Innovative prognostic methodology for pipe defect detection leveraging acoustic emissions analysis and computational modeling integration

Abstract

Pipelines play a crucial role in transporting essential resources such as water, oil, and gas across industrial, urban, and environmental infrastructures. Clogging remains a persistent challenge, potentially resulting in catastrophic failures, operational disruptions, increased maintenance costs, and serious safety risks. This study presents a novel prognostic and health monitoring approach that utilizes bubble-induced acoustic emissions to detect and characterize pipeline blockages. An analytical model is developed to capture the acoustic signatures of detaching bubbles, revealing features highly sensitive to clogging. Finite element simulations using Abaqus further investigate how different clogging conditions affect acoustic wave propagation. These insights drive the development of a machine learning-based predictive maintenance strategy, validated on real-world datasets. The results demonstrate exceptional accuracy, with most classifiers achieving 100% detection rates for clogging presence, shape, and severity. Additionally, model generalization tests show that machine learning algorithms adapt more effectively to varying clogging thickness than clogging shape. This research paves the way for a highly accurate, non-destructive monitoring solution, enhancing predictive maintenance and ensuring the reliability of industrial pipelines.