Research on Monitoring Data Anomaly Recognition Algorithm Based on Time Series Compression and Segmentation

被引：0
|
作者：
Pu, Qianhui [1 ]
Zhang, Ziyi [1 ]
Xiao, Tugang [1 ]
Hong, Yu [1 ]
Wen, Xuguang [2 ,3 ]
机构：
[1] School of Civil Engineering, Southwest Jiaotong University, Chengdu
[2] Guangxi Key Lahoratory of International Join for China-ASEAN Comprehensive Transportation, NanNing University, Nanning
[3] NanNing University, Nanning
来源：
Bridge Construction | 2024年 / 54卷 / 03期
关键词：
anomaly detection; cable-stayed bridge; Euclidcan distance; health monitoring data; local outlier factor; LOF algorithm; PLR SIP algorithm; time series;
D O I：
10.20051/j.issn.1003-4722.2024.03.003
中图分类号：
学科分类号：
摘要：
To cffccüvcly idcnüfy anomalics in bridge monitoring data and reduce the occurrence of falsc and missed alarms, and ensure thc quality and cffectiveness of bridge monitoring data as well, a monitoring data anomaly recognition algorithm based on time series compression and segmentation is proposed, counting on the anomalies (missing, outlier, and drifting)in the long-term monitoring data of long-span cable-stayed bridges. In this algorithm, the original monitoring data time series is segmented into multiple shorter time series using thc pieeewise linear regression with spectral information preservation algorithm based on important points(PLRSIP). Then, the similarity analysis of time series is performed using Euclidean distance, and the local outlier factor(LOF) algorithm is used to calculate the local outlier factor of cach time series. Finally, thc anomalies in the monitoring data are identified by comparing them with the set thresholds. The proposed algorithm has been applied to detect anomalies in the health monitoring data of an existing long-span highway bridge for verification. It is shown that the time sub-series can accurately reflect the trend and ränge of original series by using PLR _ SIP algorithm to compress and segment thc original ümc scrics. The improved LOF algorithm breaks through the limitations of thc traditional LOF algorithm which can only recognize outlicrs without duration. 1t can eliminate the interference of noise and realize the recognition of outliers, missing data and data drifting. The algorithm directly takes the original monitoring data as the input of the algorithm, with no need of defining the training sets, and can adaptively adjust the threshold Parameters. With sound scalability, real-time Performance, aecuraey, and efficiency, thc algorithm is capablc of processing real-time and mass bridge health monitoring data. © 2024 Wuhan Bridge Media Co., Ltd., MBEC. All rights reserved.
引用
收藏
页码：15 / 23
页数：8
相关论文
共 18 条
  • [1] YUE Jiali, HAO Jing, LU Hailin, Outliers Processing Method of Bridge Structure Monitoring Data, Journal of Wuhan Institute of Technology, 44, 1, pp. 107-111, (2022)
  • [2] YUEN K V, ORTIZ G A., Outlier Detection and Rohust Regression for Correlated Data, Computer Methods in Applied Mechanics and Engineering, 313, 2, pp. 632-646, (2017)
  • [3] BAO Y Q, TANG Z Y, LI H, Computer Vision and Deep Learning-Based Data Anomaly Detection Method for Structural Health Monitoring, Structural Health Monitoring, 18, 2, pp. 401-421, (2019)
  • [4] TANG Zhiyi, Data Anomaly Diagnosis and Reconstruction Based on Deep Learning for Structural Health Monitoring, (2021)
  • [5] WANG Wei, Early-Warning Method of Bridges Anomaly Monitoring Data Based on Time-Frequency Analysis, (2020)
  • [6] MAO J X, WANGH, SPENCER B F., Toward Data Anomaly Detection for Automated Structural Health Monitoring: Exploiting Generative Adversarial Nets and Autoencoders[J], Structural Health Monitoring, 20, 4, pp. 1609-1626, (2021)
  • [7] PANG Zhiyuan, Study of Intelligent Anomaly Detection and Repair Method for Bridge Acceleration Monitoring Data, (2022)
  • [8] YANG Na, FU Yingyu, LI Tianhao, Data Anomaly Identification Method Based on Local Outlier Factor and Application in Monitoring Data of Heritage Building Structure, Journal of Building Structures, 43, 10, pp. 68-75, (2022)
  • [9] LIU G, LI L L, ZHANG L L, Sensor Faults Classification for SHM Systems Using Deep Learning-Based Method with Tsfresh Features, Smart Materials and Structures, 29, 7, (2020)
  • [10] CHEN Zhen, Research on Anomaly Detection and Data Quality Assessment of Bridge Health Monitoring Data[D], (2017)
12