Repair method of structural health monitoring data based on probabilistic principal component analysis

被引：0

作者：

Ma Z. ^{[1
]}

Luo Y. ^{[1
]}

Wan H. ^{[1
]}

Yun C.B. ^{[1
]}

Shen Y. ^{[1
]}

Yu F. ^{[1
]}

机构：

[1] School of Architecture Engineering, Zhejiang University, Hangzhou

来源：

Zhendong yu Chongji/Journal of Vibration and Shock | 2021年 / 40卷 / 21期

关键词：

Expectation maximization(EM) algorithm; Missing data recovery; Probabilistic principal component analysis(PPCA); Revolving auditorium; Structural health monitoring (SHM);

D O I：

10.13465/j.cnki.jvs.2021.21.019

中图分类号：

学科分类号：

摘要：

Structural health monitoring (SHM) has increasingly become an effective means to study structural damage evolution behavior and an important technology to ensure operational safety. In long-term monitoring process, due to monitoring equipment fault, energy supply interruption, data transmission fault and many other factors, lack of monitoring data is inevitable.Repairing missing data is helpful to ensure the integrity and reliability of monitoring data.Here, the probabilistic principal component analysis (PPCA) method was proposed to repair SHM data. PPCA method could not need to train the complete data, and was especially suitable for the situation of less complete data and missing data at multiple measuring points. PPCA could estimate the uncertainty level of repair data and give the corresponding confidence interval. The monitoring data of Wuyishan revolving auditorium were used to verify the effectiveness of the proposed method. Finally, PPCA method was compared with 4 data recovery methods of the traditional principal component analysis, the multi-variant linear regression method, K-nearest neighbor method and the compressed sensing method. The results showed that PPCA has the best repair effect under different missing conditions and different missing rates. © 2021, Editorial Office of Journal of Vibration and Shock. All right reserved.

引用

页码：135 / 141and167

共 18 条

[1]

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, Journal of Building Structures, 34, 11, pp. 24-32, (2013)

[2]

ZHOU Yi, SUN Limin, MIN Zhihua, Et al., Girder strain analysis of a cable stayed bridge, Journal of Vibration and Shock, 30, 4, pp. 230-235, (2011)

[3]

LI Hui, ZHOU Feng, ZHU Yanhuang, Et al., An analysis of monitored and computed strain of the National Aquatics Center in the states of unloading and daily use, China Civil Engineering Journal, 45, 3, pp. 1-9, (2012)

[4]

CHEN Weihuan, LU Zhongrong, CHEN Shuhui, Et al., Monitoring dynamic characteristics of Canton Tower under different excitation, Journal of Vibration and Shock, 31, 3, pp. 49-54, (2012)

[5]

KO J M, NI Y Q., Technology developments in structural health monitoring of large-scale bridges, Engineering Structures, 27, 12, pp. 1715-1725, (2005)

[6]

OU J, LI H., Structural health monitoring in mainland China: review and future trends[J], Structural Health Monitoring, 9, 3, pp. 219-231, (2010)

[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, Structural Health Monitoring, 12, pp. 78-95, (2013)

[8]

NI Y Q, LI M., Wind pressure data reconstruction using neural network techniques: a comparison between BPNN and GRNN[J], Measurement, 88, pp. 468-476, (2016)

[9]

XIE Xiaokai, LUO Yaozhi, ZHANG Nan, Et al., Missing data reconstruction in stress monitoring of steel spatial structures using neural network technique, Spatial Structures, 25, 3, pp. 38-44, (2019)

[10]

ZHANG Z, LUO Y., Restoring method for missing data of spatial structural stress monitoring based on correlation[J], Mechanical Systems and Signal Processing, 91, pp. 266-277, (2017)

← 1 2 →