Work

A Comparative Performance Evaluation of Neural Network Algorithms Based State of Charge Estimation for Lithium-ion Battery

Year: 2020
Type: article
Abstract: This work presents a comparative analysis of state of charge (SOC) estimation for lithium-ion battery using neural network algorithms. The lithium-ion battery has been operating successfully in the automotive industry due to the long-life cycles, low memory effect, high voltage, and high energy density. As such, numerous research works have been conducted on lithium-ion battery towards estimating SOC. The conventional and model-based SOC estimation approaches have shortcomings including heavy computational calculation and inaccurate battery model parameters determination. Therefore, neural network algorithms based SOC estimation have received huge attention since they have the adaptively to adjust the network parameters automatically without battery model. Three prominent neural network algorithms including backpropagation neural network (BPNN), radial basis function neural network (RBFNN) and recurrent nonlinear autoregressive with exogenous inputs neural network (RNARXNN) are used to compare the SOC estimation results. The three methods are validated by battery experimental tests and electric vehicle (EV) drive cycles. The results demonstrate that RNARXNN is dominant to BPNN and RBFNN algorithms in obtaining high SOC accuracy with the low computational cost.
Language: English
Open Access status: bronze
OpenAlex ID: W7215002932
FWCI:
Cites: 18
Cited by: 1
Related to: 10