Methods and systems for reservoir history matching for improved estimation of reservoir performance

Abstract

The present disclosure presents methods and apparatuses for forecasting geological formation production data. For example, some example methods include identifying an array of parameter sets, determining a fitting error between a first set of the historical production data and modeled production data, wherein the modeled production data is obtained through execution of a simulation model based on each parameter set of the array, determining a validation error between a second set of the historical production data and extrapolated production data for each parameter set of the array, determining a combined error for each parameter set of the array based on the fitting error and the validation error, and identifying an optimal parameter set size for modeling the target geological region, wherein the optimal parameter set size is determined in reference to a minimum combined error of the combined errors determined for each parameter set of the array.

Description

technical field [0001] Embodiments described herein generally relate to reservoir modeling. Some embodiments relate to predictive modeling. Background technique [0002] Currently, engineers use mathematical models to estimate future oil and gas production from underground hydrocarbon reservoirs. Such models usually involve solving a system of differential equations to arrive at a desired estimate. Typically, many reservoir-related parameters appear in these equations, such as rock permeability, porosity, relative permeability, and others. These models are typically implemented by a computer, which typically takes these parameter values ​​as input for solving differential equations. [0003] To estimate the values ​​of these input parameters, engineers typically rely on historical production data (production rates, pressures, etc.) Production Data. During parameter estimation from historical production data, parameter values ​​associated with the geological region of in...

AI Technical Summary

Problems solved by technology

As a result, these parametric techniques are relatively inadequate for estimating reservoir parameter values ​​such that future production predictions are reliable

Images