Abstract
Increase recovery from mature oil reservoirs requires the definition of enhanced reservoir management strategies, involving the implementation of advanced methodologies and technologies in the field's operation.
This paper presents a digital workflow enabling the integration of commonly isolated elements such as: gauges, flowmeters, inflow control devices; analysis methods and data, used to improve scientific understanding of subsurface flow dynamics and determine improved operational decisions that support field's reservoir management strategy. It also supports evaluation of reservoir extent, hydraulic communication, artificial lift impact in the near-wellbore zone and reservoir response to injected fluids and coning phenomenon. This latest is used as an example to demonstrate the applicability of this workflow to improve and support operational decisions, minimizing water and gas production due to coning, that usually results in increasing production operation costs and it has a direct impact decreasing reservoir energy in mature saturated oil reservoirs.
This innovative workflow consists on the continuous interpretation of data from downhole gauges, referred in this paper as data-driven; as well as analytical and numerical simulation methodologies using real-time raw data as an input, referred in this paper as model-driven, not commonly used to analyze near wellbore subsurface phenomena like coning and its impact in surface operation. The resulting analyses are displayed through an extensive visualization tool that provides instant insight to reservoir characterization and productivity groups, improving well and reservoir performance prediction capabilities for complex reservoirs such as mature saturated reservoirs with an associated aquifer, where undesired water and gas production is a continuous challenge that incorporates unexpected operational expenses.
Implementation of multilayer perceptron (MLP) and radial basis function (RBF) neural networks to predict solution gas-oil ratio of crude oil systems