摘要
为避免大跨度钢结构健康监测中数据缺失的问题,保证监测数据的准确性和完整性,本文在研究监测点的温度与应力相关性基础上,提出改进的线性回归模型,重构缺失的应力数据,使缺失的数据达到较高的拟合精度和最佳的重构效果,并对不同缺失率的数据重构效果进行分析。结果表明:当数据缺失率控制在30%以内,相关系数均达到0.9以上,重构应力缺失数据与实测数据的平均误差小于5%。该方法对钢结构应力缺失数据的重构有着较高的精度。
In order to avoid the problem of missing data in the health monitoring of long-span steel structures and ensure the accuracy and integrity of the monitoring data,this paper proposes an improved linear regression model based on the study of the correlation between temperature and stress at the monitoring points to reconstruct the missing stress data.With the proposed model,the missing data achieved higher fitting accuracy and the best reconstruction effect,and the reconstruction effect of data with different missing rates was analyzed.The results showed that when the data missing rate was controlled within 30%,the correlation coefficients were all above 0.9,and the average error between the reconstructed stress missing data and the measured data was less than 5%.This method had a high accuracy for the reconstruction of steel structure stress missing data.
作者
游颖
王建
刘游谦
周清富
彭宁
YOU Ying;WANG Jian;LIU Youqian;ZHOU Qingfu;PENG Ning(School of Mechanical Engineering,Hubei Univ.of Tech.,Wuhan 430068,China;School of Engineering and Technology,Hubei Univ.,of Tech.,Wuhan 430068,China)
出处
《湖北工业大学学报》
2023年第4期44-48,54,共6页
Journal of Hubei University of Technology
基金
国家自然科学基金(52105550)
湖北省自然科学基金(2013CFB025)。
关键词
空间钢结构
改进线性回归法
数据重构
健康检测
space steel structure
improved linear regression method
data reconstruction
health check
