Soft Faults Diagnosis Based on LightGBM in Lithium-Ion Battery Module

Lithium-ion battery energy storage system (BESS) is an important flexible regulation resource in new power systems. Nowadays, lithium-ion BESS has entered a period of large-scale development. Soft faults in lithium-ion BESS threaten their safe operation and restrict their rapid development. Therefore, timely detection and accurate diagnosis of different types of soft faults in battery module are crucial for the safe operation of BESS. Therefore, a lightGBM based method for detecting soft faults in lithium-ion battery module was proposed, which can effectively diagnose two types of soft faults. Firstly, in order to simulate the soft faults of battery module, a second-order RC model with unknown battery parameters was established based on Whale Optimization Algorithm (WOA). Secondly, a fault diagnosis method based on lightGBM was proposed in order to enhance the ability to distinguish large sample and high-dimensional battery module fault data sets. Finally, the effectiveness of the proposed method is verified through simulation.