Title: Adaptive neuro control of parallel hybrid electric vehicles

Authors: Morteza Mohebbi, Mohammad Farrokhi

Addresses: Department of Electrical Engineering, Iran University of Science and Technology, Narmak, Farjam Street, Tehran, 16846-13114, Iran. ' Department of Electrical Engineering, Iran University of Science and Technology, Narmak, Farjam Street, Tehran, 16846-13114, Iran

Abstract: In this paper, an adaptive control method based on neural networks for controlling parallel hybrid electric vehicles is presented. Power sharing between the internal combustion engine and the electric motor is the key point for efficient driving. The control strategy will be implemented using a neural network. The controller will be designed based on the powertrain desired torque and the state of charge of the batteries. The output of the controller adjusts the fuel throttle angle in the combustion engine. The main contribution of this paper is the development of an online controller based on neural networks, which maximises the output torque of the engine while minimising fuel consumption. In other words, a compromise solution between torque and fuel can be achieved. Also, the state of charge of the batteries, which has been estimated by the ampere-hour counting technique, has been considered. Simulation results show good performance of the proposed controller.

Keywords: adaptive control; fuel minimisation; neural networks; parallel hybrid vehicles; intelligent control; hybrid electric vehicles; online controllers; fuel consumption; engine torque; batteries; battery charge; simulation.

DOI: 10.1504/IJEHV.2007.014444

International Journal of Electric and Hybrid Vehicles, 2007 Vol.1 No.1, pp.3 - 19

Published online: 08 Jul 2007 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article