Eventos Anais de eventos
COBEM 2023
27th International Congress of Mechanical Engineering
Identification and Classification of Leaks in Dense-Gas/Liquid Pipe Flow through Pressure Signature Analysis
Submission Author:
Carlos Mauricio Ruiz Diaz , SP , Brazil
Co-Authors:
Carlos Mauricio Ruiz Diaz, Johann Eduardo Castro Bolivar, Cristhian Alvarez Pacheco, Marcus da Silva Carr, Oscar Mauricio Hernandez Rodriguez
Presenter: Carlos Mauricio Ruiz Diaz
doi://10.26678/ABCM.COBEM2023.COB2023-0355
Abstract
In the hydrocarbon industry, flow assurance is an important factor for the operation of oil and gas wells, which can be exposed to deviations in their normal flow conditions. This increases the cost of system maintenance and results in a considerable loss of resources, ultimately affecting the environment. Therefore, the accurate and fast detection of anomalies or deviations in pipelines is critical for the safe and profitable production and transportation of oil and gas, as it prevents unexpected failures or shutdowns in production. However, it is worth highlighting the difficulty of finding sensors installed in pipelines that can directly detect the occurrence of undesired events. When sensors are found, it is often impractical to install them. Traditional prediction and classification methods based on physical models or purely on real data collected from wells or simulated data using specialized software have limitations, such as high uncertainty and high computational cost. In this research, an experimental campaign was carried out to collect experimental data under hydrodynamically similar conditions to the Brazilian pre-salt layer, where controlled and monitored normal condition without leak-type anomalies was developed. A total of eighteen experimental data points were generated in the Industrial Multiphase Flow Laboratory - LEMI under specific conditions to investigate the behavior of two-phase dense-gas/oil flow in horizontal position. These measurements allowed for the determination of the pressure drop and void fraction associated with each flow pattern. Additionally, a reference state was established for pressure drop in a SF₆-oil horizontal pipe through global and windowed signal analysis based on pressure drop signature from a differential pressure sensor. The results show that the oscillatory process is stationary with a tendency towards a constant value that depends on the operational conditions.
Keywords
Leakage, dense gas, Two-phase Flow, differential pressure signature
