Lithium Battery Thermal Runaway Warning Method Based on Multi-Feature Fusion

[Objective] During the operation and storage of lithium batteries, substantial heat is generated. Anomalies in temperature can impact the lifespan and cycling efficiency of lithium batteries, and in extreme cases, may lead to explosions. Therefore, research on thermal runaway warning for lithium batteries is crucial for ensuring their operational safety. [Method] The DTW-Kmeans algorithm was employed to identify anomalies in the temperature rise rate of lithium batteries. Subsequently, the physical characteristic of surface temperature decrease following the opening of the lithium battery safety valve was incorporated. A dual-feature fusion approach was utilized to propose a thermal runaway warning mechanism for lithium batteries. [Result] Repetitive experiments have validated the effectiveness of the proposed early warning algorithm in distinguishing abnormal lithium batteries based on temperature rise rates. Furthermore, it is capable of identifying the sudden change in temperature rise rates from positive to negative in abnormal lithium batteries, achieving a comprehensive recognition accuracy rate exceeding 90%. [Conclusion] The early warning algorithm is able to accurately identify lithium batteries with abnormal temperature rise rates, and can promptly and precisely detect the timing and location of the opening of the safety valve in the lithium battery. Consequently, this early warning algorithm serves as a preemptive measure against thermal runaway in lithium batteries, thereby safeguarding the safe operation of lithium-ion battery packs.