Transferring Self-Supervised Pretrained Models for SHM Data Anomaly Detection With Scarce Labeled Data

Abstract

Structural health monitoring (SHM) has experienced significant advancements in recent decades, accumulating massive monitoring data. Data anomalies inevitably exist in monitoring data, posing significant challenges to their effective utilization. Recently, deep learning has emerged as an efficient and effective approach for anomaly detection in bridge SHM. Despite its progress, many deep learning models require large amounts of labeled data for training. The process of labeling data, however, is labor-intensive, time-consuming, and often impractical for large-scale SHM datasets. To address these challenges, this work explores the use of self-supervised learning (SSL), an emerging paradigm that employs unsupervised pretraining. The SSL-based framework aims to learn from only a very small quantity of labeled data by fine-tuning, while making the best use of the vast amount of unlabeled SHM data by pretraining. Basic and representative models from generative, contrastive, and generative–contrastive SSL categories are employed. These SSL models are compared and validated on the acceleration data of two in-service bridges, which is one of the most widely utilized types of measurements in SHM. Comparative analysis demonstrates that SSL techniques boost data anomaly detection performance, achieving increased F₁ scores compared to conventional supervised training, especially given a very limited amount of labeled data.