In this study, effectiveness of combinatorial algorithms for describing reservoir properties has been evaluated. Three different types of algorithms: genetic algorithm, simulated annealing, and stochastic hillclimbing method are tested, and the performances of the algorithms are compared through numerical experiments. It was observed that the modified stochastic hillclimbing is the most efficient method for variogram based objective function. The method was also validated by applying it to a case where both variogram and well testing information were available. We concluded that modified stochastic hillclimbing is order of magnitude times faster than the simulated annealing method. The reason for the success of the modified stochastic hillclimbing method is provided through objective function analysis. Finally, the method is applied to a field case to generate interwell porosity distribution.

Keywords:
application, xxxxxxixxxxxxxxxxx, geological modeling, shc method, reservoir description, reservoir property, variogram, artificial intelligence, reservoir characterization, society of petroleum engineers
Subjects:
Reservoir Characterization, Information Management and Systems, Geologic modeling, Artificial intelligence
Copyright 1994, Society of Petroleum Engineers
You can access this article if you purchase or spend a download.