Prediction of state-of-charge effects on lead-acid battery characteristics using neural network parameter modifier

摘要

In this study, impedances of SABA BATTERY 6SB6 in different SOCs are applied to obtain the equivalent circuit parameters using Champlin method in different SOCs. Champlin method answers are used as Zview initial values to get fit results and the Artificial Neural Network (ANN) is trained by these final results. The presented ANN inputs are SOCs and outputs are equivalent circuit parameters. The completed network responses are perfectly adjusted to the experimental parameters. Accuracy of this method has been verified by using the measured data and they have shown a high consistency to experiment. So that a model is extracted in which one can approach an equivalent circuit model with specified parameters simply by entering the SOC.