Unconventional volatile-oil reservoirs exhibit near-constant produced gas-oil ratio (GOR) at pressure below the bubble point, a behavior that is quite different from that in conventional reservoirs. The primary purpose of this paper is to explain the reasons for this observed anomalous behavior using compositional reservoir simulation, with field examples from Eagle Ford shale play. Then it discusses how GOR history can help us interpret rate transient analysis (RTA) diagnostic plots and forecast multiphase well production using decline curve analysis (DCA) methods, which were originally developed for single-phase flow.
We have conducted comprehensive reservoir numerical simulations to model the non-linearity caused by the impact of nano-pores on rock and fluid properties, complex pore networks and the multiple porous media flow systems created by multi-stage fracture stimulation. We used the numerical model to generate synthetic datasets based on realistic reservoir fluid and rock properties. volatile-oil reservoirs produced under various rate/ flowing pressure production conditions. The fundamental difference between GOR during linear vs. radial flow were explained using our simulation model. We used traditional RTA approaches to analyze producing GOR and pressure and saturation changes within the reservoir. We then modeled production rate at varied pressure drawdowns and performed DCA on data from flow regimes identified with RTA plots and GOR behavior.
We found that the flattened GOR behavior is function of fracture conductivity, fracture spacing, critical gas saturation, fluid volatility, matrix permeability and bubble-point suppression. Analysis of GOR behavior helps us interpret where a well is in its GOR history and flow regime. We observed that GOR changes correlate with flow regime changes identified on RTA diagnostic plots. GOR is constant at a level higher than original solution gas-oil ratio during transient linear flow, and during most of the transition period that follows. A log-log unit-slope line forms only when pressure throughout the entire reservoir drops below the bubble point, and gas saturation exceeds the critical threshold, thus causing a sharp increase in GOR. This GOR can decline later if compound linear flow develops. Our study showed that DCA methods that employ different Arps b-parameters to represent different flow regimes, combined with drawdown-adjusted production rates, allow us to predict tight volatile oil reservoir performance with improved accuracy.