Abstract

Resource estimation is a process by which the economically recoverable hydrocarbons (HC) in place within a reservoir are calculated. The parameter that goes into resource estimation is variable and contains some aspect of risks. Resource can be defined as the total volume of hydrocarbon in place, which is further classified as reserve when proved commercially viable. Reserve estimation can be carried out deterministically while resource estimation is done probabilistically. In this paper we have utilized a probabilistic approach using genetic algorithm (GA) for resource estimation. GA is an adaptive heuristic approach which uses an analogy to the mechanism of Darwin’s theory of natural selection. This paper discusses the basic structure of GA and its operators that is, encoding, reproduction and termination, followed by the application of GA to a synthetic model for estimating the oil initially in place (OIIP) and recoverable resource, using triangular and lognormal probability distributions.

Key words: Genetic algorithm, resource estimation, crossover, mutation.