SOC estimation of NiMH battery for HEV based on adaptive neuro-fuzzy inference system.

Author

Sun, BingXiang ; Wang, Lifang ; Liao, Chenglin

Author_Institution

Inst. of Electr. Eng., CAS, Beijing

fYear

2008

fDate

3-5 Sept. 2008

Firstpage

1

Lastpage

4

Abstract

The input variables of adaptive neuro-fuzzy inference model are selected by correlation analysis based on the parameter options of NiMH batteries. The output characteristic of NiMH battery is generalized by Expert experience, and the ANFIS model is established including the selection of input variables and membership functions. Then the fuzzy rules are adjusted by BP neural network algorithm and least-squares algorithm. The data used in model identification and model validation is close to the data of actual working condition. The SOC estimation method of adaptive neural fuzzy inference combines the advantages of fuzzy inference and artificial nerve network.

Keywords

backpropagation; fuzzy set theory; hybrid electric vehicles; least squares approximations; neural nets; nickel; power engineering computing; secondary cells; BP neural network algorithm; HEV; Ni-JkH; SOC estimation; adaptive neuro-fuzzy inference system; artificial nerve network; battery; correlation analysis; hybrid electric vehicles; least-squares algorithm; state of charge; Adaptive systems; Artificial neural networks; Battery management systems; Electrical resistance measurement; Fuzzy neural networks; Hybrid electric vehicles; Input variables; Power system modeling; State estimation; Voltage; Adaptive Neuro-Fuzzy Inference System (ANFIS); Hybrid electric vehicle; Nickel-metal hydride battery; State of charge;

fLanguage

English

Publisher

ieee

Conference_Titel

Vehicle Power and Propulsion Conference, 2008. VPPC '08. IEEE

Conference_Location

Harbin

Print_ISBN

978-1-4244-1848-0

Electronic_ISBN

978-1-4244-1849-7

Type

conf

DOI

10.1109/VPPC.2008.4677676

Filename

4677676