Corrosion under insulation (CUI) is a severe problem of many multi-layer equipment operating in the oil sector. The evolution of the damage mechanism usually stays invisible until leakage or a catastrophic failure occurs. According to consolidated recommended practices, the trapped humidity inside the insulation system is the major responsible for the development of the corrosion process and the high process temperature increases the probability of equipment failure. Methods to reliably detect trapped water in the internal layers of the insulation systems is a key factor to contribute to more efficient corrosion mitigation plans.As the presence of water changes the effective thermal properties of the insulation layer, locally trapped humidity generates thermal contrasts that can be detectable by infrared thermography. The heating source provided by the process fluids generates abnormal thermal gradients or hot spots resulting from variation of the local thermal resistance and heat capacity. However, the low emissivity of the aluminum cladding makes more difficult to detect – and to differentiate from reflection artifacts – the abnormal thermal patterns produced by the humidity trapped in the insulation, reducing then the sensitivity of the inspection and prompting false positives to the analysis. A more robust and quantitative method is then needed to efficiently discriminate actual defect indications from noise.In this regard, the Partial Least-Squares Thermography (PLST) is applied to the acquired thermal data to increase defect thermal contrast and to reduce the noise level. Based on Partial Least-Squares regression, PLST has the capability to separate the information from distinct physical phenomena in different latent variables (called components) without losing the time tracking. Ambient interactions, process variations and the water spreading itself inside the insulation can act as variance sources to the mentioned processing method. The results from the decomposition of the original thermal data are then reconstructed in thermal maps in which the temperature data is linked to time variance changes.An oil conducting insulated pipe, containing inserts of wet insulation, was monitored by an infrared camera during ambient and process fluid temperature variations. The acquired thermal sequence was then divided into partial datasets referents to each test period, process temperature risings and ambient cooling, to be analyzed by PLST.We compare and report the PLST method achieved higher detection performance than other consolidated thermographic processing algorithms with comparable noise filtering when applied to the same datasets. The PLST also presents the advantage to characterize individually each physical phenomenon involved in the inspection process, helping to understand the thermal behaviors of all indications’ sources.

Humidity Detection in Multi-Layer Components by Partial Least-Squares Thermography

Paper Proceedings: https://asmedigitalcollection.asme.org/qnde/QNDE2021/volume/85529

ASME 2021 48th Annual Review of Progress in Quantitative Nondestructive Evaluation (QNDE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.