MATCHING AN AUTOMATED PRODUCTION HISTORY USING BAYESIAN OPTIMIZATION

摘要

An oil field model mapped to a history that facilitates well system operations for an oil field is generated using Bayesian optimization of adjustable parameters based on a complete production history. The Bayesian optimization process (200) includes stochastic modifications of the adjustable parameters based on an earlier probability distribution for each parameter and a model error generated using historical production measurement values and values of corresponding model prediction for various sets of test parameters.