Improve oil extraction with new techniques

The exact map of a large underground oil field was developed to direct engineers' efforts to bring oil from large rocks into wells so that the oil could be pumped up for storage or transport. moved on.

MIT researchers have developed technology that can build a map that has the potential to significantly increase oil extraction from mines.

The new technology uses JPEG's digital imaging technology to create complete, realistic-looking maps of underground oil fields using scattered oil well calculations.These maps are the first tool to provide complete details of an oil field to guide actual oil extraction.

Behnam Jafarpour, MIT graduate student, is currently a professor of petroleum engineering at Texas A&M University, saying: 'Our simulation studies show that this new method has the potential to improve those Current technology and provide predictions of oil field production. Hopefully, better prediction will be a premise for more efficient operation and increased oil production. '

The image above shows the oil saturation map in the standard oilfield model with known properties (real model). The image below shows the almost similar reconstruction of the same oil field using the McLaughlin / Jafarpour treatment method, which uses independent data from 23 wells and recreates a sustained recovery effort of 36 month. Black dots represent underground oil wells. ( Photo: Behnam Jafarpour). Jafarpour and Dennis McLaughlin, Professor of Water Management at MIT, have published two papers describing the technique and will appear in the next issue of the Society of Petroleum Engineering, the third article appeared in Computational magazine. Geosciences number June 2008.

The spatial structure of geological formation allows for the mapping of rock properties.However, JPEG compresses a lot of pixels in a detailed image down into a few pieces of information that need only very small capacity. In the MIT researchers' oil simulation simulation, the same mechanism is used to provide brief descriptions of the rock properties of the oilfield. The new technique uses data on oil flow rates and pressures from oil wells to form a real picture of the underlying surface of the oilfield.

Extraction of petroleum is expensive and inefficient - sometimes only 1/3 of the oil in a mine is extracted. So engineers rely heavily on mining techniques such as spilling water to mobilize oil. To guide and control this work, they make real-time predictions of sub-surface changes, including saturation and oil pressure, but they often work groping. The stone properties needed to make these predictions (eg the liquid conductivity of a rock at a certain depth) cannot be observed or measured.

Instead, engineers deduced indirect geological attributes from seismic data and measurements taken in scattered oil wells.

McLaughlin, who just walked the project, said: 'In a normal oil field, millions of pixels are needed to fully describe the complicated sub-surface road that transports oil to wells. Unfortunately, the number of seismic and observing available oil wells to determine the value of these points is often very limited. The methods we developed get more information from those limited measurements to provide better predictions about the sub-surface lines and the amount of oil moving inside. '

During the 36-month recovery process, McLaughlin and Jarfarpour's method accurately identified the main trends and properties of the liquid conductivity of an oil field, proving that this new technique is accurate and effective. .

Jafarpour said: 'Our next step - already in progress - is to test this idea in real oil fields and evaluate its impact on oil exploitation under real conditions'.

The study was funded by Shell International Petroleum Company.