Mechanistic modeling of pipeline leak detection at fixed inlet rate

Mechanistic modeling of pipeline leak detection at fixed inlet rate

deepwater

leak

leak detection

multiphase flow

pipeline

A leak from subsea pipelines has been a serious problem in deepwater operations because of its potential threat to surrounding marine environments, significant monetary loss from the delay of hydrocarbon production, and difficulty in detection and remedial processes. For the first time, this study shows how a leak in long horizontal subsea pipelines can be modeled mechanistically using pressure traverse calculations, and presents the results in a form of contour plots, with “a change in inlet pressure (ΔPin)” or “a change in outlet flowrate (Δqtout)” induced by a leak in z axis as a function of leak opening size and longitudinal pipe location in the x and y axes. Beggs and Brill's correlations provided a means of deciphering complicated gas–oil two-phase flow mechanisms in pipelines. The calculated steady-state response of the system before and after the leak was investigated in terms of pressure gradient, flow rate, holdup, and flowing fraction, which were validated and well agreed with results from a quasi-dynamic numerical simulation. The results showed that the outlet flowrate monitored at the receiving facilities (qtout) could be a good indicator to detect a leak, while the pressure monitored at the inlet (Pin) was less reliable. The use of inlet pressure became more trustworthy with decreasing backpressure downstream (or increasing gas compressibility equivalently). The results also demonstrated that a leak is more easily detected if located further upstream with a larger opening size, and the presence of a compressible phase in pipelines makes leak detection more favorable leading to larger Δqtout. The effects of gas–oil ratio of leaking fluids, gas compressibility, leak size, and leak coefficient were investigated, and the nature of transient behavior between the two steady states, before and after the leak, was also examined.

145-156

3

70

Kam, Seung Ihl

Journal of Petroleum Science and Engineering

2010

journalArticle

0920-4105

10.1016/j.petrol.2009.09.008