Artificial neural network in estimation of battery state-of-charge (SOC) with nonconventional input variables selected by correlation analysis

The selection of input variables is important to improve prediction accuracy of artificial neural network (ANN). A three-layer feedforward back-propagation (BP) ANN is presented to estimate and predict battery state-of-charge (SOC) with nonconventional input variables selected. Initially in this paper, a few candidate input variables are derived from three basic input variables: discharging current, discharging time and battery terminal voltage. Then, three techniques of correlation analysis, linear correlation analysis (LCA), nonparametric correlation analysis (NCA) and partial correlation analysis (PCA), are used to select input variables and the results are compared. With several nonconventional input variables included in the input sets, high prediction accuracy of the ANN model is obtained.