Abstract
Lithium-ion batteries are widely used in energy storage systems and electric vehicles because of their high power density, energy density, long cycle life, low self-discharge rate, etc. However, with the popularization of a large number of batteries, a series of accidents caused by battery failure occur frequently. In particular, accidents such as spontaneous combustion and explosion caused by thermal runaway will cause serious consequences. In this paper, a hybrid model based fault diagnosis method is proposed to reduce the occurrence of accidents. The internal temperature and SOC of batteries are estimated by the physical model of batteries. This method estimates the internal temperature and SOC of the battery through the battery electrothermal coupling model. Combined with the battery surface temperature, battery voltage and battery currentr as the input of LSTM(Long Short-Term Memory Model), the hybrid model can accurately predict the surface temperature and internal temperature of the battery. Through the threshold method to determine the occurrence of thermal runaway and analyze the causes, the accurate prediction of battery thermal runaway can be realized.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call