The Maximum Charge and Discharge Power Estimation in Hybrid Electric Vehicle Based on Artificial Neural Network

Abstract

AbstractIn view of the hybrid electric vehicle’s Ni-MH battery system, the paper predicts the maximum charge and discharge power through artificial neural network. The maximum discharge and charge power directly affect the efficiency operation of hybrid electric vehicle. It is one of the most important parameters of Ni-MH battery systems. The estimation can avoid Ni-MH battery systems overcharge and over discharge phenomenon and prolong the service life of battery system. It has important practical value and theoretical significance for reasonable use of Ni-MH battery systems. Firstly from the view of engineering application, check the matching table to get the target value’s relation with the SOC (the rest charge value), voltage, temperature. Secondly select some of the voltage, temperature and the last moment of the predicted maximum charging and discharging power, to be the input variables of the network. Thirdly through the tests, design and optimization the artificial neural network model structure. Finally get the artificial neural network model which can successfully predict the target. As a result, its error is within 10% to meet the expected requirements.Keywordshybrid electric vehiclebatterypowerartificial neural network