基于K均值聚类的油气管道漏磁缺陷标记方法

为了提高漏磁数据缺陷区域标记能力,将聚类算法应用于漏磁检测数据分析中,提出了一种基于K均值聚类的管道漏磁缺陷信号标记方法,并进行了不同口径和不同壁厚管道检测试验验证。结果表明:该方法可有效识别出漏磁数据中的缺陷区域,识别准确度满足工程要求。由于该方法无需根据检测器和管道情况单独设置阈值,因此其具有较广泛的适应性。

Defect Marking Method of Magnetic Flux Leakage in Oil and Gas Pipelines based on K-Means Clustering

In order to improve the marking ability of magnetic flux leakage data, the clustering algorithm is applied to the analysis of magnetic flux leakage detection data. A method for marking the magnetic flux leakage defects of pipelines based on K-means clustering is proposed. The test of different caliber and wall thickness pipeline tests was carried out. The results show that the method can effectively identify the defect area in the magnetic flux leakage data, and the recognition accuracy meets the engineering requirements. Since the method does not need to set the threshold separately according to the detector and the pipeline condition, the method has a wide range of adaptability.