摘要
为精确评估输电塔运行状态,以1E3-SZ2输电铁塔为研究对象,建立不同工况下输电塔应力及位移数据库。同时,基于决策树改进的XGBoost模型进行数据挖掘,得到全塔杆件对工况指标的影响权重,确定了20个适合于各工况下的敏感杆件。针对选取的敏感杆件开展1E3-SZ2输电铁塔应力监测真型实验,基于实测数据,运用机器学习方法评估输电塔运行状态。结果表明:在不同工况下,机器学习方法能够准确评估各杆件的应力状态,实时快速掌握输电塔运行状态,并对潜在危险发出预警,具有重要的理论意义和工程价值。
In order to accurately evaluate the operating status of transmission towers,the 1E3-SZ2 transmission tower was taken as the research object,and a database of stress and displacement of the transmission tower under different working conditions was established.At the same time,data mining was carried out based on the XGBoost model improved by decision tree to obtain the influence weights of all tower members on the working condition indicators,and 20 sensitive members suitable for each working condition were determined.For the selected sensitive components,a true-scale stress monitoring experiment was conducted on the 1E3-SZ2 transmission tower.Based on the measured data,machine learning methods were used to evaluate the operating status of the transmission tower.The results show that in different working conditions,the machine learning methods can accurately evaluate the stress state of each member,quickly grasp the operating status in real time of the transmission tower,and issue warnings for potential dangers.This research has important theoretical significance and engineering value.
作者
张亮
裴浩威
牛凯
吴雨桐
鞠彦忠
白俊峰
ZHANG Liang;PEI Haowei;NIU Kai;WU Yutong;JU Yanzhong;BAI Junfeng(Economic and Technological Research Institute,State Grid Henan Electric Power Company,Zhengzhou 450052,China;School of Civil Engineering and Architecture,Northeast Electric Power University,Jilin 132012,China)
出处
《武汉大学学报(工学版)》
北大核心
2026年第1期156-166,共11页
Engineering Journal of Wuhan University
关键词
输电塔
状态评估
监测
机器学习
transmission tower
state assessment
monitoring
machine learning
