电力学报2025，Vol.40Issue(5)：339-351,13.DOI:10.13357/j.dlxb.2025.037
储能舱排气防爆机器人排气控制方法
Exhaust Control Method for Exhaust Explosion-proof Robot in Energy Storage Cabin
摘要
Abstract
Aiming at the risk of explosion caused by combustible gas accumulation induced by thermal runaway in lithium battery energy storage cabin,an exhaust control method based on long short-term memory network(LSTM)is proposed,and an exhaust explosion-proof robot in energy storage cabin is designed in this paper.The mechanical structure,software system and hardware system of the robot are designed according to the envi-ronmental characteristics of the energy storage cabin,so that it has the ability of autonomous operation and data acquisition under the failure of the battery management system.The gas evolution law of thermal runaway of lith-ium battery is analyzed through experiments,and the explosion limit threshold is determined by combining the explosion formula of mixed gas.The electro-thermal coupling model is used to integrate the data of battery tem-perature,state of charge and combustible gas concentration,and a prediction model is constructed to predict the future trend.Based on the predicted results and the explosion limit threshold,the proportional integral(PI)con-trol method is used to dynamically adjust the exhaust rate and time to ensure that the concentration of combusti-ble gas in the cabin is always below the lower explosion limit.The test accuracy of the LSTM prediction model is 99.54％.The simulation results show that,this proposed exhaust control method can predict the change trend of temperature and gas concentration in advance,and reduce the concentration of combustible gas to a safe range within 130 s which can effectively achieve explosion-proof of energy storage cabin.The exhaust explosion-proof robot and control method proposed in this paper realize the closed-loop application from mechanism research to intelligent control,providing an efficient and reliable solution for the fire safety of lithium battery energy storage cabin,and significantly improving the safe operation level of energy storage system.关键词
电力机器人/锂电池储能舱/长短期记忆网络/自动控制/防爆/排气/消防Key words
electriced robot/lithium battery energy storage cabim/long short-term memory(LSTM)/automatic control/explosion-proof/exhaust/firefighting分类
信息技术与安全科学引用本文复制引用
邓贝,樊绍胜,熊铭轩..储能舱排气防爆机器人排气控制方法[J].电力学报,2025,40(5):339-351,13.基金项目
国家自然科学基金项目(62271087). （62271087）
