Font Size: a A A

Energy Management Strategy Based On Neural Network For HEV Lithium Power Batteries

Posted on:2013-08-06Degree:MasterType:Thesis
Country:ChinaCandidate:R ZhaoFull Text:PDF
GTID:2232330362471477Subject:Control theory and control engineering
Abstract/Summary:PDF Full Text Request
Hybrid Electric Vehicle (Hybrid Electric Vehicle,HEV) power battery packstate of charge (State of charge, SOC) estimation is the most difficult and the key partof the HEV battery management system. Accurate estimate of the battery state ofcharge is not only able to provide for the HEV driving when the battery energy storagesituation, but also a battery management system to charge and discharge managementand balance control management.Estimation algorithm about HEV lithium-ion battery state of charge is researchedin this paper. By the analysis of lithium-ion battery working principle, the emergenceand change mechanism of electric quantity, the battery charging and dischargingcharacteristics, and considering of the battery current, temperature, other factors whichaffect SOC algorithm like self-discharge rate, cycle number and working conditions, anew method that Estimation algorithm about HEV lithium-ion battery state of chargebased on adding momentum self-recurrent wavelet neural network is proposed in thispaper. At the same time, the models of neural network are built in MATLAB, and itsconvergence is proved. Estimation algorithms about HEV lithium-ion battery packSOC, such as artificial neural network, BP neural network, wavelet neural network andSelf-recurrent wavelet neural networks are simulated, The respective results and errorare compared and analyzed. The results show that this method can not only improvethe estimation accuracy of the SOC, and network convergence fast on-line estimation.In order to improve the efficiency of battery balanced charge and discharge andreduce overall energy consumption in Parallel Hybrid Electric Vehicle, this paperpresents battery power real-time balance control strategy. The strategy not only canreduce overall energy consumption, but also can extend the service life of the powerbattery. The simulation results show the effectiveness of the proposed method. Theresults show that, the presented method reduces the energy consumption and prolongs the life of the power batteries, improves the efficiency of battery charging anddischarging for HEV.
Keywords/Search Tags:HEV power battery pack, State of charge(SOC), Self-Regressionwavelet neural network(SRWNN), balanced charge and discharge
PDF Full Text Request
Related items