Algorithm and Application of Pipeline In-Line Inspection Data Alignment

Abstract

Abstract With the publication of GB 32167, pipeline in-line inspection has been executed comprehensively in China, which causes many problems such as mismatching, missing and misreporting of pipeline features, commanding inspection level of different providers to improve the inspection quality, data mining of multiple rounds of inspections and so on. In-line inspection data alignment can solve these problems efficiently. In this paper, a series of in-line inspection data automatic alignment experiences and rules are put forward, which can find out the error of in-line inspection data initially, realize the alignment of above-ground and underground features of pipeline, search and locate the combined defects of one inspection data, and match defects of different rounds of inspections. Thus problems such as locating pipeline defects and segments above ground, judging and analysis of different kinds of combined defects, quality verification of inspection data from different service providers, knowing the evolution of defects for pipeline, etc. can be solved. The methodology is applied to one certain product oil pipeline in China and the following results are obtained. Two ILIs data alignment rates of bends, girth wields and defects are 100%, 99.05% and 28.6%. The result shows that 2 dents intersecting the metal loss in 2016 ILI and 4 metal losses in the heat affected zone of girth weld in 2021 ILI are unreported. Reasons for low defects alignment rate are proposed. The results demonstrate problems such as misreporting and missing of combined defects, bends, girth welds and other features, as well as quality problems and logical errors existed in two rounds of ILIs data. The relevant advices were given to the pipeline enterprise finally.