A 3D leakage monitoring method for pressure vessel based on region segmentation and time difference estimation

Abstract

Leak monitoring and location technology is an important guarantee for the safety of pressure vessels. However, the first-time capture of sudden leakage incidents and the structure of the revolving body have brought challenges to the full range of pressure vessel monitoring and location. In this paper, based on the structural characteristics of the cylindrical pressure vessel, a 3D monitoring and location method based on region segmentation and time difference estimation is proposed, which combines regional location and time difference location to improve the location speed and accuracy of the algorithm. The adaptive threshold is adopted to capture the first moment of leakage, and the regional cross-correlation calculation is used to locate the leakage source in the area. The performance of the method is verified by liquid leakage experiment based on distributed acoustic emission sensor. The results show that the proposed method has good adaptability. It can capture leakage events and locate the leakage events at different positions and apertures in the first time, with an average location error of 2.44 cm. This study will have a new understanding of the location technology of pressure vessels and provide a reference for the health monitoring of large pressure vessels.