Computational methods for pipeline leakage detection and localization: A review and comparative study

Abstract

Pipelines are one of the least expensive means of transporting fluids in long distances and distributing fluids in large areas and cities. Fluids transported and distributed by pipelines are often potentially hazardous, can pollute the environment, and are of high economic value. As such, monitoring these pipelines to predict and detect leakage accurately and promptly, and to determine the location of the leak is of importance. This article reviews and evaluates existing computational methods of pipeline leakage detection and puts recent advances in this area into perspective. The methods are of the following types: mass/volume balance, negative pressure wave, pressure point analysis, statistical methods, and real-time transient modeling. The strengths, weaknesses, and limitations of the five types are discussed in terms of the person-hours that they need to detect a leak, and the certainty and speed of the leakage detection and localization. Future outlook for this field is also provided. To substantiate the evaluation, three of these methods are implemented and tested in a pipeline case study.