Optimal Neuro-Fuzzy Control of Parallel Hybrid Electric Vehicles

Abstract

In this paper an optimal method based on neuro-fuzzy for controlling parallel hybrid electric vehicles is presented. In parallel hybrid electric vehicles the required torque for driving and operating the onboard accessories is generated by a combination of internal combustion engine and an electric motor. The power sharing between the internal combustion engine and the electric motor is the key point for efficient driving. Therefore, we are dealing with a highly nonlinear and time varying plant. Moreover, the estimation of the state of charge of the battery pack is a very important point, which has been considered in this paper. The control strategy will be implemented using the ANFIS method. The controller will be designed based on the desired torque for driving and the state of charge of batteries. The output of controller adjusts the throttle in the combustion engine. The main contribution of this paper is the development of optimal control based on neuro-fuzzy, which maximizes the output torque of the vehicle while minimizing fuel consumption used by the internal combustion engine. Simulation results show very good performance of the proposed controller.