Trend Detection and Adaptive Removal in Pressure Signal for Pipeline Leak Monitoring

The presence of a trend component in the pressure signal has a detrimental effect on the accuracy of pipeline leak detection based on the negative pressure wave (NPW). To address this issue, a trend judgment criterion based on the signal interval statistical distribution (SISD) is introduced. In addition, an adaptive trend removal method is developed by combining trend detection and removal techniques. This method removes the trend component by iterative calculations without setting parameter thresholds. The SISD-based trend detection method in this article can be combined with different trend removal methods to remove the trend component adaptively, and it performs excellent real-time and universality. Experiments on a real-world pipeline dataset show that the proposed trend detection method combined with the Butterworth digital high-pass filter can remove the trend component in pressure signal adaptively, while the time-domain transiency of NPW is well retained and the signal-to-noise ratio (SNR) of NPW is improved, which has a positive effect on improving the detection accuracy of the leak signal.