Field production constrained with surface facilities on gas handling have to deal with well rates optimization by reducing gas oil ratio of the field production. This means the best way of reducing gas oil ratio on field level is not by closing wells with the highest gas oil ratio but chocking back wells where gas breakthrough occurred and GOR of a well is rate dependent [1].

In this paper, authors modeled and analyzed wells with gas breakthrough in single porosity and dual porosity sector models. The analysis showed single porosity models underestimate severity of gas breakthrough and fail to predict rate dependent GOR of a well in the field.

Also, based on the sector model using machine-learning technique an empirical equation was developed to estimate rate dependent GOR of a well which can be further used in field level production optimization exercise to reach maximum liquid production under gas processing constraints.

Keywords:
machine learning, flow in porous media, fluid dynamics, condensate reservoir, upstream oil & gas, enhanced recovery, complex reservoir, gor behavior, artificial intelligence, reservoir simulation
Subjects:
Reservoir Fluid Dynamics, Improved and Enhanced Recovery, Reservoir Simulation, Unconventional and Complex Reservoirs, Flow in porous media, Gas-condensate reservoirs
Copyright 2021, Society of Petroleum Engineers
You can access this article if you purchase or spend a download.