Tracking the progress of corrosion is a key part of ensuring safety and planning maintenance in the petrochemical industry. For the vast majority of engineering components, inspections are performed using traditional non-destructive testing (NDT) techniques, over the entire component. More often than not, plant operations must be shutdown to allow these inspections to occur. This proves expensive, as the plant operator incurs a loss of revenue on top of the cost of the inspection. Consequently, there has been a drive in industry to reduce the cost of these inspections and the time required to make an assessment.
The picture below shows typical C-scan data represented as a colour map and empirical cumulative distribution function (ECDF). Pits in the C-scan show up as a tail in the ECDF, by systematically adjusting the threshold that is applied to the ECDF the pits can be separated out from the general background variation in thickness on the C-scan. This project focuses on partial coverage inspection (PCI) as a potential to reduce the overall cost of inspection. In a typical PCI an inspector would inspect a fraction of the area of the component. The data collected from this area can then be analysed within a statistical framework to extrapolate to the condition of the entire component. We have been investigating the effects of measurement errors on the results, the circumstances in which this approach is suitable and are developing a standardized methodology for PCI data analysis.
References
Benstock, Daniel, and Frederic Cegla. "The effect of surface roughness on extrapolation from thickness C-scan data using extreme value theory." In 41ST ANNUAL REVIEW OF PROGRESS IN QUANTITATIVE NONDESTRUCTIVE EVALUATION: Volume 34, vol. 1650, pp. 1677-1687. AIP Publishing, 2015. (web: http://scitation.aip.org/content/aip/proceeding/aipcp/10.1063/1.4914789)
Benstock, Daniel, Frederic Cegla, and Mark Stone. "The influence of surface roughness on ultrasonic thickness measurements." The Journal of the Acoustical Society of America 136, no. 6 (2014): 3028-3039. (web: http://scitation.aip.org/content/asa/journal/jasa/136/6/10.1121/1.4900565)