Oil/gas pipeline pipe-segment dividing method based on improved FPPC algorithm

Abstract

The invention discloses an oil / gas pipeline pipe-segment dividing method based on an improved FPPC algorithm. The method comprises the following steps: 1) obtaining oil / gas pipeline risk index featurevalues; 2) obtaining projection values of all pipe-segment samples in an oil / gas pipeline; 3) constructing a membership degree matrix and clustering centers of an objective function in a fuzzy clustering iteration algorithm after dimension reduction, calculating overlapping degrees among all classes, and then using a real-number coding-based accelerated genetic algorithm to calculate optimal projection direction values, an optimal membership degree matrix and optimal projection clustering centers at the time, when a projection index function value is smallest, according to the projection values which are of all the pipe-segment samples and are obtained in the step 2) ; 4) calculating category feature values of all the pipe-segment samples; 5) calculating evaluation index function values of FS<( )>, XB<( )> and SC<( )>; and 6) obtaining category feature values of all pipe-segment samples corresponding to an optimal clustering center number, and thus completing oil / gas pipeline pipe-segment dividing based on the improved FPPC algorithm. According to the method, pipe-segment dividing of oil / gas pipelines can be accurately realized, and adaptability is higher.

Description

technical field [0001] The invention belongs to the technical field of risk assessment of oil and gas pipelines, and relates to a method for dividing sections of oil and gas pipelines based on an improved FPPC algorithm. Background technique [0002] In recent years, with the rapid development of China's economy, people's consumption of energy is increasing day by day. Oil and natural gas are one of the main energy sources consumed in China, and the safety of oil and gas pipelines as transportation carriers has also received extensive attention. Most of the oil and gas pipelines are buried in the buried environment, and the areas they pass through are vast and complex, which will cause changes in the properties of the pipelines, which in turn will cause continuous changes in the risks along the pipelines. Therefore, once a leak occurs, it is difficult to locate and may be Trigger a chain reaction, causing great property damage and casualties. [0003] In the late 1990s, som...

AI Technical Summary

Problems solved by technology

[0005] The above methods have also achieved certain results, but there are still the following problems: (1) the artificial neural network has high requirements on the sample size, and it is easy to fall into local optimum; (2) the fuzzy clustering method and the projection pursuit clustering algorithm are subjective (3) Fuzzy projection pursuit clustering only considers intra-class closeness and inter-class separation, and it is difficult to identify small or low-density classes in the pipeline sample set. Applicability Not strong

Images