Researchers at the UK’s University of Aberdeen have launched an unmanned flying drone programme in a bid to identify ways of recovering more oil reserves from the North Sea
John Howell, geoscientist at the University of Aberdeen, said, “When you drill a well in the North Sea, you can directly measure the rocks in the borehole. However you have much less certainty about what is going on away from the well. Given that two wells are often several miles apart, predicting what the rock layers in between the boreholes look like is a huge challenge.
“To solve this problem, we look at similar rock units which occur in cliffs above sea level and we use the drone to make extremely detailed 3D models, which we can then adapt for the subsurface. This gives us a much better idea of what conditions are like between these two boreholes and then allows us to predict how the oil will follow and how much we can recover.”
According to the university, the research programme is supported by more than 24 global oil firms.
Howell added that the advantage of the drone is that it allows collecting large volumes of data from otherwise inaccessible cliff sections in remote and often dangerous places
Data technology
The US$15,000 project called Safari generates not just pictures, but accurate 3D models of entire rock faces.
“The drone uses twin cameras to take pictures that are offset at slightly different angles and perspectives. Then, using a single point of reference on each, the photos are combined using computer software to triangulate a 3D projection from a flat, two dimensional image. This technique, called stereophotogrammetry, is actually quite old, but digital cameras and advances in computing power have made it much more useful,” Howell noted.
The overall project’s goal is to develop a fully searchable database of relevant rock formations, which will help oil companies build better models of the subsurface and improve recovery from oilfields, concluded the geologist.
To register for Big Data Analytics for Oil & Gas, please visit: http://www.oilandgasbigdata.com/register
