Research on intelligent early warning method for battery safety based on long and short-term memory loop neural network

With the wide application of batteries in mobile devices, electric vehicles and other fields, batteries may suffer from performance degradation, overheating, overcharging, overdischarging and other problems during the process of use, which may lead to battery failures, damages and even dangerous situations, so the safety of batteries is of great concern. There are some battery safety warning methods available, but there are shortcomings such as insufficient data series modelling capability and low prediction accuracy. Therefore, a more efficient and accurate battery safety warning method is needed. In this paper, a battery safety intelligent warning method based on long and short-term memory recurrent neural networks is provided, which utilises a highly intelligent method to quickly and accurately warn the battery safety state. The method is characterised in that the method monitors and analyses real-time data of the battery using a long and short-term memory recurrent neural network to predict possible safety problems or failures of the battery, thereby realising rapid and accurate early warning of the battery safety state. The method has a high prediction accuracy and real-time performance, and is capable of timely detecting potential safety problems of the battery and improving the safety and reliability of the battery use.