Sodium sulfur batteries contain a large amount of high-purity sodium,which leads safety risks such as combustion and explosion during the automated disassembly process.A modified SSA-BP neural network algorithm was proposed to predict the maximum temperature of tool processing in response to the safety issues of sodium sulfur batteries during turning and disassembly.The real-time temperature of tool machining was calculated using ABAQUS software,and the reliability of the simulation data was verified through battery disassembly experiments.Then,samples were established based on simulated temperature data,and the SSA-BP neural network algorithm was optimized using Tent chaotic mapping to establish a tool temperature simulation prediction model.The experimental results show that the simulation prediction model has fast convergence speed,strong robustness,and small model error.