A hierarchical warning control method for battery modules based on multi-parameter fusion

Early warning control integrating multiple parameters is crucial for the safe operation of battery modules. This paper proposes a hierarchical early warning control method for battery modules based on multi-parameter fusion. Firstly, considering battery failure events and their underlying internal mechanisms changes, Fault Tree Analysis (FTA) is employed to analyze various abuse behaviors leading to battery failures. Structural Importance Analysis is then integrated to extract the critical abuse behaviors. Consequently, the State of Charge (SOC), temperature, State of Health (SOH), and internal stress of the battery module are identified as key indicator parameters affecting its safety performance. Secondly, a generic State of Safety (SOS) model for the battery module is established based on an “abuse function-exponential function” coupling model. This SOS model under multi-parameter fusion is constructed by integrating the SOC, temperature, SOH, and internal stress of battery modules, serving as a warning indicator. Finally, the Pelican Optimization Algorithm-Long Short Term Memory (POA-LSTM) network is utilized to predict the individual parameters and the SOS of the battery module. Based on the SOS prediction results and prediction absolute errors, a hierarchical early warning control method is proposed to ensure the timeliness and reliability of the warning control. Simulations and experiments are conducted to validate the proposed method. The experimental results show that this warning control method ensures the reliability of warning control while providing a warning of serious abuse of battery modules 100 s in advance. This method provides a new solution for the safe operation and early warning of energy storage battery modules.