State-of-charge (SOC) estimation using T-S Fuzzy Neural Network for Lithium Iron Phosphate Battery

摘要

Although lithium battery has the characteristics of high charge and discharge rate and energy density, its chemical activity is very high. Since the SOC of lithium battery cannot be directly tested, this paper presents a method of estimating the SOC of the battery by the T-S fuzzy neural network regression. Firstly, a T-S fuzzy neural network regression model was constructed. Take the battery voltage, battery current and battery temperature as the training input of the model, and take the corresponding SOC as the training output of the model. And then, used the T-S fuzzy neural network algorithm for model training . Finally, the training model was applied to the battery SOC estimation. The experimental results show that this method can estimate the SOC effectively, improve the estimation accuracy, and has high computational efficiency. This model may provide a theoretical reference for the model construction of future battery charge estimation system.