Warning method of fault for lead-acid battery of energy storage system based on its resistance prediction

The fault of the battery affects the reliability of the power supply, thus threatened the safety of the battery energy storage system (BESS). A fault warning method based on the predicted battery resistance and its change rate is proposed. The causes of the resistance change of the battery are classified, and the influencing factors of battery internal resistance are quantified. The prediction model of the battery resistance is constructed by the long-and-short-time memory neural network (LSTM) with the inputs of the state, voltage, current, temperature of the battery. The signal of the fault warning is issued and the cause of the fault is determined with the predicted battery resistance and its change rate. The results of the experiment show that the error of the prediction method of the battery resistance is within 3%. The fault warning method is beneficial to the safe operation of the BESS.