An adaptive PCA‐based method for more reliable ultrasonic guided waves SHM: Data‐driven modeling and experimental validation in high attenuating medium

Abstract

This paper proposes a monitoring method for defect detection and localization in a structure operating under environmental and operational conditions (EOCs) variation. This method is based on a model obtained using PCA of the healthy state data. To account for variation in EOCs, more particularly temperature change, the model is updated using a moving window over the collected signals. Defect detection was performed by calculating the squared prediction error between the current measured signal and its estimation predicted by the model. Once a defect is detected, its localization is performed by applying the PCA-based model on a sliding window over the signal. The test and validation of the proposed method were achieved on two databases collected from two pipeline segments where each one has high attenuating composite reparation. The first database was built under small temperature variation, whereas the second one was under a relatively high temperature variation. To simulate a minor corrosion, a relatively small defect was created by removing material inside each pipeline segment. To mimic its time evolution in real world, this corrosion-like defect was grown in different steps. The data were collected using ultrasonic guided waves (UGW) technique. Despite the high attenuation caused by the composite reparation, the specific placement of the defect, and its small size, it was successfully detected and localized. The proposed method for defect detection is not limited to UGW and could be applied to any active SHM technique that provides time signals.