Deep learning-based data anomaly detection for highway slope structural health monitoring: A comparative study

作     者：Dong, Shi Long, Zhiyou Zhang, Shiyuan Wang, Jianwei Zuo, Chen Yang, Chao Jiang, Jinyi Cui, Zhiwei Wan, Zhaolong 

作者机构：Changan Univ Sch Transportat Engn Xian 710064 Peoples R China Changan Univ Engn Res Ctr Highway Infrastruct Digitalizat Minist Educ PRC Xian 710064 Peoples R China Zhejiang Sci Res Inst Transport Hangzhou 310023 Peoples R China Shaanxi Expressway Engn Testing Inspect & Testing Xian 710086 Peoples R China Zhejiang Inst Commun Zhejiang Engn Res Ctr Digital Highway Appl Technol Hangzhou 311112 Peoples R China 

出 版 物：《TRANSPORTATION GEOTECHNICS》 (Transp. Geotech.)

年 卷 期：2025年第51卷

核心收录：

学科分类：08[工学] 0818[工学-地质资源与地质工程] 0814[工学-土木工程] 0823[工学-交通运输工程] 

基　　金：National Natural Science Foundation of China National Key Research and Development Program of China [2020YFC1512000] Special Funds for Central Administration Guiding Local Science and Technology Development of Shaanxi Province, China [2024GX-YBXM-528] The 2023 Transportation Research Project, Shaanxi Provincial Department of Transportation, China [23-04X] IOT Technology Application Transportation Industry R & D Center, Hangzhou, China [2023-03] The 2022 Transportation Research Project, Shaanxi Provincial Department of Transportation, China [22-01X] 

主　　题：Slope structural health monitoring (SHM) Big data Data anomaly detection Deep learning Comparative study 

摘      要：Highway slope instability has a significant influence on traffic safety. However, there are many anomalies in slope SHM data, which is critical to timely warnings and safety assessments of slopes. In this paper, we carried out a multi-case comparative study of deep learning models to examine the recognition accuracy of anomalous data. First, our program collected the monitoring data from the Baihe and Lueyang slopes of the Shi-Tian expressway in Shaanxi province. Six categories of abnormal data were found using K-means clustering. Second, based on the time response images, the frequency response and Gramian Angular Field (GAF) images were superimposed to improve the model s ability to identify six types of data anomalies. Third, we employed combo loss to tackle the data imbalance problem by incorporating dice loss and focal loss. Last, we conducted a comparative study of nine deep learning models to investigate the anomaly detection capability with the combo loss function. The results indicated that a combination of frequency response and GAF images can effectively improve the identification of abnormal data. The combo loss with equal weights significantly improved specific anomaly categories performance. In addition, the performance of ResNet50, EfficientNetB1, and Vision Mamba exhibited impressive classification accuracy, generalization ability, and computational efficiency. Our investigation has the potential to recognize slope SHM data abnormal types in various scenarios, further improving the precision of subsequent slope capability analysis.