MULTI-SENSOR INFORMATION FUSION FOR STRUCTURAL DAMAGE DETECTION
In long-term structural health monitoring,a huge number of data are measured from multi-sensors. The different sets of data and different types of data may offer different results. How to interpret the data and how to integrate the results are critical to effective use of SHM system for structural condition assessment and damage detection. In this study,the Dempster-Shafer (D-S) evidence theory-based approach for structural damage detection is presented. First,the damage basic probability assignment function (BPA) using each data set measured from the monitored structure is calculated. Then,the D-S evidence theory is employed to combine the individual damage BPAs and obtain the final damage detection results. An experiment on a three story frame is investigated to demonstrate the effectiveness of the proposed technique with consideration of the model uncertainty and measurement noise. It shows that the damage detection results obtained by combining the damage BPAs from each tested data set are more accurate than those obtained from each test data separately.
Structural damage detection Information fusion Dempster-Shafer evidence theory Bayesian theory Uncertainty
Yue-Quan Bao Yong Xia Hui Li You-Lin Xu
Department of Civil and Structural Engineering,The Hong Kong Polytechnic University,Hong Kong,P.R. C Department of Civil and Structural Engineering,The Hong Kong Polytechnic University,Hong Kong,P.R.Ch School of Civil Engineering,Harbin Institute of Technology,Harbin 150090,P.R. China
国际会议
The Eleventh International Symposium on Structural Engineering(第十一届结构工程国际研讨会)
广州
英文
1648-1653
2010-12-01（万方平台首次上网日期，不代表论文的发表时间）