热失控
残余物
热的
锂(药物)
电池(电)
控制理论(社会学)
计算机科学
功率(物理)
热力学
物理
算法
医学
人工智能
控制(管理)
内分泌学
作者
Xi Cao,Jin Du,Qu Chang,Jiabin Wang,Ran Tu
标识
DOI:10.1016/j.est.2023.109661
摘要
Addressing the challenges in detecting the early stage of thermal runaway caused by overcharging of lithium-ion batteries. This paper proposes an early diagnosis method for overcharging thermal runaway of energy storage lithium-ion batteries, which is based on the Gramian Angular Summation Field and Residual Network. Firstly, the surface temperature data of normal charging and overcharging of lithium iron phosphate battery are obtained through experiments; then the temperature data of different charging periods are divided into stages, and the different stages are converted into a two-dimensional thermodynamic chart by using Gramian Angular Summation Field encoding; finally, using the Residual Network to extract features from temperature thermodynamic charts, so as to achieve the classification of normal charging, very early and early-mid stage of overcharging thermal runaway. According to the experimental results, the accuracy of this classifier can reach 97.7 %, it can diagnose whether the batteries have undergone thermal runaway before the surface temperature is lower than 50 °C after overcharging, and accurately distinguish the current stage of overcharging thermal runaway. This diagnostic method can provide a reference for the safe monitoring and early warning of lithium-ion batteries in energy storage power stations.
科研通智能强力驱动
Strongly Powered by AbleSci AI