Diagnosis of battery external short circuits based on an improved second-order RC fault model and recursive least squares identification method

The external short-circuit fault of Li-ion batteries poses significant safety hazards, potentially leading to thermal runaway, fires, and even explosions, which severely threaten personal and property safety. Currently, the detection methods for Li-ion battery external short-circuit faults are still inadequate, making timely and accurate diagnosis of such faults crucial. This paper proposes a method for diagnosing external short-circuit faults in Li-ion batteries, based on an improved circuit fault model and the recursive least squares method. First, to obtain more accurate and safer real-world external short-circuit data of Li-ion batteries, a multi-dimensional data acquisition system was established, allowing external short-circuit experiments under different ambient temperatures and states of charge (SOC) while precisely collecting data such as battery voltage, current, and temperature. Second, an improved second-order RC circuit fault model specifically for the external short-circuit state of Li-ion batteries is proposed, which effectively captures the nonlinear characteristics of Li-ion batteries under external short-circuit conditions. Finally, experimental results show that the fault diagnosis strategy based on this model can accurately detect faults within 4 s of the occurrence of an external short circuit and demonstrates robust performance under different temperature and SOC conditions.