Threshold Value Classification of Bridge Health Monitoring
The classification of the threshold values (CTV) in bridge health monitoring system plays an important role for the bridge assessment. Because of the lack of classification standards, the bridge monitoring system sometimes gives incorrect alarm information. However, CTV is so a complicated problem which refers to the measured data accumulation, the theory of dynamics, signal analysis and processing, parameter identification etc. that there is few theory about the issue. Fisher optimal division (FOD), as an effective mathematical method to decompose an ordered sequence, is used in this paper to solve the CTV of the Nanjing Yangtze River Bridge (NYRB). Firstly, by ANSYS software, the finite element model of NYRB for the simple-supported span is built and then stiffness of parts of members is reduced. Secondly, a series of data obtained from the model are sorted and then form an ordered sequence. Finally, by using MATLAB as a platform, programs are developed to obtain classification information and the data is classified for three clusters: the normal value, the abnormal value and the alarm value.
Threshold values Classification Fisher optimal division Bridge health monitoring Assessment system
Qiufen Wang Fanglin Huang Yingnan Gao
School of Civil Engineering and Architecture, Central South University, Changsha 410075 China
国际会议
南京
英文
337-341
2007-10-16（万方平台首次上网日期，不代表论文的发表时间）