Grant: Alliance Grants

Dents are a common type of mechanical damage on buried oil and gas pipelines and result from excavation activities near the pipeline right of way and rock impact. A dent is defined as a depression that produces a gross disturbance in the curvature of the pipe wall. Many codes and standards such as ASME B31.8 and CSA Z662 have specified the strain-based assessment criteria for dents, which typically require a dent with the maximum strain exceeding a critical value to be repaired. To compute strains in the indented area of a pipeline requires the detailed deformation profile of the dent, which is measured using the inline caliper tool in practice. Due to fluctuations in the tool speed and equipment wear, the obtained dent signals inevitably contain noises. Such noises could markedly impact the accuracy of dent strains evaluated using the caliper signal and subsequent decision-making for mitigation actions. To minimize the influence of the noise, it is essential to denoise the measured dent signal. There is currently a lack of a clear understanding of the effectiveness and applicability of different denoising methods, and as a result, a lack of quantitative knowledge of the accuracy of dent strains evaluated based on denoised signals. This knowledge gap leads to poor maintenance decisions, e.g. critical dents unidentified or non-critical dents repaired unnecessarily. Furthermore, there is inherent measurement error associated with the caliper tool. It is currently unclear if and how the magnitude of the measurement error is related to the basic dent characteristics. An effective framework to incorporate the measurement error in the dent assessment in practice is unavailable. The proposed project is aimed at addressing the above-indicated challenges associated with the strain-based dent assessment in the pipeline industry. The project will last one year and train one PhD student and two MESc students.