摘要
在新能源产业快速发展背景下,磷酸铁锂(LiFePO4)电池在储能化工领域得到广泛应用。随着人工智能技术的进步,基于机器学习和深度学习的方法已广泛应用于电池材料开发与寿命预测中。通过建模产气、热特征与循环数据,人工智能可有效识别电池失效模式,并为材料与电解液优化提供支持。本文以5Ah LiFePO4软包电池为对象,研究了不同过充电压条件下的产气行为与失效机理,并采用原位电化学质谱、气相色谱及热电偶监测,细致探究了特征气体演化与温度变化。该研究对开发预警装置、提升储能电池安全可靠性具有重要指导意义,并对皮革等复杂化工过程的安全监测也有参考价值。
With the rapid growth of the new energy sector,LiFePO₄batteries are widely employed in energy storage and chemical engineering.Artific Intelligence advances have enabled ML and DL to be extensively used in battery material development and lifespan prediction.By modeling gas emission,thermal behavior,and cycling data,Artific Intelligence can identify failure modes and assist in material and electrolyte optimization.This study examines gas generation and failure mechanisms in a 5 Ah LiFePO₄pouch cell under various overcharge conditions using in-situ differential electrochemical mass spectrometry(DEMS),gas chromatography,and thermocouple monitoring to track gas evolution and temperature changes.The results offer key insights for developing early-warning systems and improving the safety and reliability of energy storage batteries,with relevance to safety monitoring in industrial processes like leather manufacturing.
作者
蔡志亮
黄杰明
姚学轩
李欣
袁淦钦
冯健磊
黎锡泉
CAI Zhiliang;HUANG Jieming;YAO Xuexuan;LI Xin;YUAN Ganqin;FENG Jianlei;LI Xiquan(Dongguan Power Supply Bureau,Guangdong Power Grid Corporation,Dongguan Guangdong 523106,China)
出处
《皮革与化工》
2025年第5期4-6,共3页
Leather And Chemicals
基金
南方电网科技项目“储能电站多参数协同极早期火灾探测预警技术研究”(031900KC23070050)。
关键词
电池材料
储能电池
人工智能
化工开发
寿命预测
battery materials
energy storage batteries
artific Intelligence
chem development
lifetime prediction
