A Production-Allocation Method for Two-Zone Downhole Commingling in Semi-Intelligent Oil Wells

Abstract

Abstract Downhole commingling is a production-optimization method for producing multiple reservoirs (zones) via a single wellbore. It offers cheap exploitation of small reservoirs that would have remained stranded for uneconomic offtake rates and recovery. Although this method reduces development costs and environmental footprints, production allocation is a concern, especially in the absence of intelligent completion. Here, an intelligent commingling employs active inflow control valves (ICV) across individual zones. Nevertheless, this ideal scenario of full intelligence is not always available in practice. Considering the opportunities and challenges of stranded hydrocarbon resources amid tightening global energy supply and drive for cleaner fuel, this paper considers semi-intelligent commingled production (SICP) as an enabler for production acceleration. An SICP has a permanent downhole gauge for real-time pressure and temperature measurements but, unlike its intelligent counterpart, SICP lacks ICVs for zonal surveillance and control. We develop a fit-for-purpose method and workflow for back-allocation in a two-zone SICP. It is a semi-empirical approach premised on pressure-volume-temperature (PVT) characterization. Different mixing proportions of fluid samples from target contributing zones are baselined in terms of compositions and aggregate properties. These same properties of commingled stream are monitored over time to back-allocate produced gas, oil and water phases. Unlike production logging and conventional fingerprinting that require frequent well re-entry and large spectrum of fluid components respectively, our method is cheap and minimizes well re-entry and its associated risks. Specifically, we exploit C1, C7+, and salinity (Cl−) as markers to uniquely characterize the gas, oil and water phases of individual zones, respectively. Given the challenges posed by highly biodegraded oil reservoirs and reservoirs with similar PVT properties, we propose augmenting our back-allocation method with the use of biomarkers from saturate and aromatic fractions of the oil in such cases.