Leakage Localization for Long Distance Pipeline Based on Compressive Sensing

Abstract

Leakage in the long distance pipelines may lead to severe impairment to water transportation. In practice, dealing with the data collected by wireless sensor networks (WSNs) is essential to achieve real-time leakage localization to reduce water loss. More sensors may improve the effectiveness of monitoring the leakage of long distance pipelines. However, to avoid deploying numerous sensors which leads to drill many holes in the pipeline to install them, a novel algorithm based on the compressive sensing (CS) theory is proposed in this paper. First, a new sensing matrix named pressure variation decay (PVD) model based on the attenuation of pressure wave is creatively established, which describes the leakage status of long distance pipeline concretely. As such, the equation of leakage localization based on CS theory reconstruction algorithm is deduced. Then, in order to assess this new method, the simulation of the method and the discussion of results are conducted. Finally, the results show that the proposed algorithm possesses validness and excellent robustness, even if the interference of noise.