In situ monitoring of corrosion in steel pipes using acoustic emission

The structural integrity of critical infrastructure, such as pipelines, is continually subjected to risks from corrosion. However, traditional acoustic emission (AE) monitoring remains limited to periodic, operationally interfering inspections that require artificial overpressurization, a constraint incompatible with continuous, in-service assessment. This research addresses the fundamental question of whether AE can reliably detect and monitor active corrosion continuously, enabling true in-service assessment. To address this problem, a multi-scale methodology was developed, integrating long-term AE data acquisition from an active pipeline with complementary environmental and parametric sensors. To overcome background noise, a filtering protocol was employed to obtain a high-fidelity dataset of corrosion-related AE activity. A clear correlation was established between AE event rates and natural temperature cycles, confirming thermal gradients as an effective and non-disruptive source of mechanical stress to stimulate corrosion emissions. To validate this finding and establish causality, field AE signatures were systematically compared against ground-truth damage mechanisms from controlled laboratory tests. This validation framework confirms that continuous, in-service AE monitoring for corrosion is viable without operational overpressurization. Integrating passive environmental stimuli enables real-time structural health monitoring, fundamentally transforming infrastructure assessment from periodic inspections to continuous predictive evaluation.