Tool Performance Estimation Considering the Effect of Fixed vs. Variable Slope

Abstract

The performance of in-line inspection (ILI) measurements is one of the primary sources of uncertainty for corrosion and crack assessments. Operators typically perform validation digs to determine how well the ILI tool performed at sizing anomalies. The API Standard 1163, “In-line Inspection Systems Qualification”, defines three levels of analysis that operators use to validate the ILI sizing accuracy. The first level relies on previous experience with the ILI system and adherence to operating procedures, the second level uses validation dig data to determine if the vendor specification can be used, and the third level uses advanced statistical methods to estimate performance directly from validation dig data. The recently published third edition of API 1163 (2021) provides two example methodologies to perform a Level 3 analysis. Both methods make simplifying assumptions to reduce the complexity of the problem, but one of the key differences between them is that the first method assumes the slope of the best-fit line on a unity plot is fixed at 1.0. In this paper, the effect of fixing the slope at 1.0 to estimate tool performance is investigated and compared against a method where slope is allowed to be variable. Several realistic example datasets are evaluated using both methods and the results are compared. The effect on calculated burst pressure for representative corrosion anomalies is also investigated for both methods.