When a reservoir is subject to double-edged water drive mechanism, another well completion may be utilized to (1) intervene in an already watered-out well, for the purpose of optimizing oil production, (2) deliver material through one well to the other, or (3) strengthen confidence on earlier results of reservoir characterization. In all these applications, communication between the wells must not be in doubt. Procedures for analyzing interference tests data to determine the degree of communication between points in such a reservoir system is the main objective of this paper.
Source and Green's functions are utilized in deriving appropriate dimensionless pressure expressions for a reservoir, bounded laterally by double-edged water, and consisting of horizontal and vertical well combinations. To strengthen analyses, dimensionless pressure derivatives are computed from the dimensionless pressure expressions derived. As a late time flow test, external distances between wells are considered in the computations. Well combinations considered are (1) horizontal-vertical; (2) horizontal-horizontal and (3) vertical-vertical. Analytical expressions for analyzing interference test data for each well combination are derived, even if well functions are reversed to strengthen the validity of results.
Results show that, if the reservoir is isotropic, interference test duration should be less than water breakthrough time tDbr ≥ x2eD/Π2 for valid results. Flow time beyond this limit gives only an indication of the strength and responsiveness of the encroaching aquifer. Furthermore, analytical expressions for calculating fluid and reservoir properties depend on the type of well used as active well during the test. Horizontal well combinations would produce earlier interference than vertical-vertical well combinations, given the same anisotropy and rate history. An active horizontal well at interference produces variation of slopes on dimensionless pressures and their derivatives plots between early flow period and commencement of irreversible steady state, while a slope of 1.151 is observed for an active vertical well during the same period. In addition to type curve matching, parameter matching is required for estimating complete reservoir properties. Finally, time of occurrence of interference is directly proportional to area between interfering wells.