Uncertainty-quantified unsupervised transfer learning for ultrasonic-based corrosion detection in underwater steel pipes
Lu, H.; Haris, M.; Chinchilla, S.; Fang, C.; Soccol, D.; Gryllias, K.; Chronopoulos, D. (2025). Uncertainty-quantified unsupervised transfer learning for ultrasonic-based corrosion detection in underwater steel pipes. Automation in Construction 178: 106385. https://dx.doi.org/10.1016/j.autcon.2025.106385
In: Automation in Construction. Elsevier. ISSN 0926-5805; e-ISSN 1872-7891, meer
|  |
| Author keywords |
Unsupervised transfer learning;Uncertainty quantification;Automation;Underwater ultrasonic wave;Underwater corrosion;Marine structural health monitoring;transport |
| Auteurs | | Top |
- Lu, H.
- Haris, M.
- Chinchilla, S.
- Fang, C.
|
- Soccol, D.
- Gryllias, K., meer
- Chronopoulos, D.
|
|
| Abstract |
Automation-ready subsea and coastal asset inspection requires reliable, low-touch structural health monitoring, critical for digitalized underwater pipeline management. This paper introduces a deep-ensemble unsupervised transfer-learning (DE-UTL) framework to automatically quantify corrosion-induced thickness loss from ultrasonic signals while quantifying aleatoric and epistemic uncertainties. DE-UTL employs an enhanced transformer architecture with adaptive weighting and multi-scale feature extraction, integrated with a Wasserstein-distance domain adapter supported by consistency regularization and reconstruction, ensuring environmental transferability and mitigating retraining needs. Ten UTL models are integrated via a deep ensemble to evaluate uncertainty. Underwater corrosion and ultrasonic experiments on two pipe types are conducted, revealing that pipe thickness, weight, and ultrasonic signal amplitude and phase exhibit trend-like variation with corrosion severity. Experimental validation confirms DE-UTL's effective corrosion thickness quantification in both geometry-only and geometry-environmental (air-to-water) transfer scenarios. Integrating corrosion experimentation, unsupervised transfer learning, and uncertainty quantification, DE-UTL shows strong potential for automated, scalable, real-time underwater asset inspection. |
|
