Recently, low-permeability (tight) gas condensate and oil reservoirs have been the focus of exploitation by operators in North America. Multi-fractured horizontal wells (MFHWs) producing from these reservoirs commonly exhibit long periods of transient flow during which two-phase flow of oil and gas initiates due to well flowing pressures dropping below saturation pressure. History-matching and forecasting of such wells can be rigorously performed using numerical simulation, but this approach requires significant data and time to setup. Analytical methods, while requiring less data and time to apply, have historically been developed only for single-phase flow scenarios. In this work, a novel and rigorous analytical method is developed for history-matching and forecasting MFHWs experiencing multi-phase flow during the transient and boundary-dominated flow periods.
The distance of investigation (DOI) concept has been used for many years in pressure transient analysis to estimate distances of reservoir boundaries to wells, amongst other applications. In the current work, the DOI concept is used to estimate dynamic drainage area (DDA) for the purpose of forecasting of tight gas condensate and oil wells – a linear flow geometry is assumed. During transient flow, the DDA is calculated at each timestep using the linear flow DOI formulation; a multi-phase version of the linear flow productivity index equation and material balance equations for gas and condensate/oil are solved iteratively for pressure, saturation, and fluid production rate. The productivity index equations for gas and oil utilize pseudopressure which is evaluated with saturation-pressure relationships derived from PVT data. For boundary-dominated flow, when the drainage area is static, the inflow equations are again coupled with material balance for both phases.
The new method is validated against numerical simulation, covering a wide range in fluid properties, and operating conditions. The new method matches simulation acceptably for all cases studied. Field examples of MFHWs are also analyzed to demonstrate the practical applicability of the approach. The three liquid-rich shale examples analyzed were also chosen to represent a wide-range in fluid properties. In all cases, acceptable history-matches are achieved.
The new analytical forecasting/history-matching procedure developed in this work provides a practical alternative to numerical simulation for tight gas condensate/oil experiencing two-phase flow during the transient flow period. The method, which does not rely on Laplace space solutions, is conceptually simple to understand, easy to implement, and avoids the inconvenience of Laplace space inversion.