Prediction method of pipeline corrosion defect size based on multi-source data fusion

Abstract

A prediction method for pipeline corrosion defect size based on multi-source data fusion. This method is implemented using a pipeline corrosion defect size prediction system based on multi-source data fusion. The system includes: sensor group, lower computer and upper computer; among them, the sensor group is at the head or end of the pipeline and in contact with the transport medium; the sensor group is connected to the lower computer, and the lower computer is connected to the upper computer; this method first uses historical data to establish a gray GM (1, 1) prediction model and BP neural network model, and then use these two models to obtain the predicted values. Next, based on the historical data and the predicted values ​​​​of these two models, an induced ordered weighted harmonic average IOWHA operator combination model is established. For new data, these three methods are used successively. Model, after predicting the size of corrosion defects in three directions: axial, circumferential and depth, finally obtains the predicted value of pipeline corrosion defect size; the present invention has higher prediction accuracy and reduces the amount of data input.

Description

technical field [0001] The invention belongs to the technical field of pipeline risk prediction, and in particular relates to a pipeline corrosion defect size prediction method based on multi-source data fusion. Background technique [0002] The role of pipeline transportation in economic development is becoming more and more important, such as urban tap water pipelines, land crude oil pipelines, and submarine oil and gas pipelines. One of the important safety hazards of pipeline transportation. In order to prevent leakage accidents, it is necessary to use pipeline detection equipment to detect the pipeline, understand the corrosion status of the pipeline, evaluate the remaining strength of the pipeline, and predict the remaining life of the pipeline, so as to maintain and repair the pipeline in a planned and targeted manner. . [0003] There are many factors for the occurrence of pipeline corrosion. The internal factors are the difference in the composition of the pipe wa...

AI Technical Summary

Problems solved by technology

At present, the methods for determining the change trend of corrosion defect size with time mainly include: online real-time detection through pipeline detectors to obtain the current corrosion defect size, using prediction algorithms such as gray GM (1, 1), based on time series, through Calculation and analysis to obtain the change of corrosion defect size with time, this method requires less data, but the prediction accuracy is not high

In addition, it is also possible to analyze the environment of the pipeline, use the knowledge of chemical disciplines to establish a specific pipeline corrosion rate model, and solve the corrosion defect size change with time. This method is more accurate than the gray GM (1, 1) method. High, but the amount of required data information is large, and it is difficult to obtain a sufficient amount of detection data, which also reduces the accuracy of this method to a certain extent

Images