Oil Reservoir Simuluation

A new method has been developed to quickly determine fine-grid reservoir simulation pressures. It is estimated that the pressures in a one-billion cell simulation model can be determined in less than one percent of the time required by other state-of-the-art solvers. Increased accuracy is obtained by using new finite-difference equations to approximate partial the differential equations governing the flow through porous media. Unlike traditional Taylor-series approximations that are based on polynomials, the new equations can accommodate highly nonlinear solutions including solutions with singularities and reservoir pressures exhibit such behaviors.

General Information

The development of new finite-difference equations that incorporate the physics of an oil reservoir to approximate the partial derivatives significantly increases the accuracy of the solutions obtained over traditional Taylor-series approximations. Taylor-series approximations are based on polynomials that do not adequately represent highly nonlinear solutions or solutions with singularities or discontinuities. Such solutions are found in oil reservoirs due to the immiscibility of the fluid displacing the oil creating near discontinuities of pressure (and resulting flow velocities), at the interface and the injection and production areas of the well, creating near singularities in pressure near those locations. Often correlations, e.g. Peaceman corrections, are used to compensate these shortcomings of the Taylor-series approximations. The new finite-differences equations have shown to be more accurate then even the correlation-corrected Taylor-series approximations.

By incorporating the new equations to more accurately model the reservoir pressures into a novel technique of iteration, the processing time and power required have been substantially reduced. An accurate fine-grid model of the pressures can be generated by first obtaining course-grid solutions with the new equations and then applying the solutions to a fine-grid and iterating between the points of known pressure.

The Market

Computer modeling of an oil reservoir is vital in determining the productivity and profitability of its petroleum reserves. Unfortunately, accurate modeling can require vast amounts of computer time and tax even exceptionally powerful computers. This new solver can be tens to hundreds of times faster then traditional methods, provide increased resolution —and therefore accuracy— and may be run on today’s personal computers. This greater accuracy of reservoir simulation is provided by allowing for fine grid simulations with amounts of data points that are impossible by current technologies. By combining very accurate finite-difference equations to approximate the partial differential equations, and the multi-step iteration process converging onto a solution, an accurate reservoir simulation can be quickly and easily obtained.

Type of Offer: Licensing



Next Patent »
« More Computer Science Patents

Share on      


CrowdSell Your Patent