An adaptive intelligent system to minimize energy use for a parallel hybrid electric vehicle

This paper elaborates on the synthesis of a neuro-fuzzy controller and also its application to hybrid vehicle powertrain control system. This technique builds a fuzzy inference system for which the parameters can be updated to achieve a preferred input–output mapping. The procedure of learning (guidelines on the parameter updates) is essentially composed of the gradient descent algorithm as well as the least square estimate method. The hybrid vehicle used in this study is the parallel hybrid vehicle. With its small engine, this type of hybrid vehicle is made for steady highway driving. In addition, it composes an electrical device that provides acceleration and deceleration along with a battery management system. A fuzzy logic inference system is used for energy administration. In which, the inference system states the guidelines on recording the nonlinearities in the mapping. The neuro-fuzzy controller is able to accomplish the task of an inverse model with an adaptive fuzzy logic controller. This paper will draw upon the growth and advancement in SOC estimation method and power management system of lithium-ion battery for future high-tech EV applications. It is shown that the intelligent management system proposed in this paper has a significant improve in performance to that of the other existing systems.