Abstract:Structural health monitoring (SHM) has become an effective technology for damage diagnosis and operational safety assurance. During the long-term monitoring process, data missing is common occurrence owing to sensor fault, power depletion, data transmission failure, etc. Recovery of the missing data contribute to the integrity and reliability of the SHM data. In this study, a missing data recovery method based on the probabilistic principal component analysis was introduced for SHM. This method does not require the complete training data in advance. It is especially suitable for the case with a limited number of complete sets of data and missing data in multiple sensors. In addition, the PPCA can estimate the noise level of SHM data, thus the confidence interval of the recovered data can be obtained. The SHM data of a revolving auditorium was used for validating the proposed method and comparing with the traditional principal component analysis, multiple linear regression, K nearest neighbor-based data recovery methods, and compressive sensing method. The result shows that the performance of the proposed method is the best among four methods under various cases with data missing type and data missing ratios.
马帜,罗尧治,万华平,YUN C B,沈雁彬,俞峰. 基于概率主成分分析的结构健康监测数据修复方法研究[J]. 振动与冲击, 2021, 40(21): 135-141.
MA Zhi, LUO Yaozhi, WAN Huaping, YUN C B, SHEN Yanbin, YU Feng. Repair method of structural health monitoring data based on probabilistic principal component analysis. JOURNAL OF VIBRATION AND SHOCK, 2021, 40(21): 135-141.
[1] 罗尧治, 梅宇佳, 沈雁彬, 等. 国家体育场钢结构温度与应力实测及分析[J]. 建筑结构学报, 2013,34(11):24-32.
LUO Yaozhi, MEI Yujia, SHEN Yanbin, et al. Field measurement of temperature and stress on steel structure of the National Stadium and analysis of temperature action[J]. Journal of Building Structures, 2013,34(11):24-32.
[2] 周毅, 孙利民,闵志华, 斜拉桥主梁应变监测数据分析[J]. 振动与冲击, 2011. 30(04): 230-235.
ZHOU Yi, SUN Li-min, MIN Zhi-hua, et al. Girder strain analysis of a cable stayed bridge[J]. Journal of Vibration and Shock, 2011. 30(04): 230-235.
[3] 李惠, 周峰, 朱焰煌, 等. 国家游泳中心钢结构施工卸载过程及运营期间应变健康监测及计算模拟分析[J]. 土木工程学报, 2012,45(03):1-9.
LI Hui, ZHOU Feng, ZHU Yan-huang, et al. An analysis of monitored and computed strain of the National Aquatics Center in the states of unloading and daily use[J]. China Civil Engineering Journal, 2012,45(03):1-9.
[4] 陈伟欢,吕中荣,陈树辉等. 广州新电视塔不同激励下动力特性监测[J]. 振动与冲击, 2012. 3(31): 49-54.
CHEN Wei-huan, LU Zhong-rong, CHEN Shu-hui, et al. Monitoring dynamic characteristics of Canton Tower under different excitation [J]. Journal of Vibration and Shock, 2012. 3(31): 49-54.
[5] KO J M, NI Y Q. Technology developments in structural health monitoring of large-scale bridges[J]. Engineering Structures, 2005,27(12):1715-1725.
[6] OU J, LI H. Structural Health Monitoring in mainland China: Review and Future Trends[J]. Structural Health Monitoring, 2010,9(3):219-231.
[7] BAO Y, LI H, SUN X, et al. Compressive sampling–based data loss recovery for wireless sensor networks used in civil structural health monitoring[J]. Structural Health Monitoring, 2013: 12, 78-95.
[8] NI Y Q, LI M. Wind pressure data reconstruction using neural network techniques: A comparison between BPNN and GRNN[J]. Measurement, 2016,88:468-476.
[9] 谢晓凯, 罗尧治, 张楠, 等. 基于神经网络的大跨度空间钢结构应力实测缺失数据修复方法研究[J]. 空间结构, 2019,25(03):38-44.
XIE Xiao-kai, LUO Yao-zhi, ZHANG Nan, et al. Missing data reconstruction in stress monitoring of steel spatial structures using neural network technique[J]. Spatial Structures, 2019,25(03):38-44.
[10] ZHANG Z, LUO Y. Restoring method for missing data of spatial structural stress monitoring based on correlation[J]. Mechanical Systems and Signal Processing, 2017,91:266-277.
[11] WAN H, NI Y. Bayesian multi-task learning methodology for reconstruction of structural health monitoring data[J]. Structural Health Monitoring, 2018:84280249.
[12] TIPPING M E, BISHOP C M. Probabilistic Principal Component Analysis[J]. Journal of the Royal Statistical Society, 1999,61(3):611-622.
[13] JOLLIFFE I T. Principal component analysis[M]. Springer Series in Statistics, Berlin: Springer, 1986.
[14] YU L, SNAPP R R, RUIZ T, et al. Probabilistic principal component analysis with expectation maximization (PPCA-EM) facilitates volume classification and estimates the missing data[J]. Journal of Structural Biology, 2010,171(1):18-30.
[15] LUO Y, YANG P, SHEN Y, et al. Development of a dynamic sensing system for civil revolving structures and its field tests in a large revolving auditorium[J]. Smart Structures and Systems, 2014,13(6):993-1014.
[16] SHEN Y, YANG P, ZHANG P, et al. Development of a Multitype Wireless Sensor Network for the Large-Scale Structure of the National Stadium in China[J]. International Journal of Distributed Sensor Networks, 2013,2013:1-16.
[17] WANG Y C, LUO Y Z, SUN B, et al. Field measurement system based on a wireless sensor network for the wind load on spatial structures: Design, experimental, and field validation[J]. Structural Control and Health Monitoring, 2018,25(9): e2192.
[18] 王凤梅, 胡丽霞. 一种基于近邻规则的缺失数据填补方法[J]. 计算机工程, 2012,38(21):53-55.
WANG Feng-mei, HU Li-xia. Missing Data Imputation Method Based on Neighbor Rules[J]. Computer Engineering, 2012,38(21):53-55.