Saudi Arabian Oil Company

Huihai Liu

Mohammed Boudjatit

Shuo Zhang

Mustafa A Basri

Rebah Mesdour

Methods for predicting hydrocarbon production rates for a hydrocarbon reservoir include receiving data from a hydrocarbon reservoir. The data includes reservoir characterization data, well log data, and hydraulic fracturing data. A physics-constrained machine learning model predicts a hydrocarbon production rate for the hydrocarbon reservoir as a function of time. The physics-constrained machine learning model includes an artificial neural network and a hydrocarbon fluid flow model. Predicting the hydrocarbon production rate includes generating, by the artificial neural network, multiple parameters of the hydrocarbon fluid flow model based on the data from the hydrocarbon reservoir. The hydrocarbon fluid flow model provides the predicted hydrocarbon production rate as a function of time based on the parameters. A display device of the computer system presents the predicted hydrocarbon production rate for the hydrocarbon reservoir as a function of time.

Determination of hydrocarbon production rates for an unconventional hydrocarbon reservoir

Changes in capillary pressure and relative permeabilities in reactive transport codes or reservoir simulators are formed by computer modeling. Based on continuum-scale hydraulic properties, the pore size distribution (PSD) is determined from a capillary pressure curve using the capillary tube concept. Changes in mineral volume through equilibrium or kinetic mineral reactions are then translated to changes in pore radii of the pore size distribution by selectively changing the radii of water occupied pores. The resulting new pore size distribution is converted back to an updated capillary pressure curve, which is then used for determining and forming models of total permeability and relative permeabilities at the continuum scale.

Growing community of inventors

Houston, TX, United States of America

Huihai Liu