Estimation of Distance Between Interfering Wells Using Python Programming Language

Abstract The knowledge of the distance between two interfering wells is very important and forms part of the objectives of interference well test analysis. Ascertaining if wells are in communication with each other, especially if the wells do not have the same ownership is key to every operator, for flow boundary definition, arbitration, and neighbouring well location, either as relief or recovery well. The procedure utilizes pressure distribution of a vertical well in a reservoir exhibiting infinite-acting behavior. The total pressure drop in the observation well is aggregated using the superposition of multiple wells with multiple rates histories. The general equation solved eventually comprises of either single or multiple exponential integrals as major variables. A single linear equation is formed if only two wells are involved. Multiple linear equations are formed if at least two wells are involved. An observation well may also be active. If only two wells are involved, the solution to the equation is provided using the secant method of finding roots to estimate the distance. Where well pairs are selected for flow and observation, multiple linear equations are obtained. In all cases, only well distance apart is involved. Python as a programming language is deployed to automatically search for the distance between two wells or an average distance between wells in a multiple array of wells completed in the same or neighboring reservoirs in communication. Major inputs into the program are observed pressure drops and time, wellbore, reservoir, and fluid properties. The program outputs well distance, reservoir, fluid, well parameters and time for every run. The roles of the wells were interchanged as observation and active wells in turn, just to strengthen confidence on results obtained. Two wells (Well 1 and Well 2) were used for demonstration successfully. Distances between the wells as computed by the program compared closely to those manually computed.