For oil or gas fields with stratified reservoir layers, detailedproduction contribution for individual layer is always desired.Unfortunately, in some particular cases, production wells arecompleted following commingled scheme. This is worsenedfurther if only very few production tests are run for the field.This is the case for the Central Sumatera field with its 95commingled production wells, among which only a few hadundergone production tests and none of them have everundergone production logging. Problems rise when theoccasion came in which detailed production contributionfrom individual reservoir layer is required for the field’sreservoir simulation modeling and productionevaluation/prediction.This paper presents an approach to solve the problem. Theapproach is basically based on the application of softcomputing (Fuzzy Logic) to investigate pattern ofrelationships between production contribution of layers incommingle wells and rock petrophysical data as well as otherrelevant geological/engineering data. For the purpose, thirteenwells (key wells) that have production tests are assigned,among which three wells are assigned for checking thevalidity of the recognised pattern. Using the validated mostvalid pattern, individual layer’s production allocation for otherwells are determined with well-log analysis data as the majorinput.Result estimates for the candidate wells are better compared toresults produced by the conventional method of productivityindex (PI) analogy. The resulted variation in water cut andseparate oil and water split factors appear to be more realisticfrom any point of view.
展开▼
