基于改进K均值聚类算法的燃气泄露检测研究

天然气管道会出现损坏现象导致天然气泄露,因此,快速准确地判断天然气的泄露具有重要意义。针对天然气泄露的检测问题,本文提出一种基于改进K均值聚类的检测方法。该方法在提取声发射信号特征的基础上,提出了基于数据点的邻点数目来选取初始聚类中心,并采用信息熵方法来确定聚类类别数目。实验结果表明,本文提出的方法较好了解决了原始K均值方法的问题,从而能准确地给出泄露检测结果。

The Research on Gas Leak Detection Based on Improved K-means Clustering Algorithm

The damage of natural gas pipeline will lead to natural gas leakage.Therefore,it is of great significance to quickly and accurately judge the leakage of natural gas.For the detection of natural gas leakage,a detection method based on the improved K-means clustering algorithm is proposed.We select the initial cluster centers based on the number of neighbors of the data point after the extraction features of acoustic emission signal. Forthermore,we determine the number of clusters with the method of information entropy.The experimental results show that the proposed method avoids the problem brought by original K-means and achieve the better leak detection result.