience values, and it cannot be guaranteed that each algorithm is in the optimal state, and the vehicle is in a complex state. Under the environment of driving conditions, different fuzzy rules are required to realize the technical problem of torque distribution of the whole vehicle. The present invention provides a method for controlling the torque distribution of the whole vehicle of hybrid electric vehicles based on the fuzzy cerebellum model (FCMAC) neural network. It comprehensively considers the SOC remaining power and the driver's operating intention under the current actual driving conditions, uses the input space division method and the activation characteristics of the associative unit, which can be adjusted in real time online and the learning accuracy is optimized due to the FCMAC neural network of the CMAC neural network. The rules effectively improve the problem that fuzzy rules in fuzzy control are only determined based on expert experience, and the fuzzy reasoning system can better judge the driver's operating intention by identifying the rate of change of the accelerator pedal, thereby optimizing the torque distribution control strategy and improving Improve the fuel economy of the engine and reduce emissions, while keeping the SOC within a reasonable range and prolonging the service life of the power battery

Images