Kavi

Problem

We help offshore drilling operators improve safety & efficiency in the process of rock cutting analysis.

In current practice, drill cuttings come out of a drilled well in mixture with drilling fluid (mud). A sample catcher goes to mud return line to take cutting samples at regular intervals. The sample is then washed, separated from contamination, dried, and prepared to be examined under optical devices. A mud logger analyzes the cutting sample and describes geological features including hydrocarbon show, rock lithology, visible porosity, color, etc. in the mud log report.

The process is heavily expertise dependent and intensive subjectivity in the interpretation result.

In case of fast drilling or geostopping application, the process is inefficient because of poor vertical resolution or slowing drilling rate of penetration – thus increases rig time (money).

Proposed Solution

I want to develop a computer vision technology to help identify and quantify rock cutting properties more accurately and in less time. The system comprises of wide spectrum microscope, and other sensing devices to capture images of rock cuttings. An image recognition algorithm is used to detect rock cuttings, and to quantify their petrogeological properties under semi-automatic human supervision.

Contact

To learn more about this project, please contact Thuy Vo at https://www.linkedin.com/in/volamnghithuy/