大数据聚类算法在成品油输油管道泄漏检测中的应用

Application of Big Data Clustering Algorithm in Leakage Detection of Oil Product Pipeline

针对石油成品油输油管道泄漏检测易出现持续性重复报警和大量误报的问题,提出采用大数据聚类算法,建立成品油管道聚类算法模型。以湖南成品油输油管道为例,通过大数据模拟不同工况下离群系数的变化规律,对离群系数阈值进行总结,设立不同工况的阈值,提高了石油成品油输油管道泄漏检测的准确度。

In order to solve the problems of continuous repeated alarm and a large number of false alarms in leakage detection of oil product pipeline,a clustering algorithm model of oil product pipeline based on big data clustering algorithm is proposed.Taking Hunan oil product pipeline as an example,through simulating the changing law of outlier coefficient under different working conditions using big data,the outlier coefficient threshold is summarized,and the threshold under different working conditions is set up to improve the accuracy of oil product pipeline leakage detection.