Morphological filtering assisted field-pipeline small leakage detection

Abstract

Field-pipelines leakage detection based on the negative pressure wave theory is mainly aiming to effectively process pressure sample signal in time domain. Previous methods often perform well for detecting large leakage but poorly for small leakage. In this paper, an alternative way using a particular morphological filter constructed by basic morphological operators is considered. The particular filter employed can remove noise and disturbance while retaining the basic geometry shape of pressure sample signal. This feature allows the filter facilitating signal small and slow change detection. The effect of the structuring element length selection is also analyzed on simulated data to acquire empirical rules for practical applications. An experiment is studied to validate the performance of the employed morphological filter.