基于大数据驱动的阀控式铅酸蓄电池剩余使用寿命预测方法  

Residual Life Prediction Method of Valve-controlled Lead-acid Battery Based on Big Data Driven Technology

作　　者：李文辉 王瑜曈 王振伟 LI Wen-hui;WANG Yu-tong;WANG Zhen-wei(State Grid Gansu Electric Power Company,Lanzhou 730050 China;Information and Communication Company of State Grid Gansu Electric Power Company,Lanzhou 730050 China;Beijing Slaxton Technology Co.,Ltd.,Beijing 100068 China)

机构地区：[1]国网甘肃省电力公司,甘肃兰州730050 [2]国网甘肃省电力公司信息通信公司,甘肃兰州730050 [3]北京斯莱克顿技术有限公司,北京100068

出　　处：《自动化技术与应用》2023年第11期151-154,共4页Techniques of Automation and Applications

摘　　要：阀控式铅酸蓄电池剩余使用寿命对其使用安全性存在较大影响,为准确掌握其使用状态,研究应用大数据驱动技术的阀控式铅酸蓄电池剩余使用寿命预测方法。分析阀控式铅酸蓄电池在充电放电过程中产生的退化健康指标,并提取主成分特征,将其作为支持向量机模型的输入,预测电池剩余使用寿命。通过免疫完全学习型粒子群优化算法优化模型中的参数,提升模型的泛化能力和预测准确度。测试结果表明:优化后模型的预测结果与实际结果的拟合程度较高,具备良好的回归性能,均方根误差值均低于0.1%的情况下,完成阀控式铅酸蓄电池剩余使用寿命预测。The remaining service life of valve regulated lead-acid battery has a great impact on its service safety.In order to accurately grasp its service state,the remaining service life prediction method of valve regulated lead-acid battery using big data-driven technolo-gy is studied.The degraded health index of valve regulated lead-acid battery during charging and discharging is analyzed,and the principal component feature is extracted as the input of support vector machine model to predict the remaining service life of bat-tery.The parameters in the model are optimized by immune complete learning particle swarm optimization algorithm to improve the generalization ability and prediction accuracy of the model.The test results show that the optimized model has a high degree of fit with the actual results,has good regression performance,and completes the prediction of the remaining service life of valve regulated lead-acid battery when the root mean square error is less than 0.1%.

关 键 词：大数据驱动技术 阀控式 铅酸蓄电池 剩余使用寿命 预测方法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP311[自动化与计算机技术—控制科学与工程]