A wave change analysis (WCA) method for pipeline leak detection using Gaussian mixture model

This paper presents a novel pipeline leak detection scheme based on gradient and slope turns rejection (GSTR). Instead of monitoring the pipeline under constant working pressure, GSTR introduces a new testing method which obtains data during the transient periods of different working pressures. A novel pipeline leak detection method based on those transient data without failure history is proposed. Wavelet packet analysis (WPA) is applied to extract features which capture the dynamic characteristics from the non-stationary pressure data. Principal component analysis (PCA) is used to reduce the dimension of the feature space. Gaussian mixture model (GMM) is utilized to approximate the density distribution of the lower-dimensional feature space which consists of the major principal components. Bayesian information criterion (BIC) is used to determine the number of mixtures for the GMM and a density boosting method is applied to achieve better accuracy of the distribution estimation. An experimental case study for oil pipeline system is used as an example to validate the effectiveness of the proposed method.

► Gradient and slope turns rejection (GSTR) is presented to detect the oil pipeline leaks.
► Wavelet packet analysis is applied to extract features.
► Principal component analysis is used to reduce the dimension of the feature space.
► Gaussian mixture model works well in classifying the density distribution of the open/shut valves, startup/shut pumps and leak features.