Stationarity index based segmentation approach in leak locating systems

This paper presents a practical criterion for extracting piecewise stationary segments from pipeline leak signals, in order to better locate the leak on the pipe. The segmentation technique, that enables detecting and avoiding the non stationary abrupt, data changes, is based on computing the stationarity index (SI) function of the acquired noisy leak signals. A comparative study, involving different segmentation thresholds on real leak signals, captured in an experimental pipeline installation, indicate an optimum threshold. According to the proposed criterion, the signals acquired with pipeline´s end obstructed, practically needed no segmentation and gave the best estimates.