Experimental study of leakage detection of natural gas pipeline using FBG based strain sensor and least square support vector machine

Abstract

Leakage is the most common cause of natural gas pipeline accidents. This work was devoted to natural gas pipeline leakage detection, which is based on detecting negative pressure wave signals caused by leakage. The FBG strain sensor, which is based on monitoring the hoop strain of a pipeline to detect negative pressure wave signals, is fabricated and experimentally tested. Compared to conventional pressure sensors, FBG strain sensors were shown to be less influenced by noise, and they have the advantage of being a nondestructive sensing method. This makes them ideal for sensing pressure transients, which could be analyzed to detect natural gas pipeline leakage. Toward this objective, a least square support vector machine (LS-SVM) classifier was developed as an automatic leakage detection technique. This technique proved to be effective at detecting leakage.