Natural gas pipeline corrosion defect typical characteristic extraction method

Abstract

The invention discloses a natural gas pipeline corrosion defect typical characteristic extraction method, and belongs to the technical field of oil and gas pipelines. The method comprises the steps: collecting defect-containing pipeline design parameters and conveying medium information; performing pre-classification processing on the corrosive defect data; performing curve fitting on the pre-classified data on the basis of the least square method and the distribution hypothesis to obtain the optimal distribution type and the probability density function of the defect size variable obeying; and carrying out clustering analysis on the defect data by adopting a clustering algorithm improved based on statistics and an outlier detection method and a roulette algorithm thought so as to obtain the defect typical size and distribution characteristic value. According to the method, the initial clustering center selection mode is improved, the adverse effect of the outlier value on clustering analysis is reduced, and a scientific and systematic method is provided for extracting the corrosion defect distribution rule, the typical size and the circumferential distribution characteristics froma large amount of detection data.

Description

technical field [0001] The invention belongs to the technical field of oil and gas pipelines, in particular to a method for extracting typical features of corrosion defects in natural gas pipelines. Background technique [0002] In the oil and gas pipeline industry, corrosion is one of the main factors that threaten the safety of natural gas pipelines. It can lead to the formation of a large number of corrosion defects such as thinning and cracks on the surface of the pipeline, burying hidden dangers for the safe operation of oil and gas pipelines. Regularly inspecting and evaluating the metal surface of oil and gas pipelines, and formulating reasonable repair measures, has become an important technical means for the oil and gas pipeline industry to deal with the threat of pipeline corrosion and reduce the risk of corrosion failure. [0003] Defect detection of metal pipelines usually produces a large amount of detection data, including information such as types and sizes of...

AI Technical Summary

Problems solved by technology

[0005] The present invention aims to solve the problems that cluster analysis is susceptible to interference from outliers and initial cluster centers, and the existing detection data processing methods cannot efficiently and reasonably extract typical characteristics of pipeline corrosion defects, etc., and proposes a method that considers the characteristics of pipe materials and corrosion environments , and based on the statistical outlier detection method and the improved clustering algorithm of the roulette algorithm, the method of extracting typical characteristics of corrosion defects from a large number of detection data

Images