Back to search

NAERINGSPH-Nærings-phd

Application of predictive data analytics to wellbore geomechanics on the Norwegian continental shelf

Alternative title: Anvendelse av prediktiv dataanalyse til borehull geomekanikk på norsk sokkel

Awarded: NOK 0.31 mill.

Project Manager:

Project Number:

317606

Application Type:

Project Period:

2020 - 2021

Funding received from:

Location:

Despite significant progress, borehole instabilities and well control incidents are still major challenges in drilling wells across the globe. An overview of wells drilled the last 25 years on the Norwegian continental shelf (NCS) show that an average of 14% of all exploration wells and 8% of all production wells encountered well control incidents. These severe incidents lead to costly delays and the need for activation of contingency measures affecting safety. The proportion of well control incidents related to uncertain pore pressure predictions is high; around 80% of the cases. Most of these problems can be avoided with a better understanding of the conditions in the subsurface and adjusted drilling practices, leading to more efficient and safer drilling. By introducing drilling mechanics data, geomechanical modelling, and advanced data analytics, reliable predictions of stress and pore pressure can be obtained. Applying data analytics and machine learning solves problems with data irregularity and allows joint analysis of a higher number of wells, allowing the datasets to “learn from each other”. Combining geophysical logs and drilling mechanics data in new ways allows for a more precise determination of the top and base of over pressured zones in the subsurface. Additionally, a quantified analysis of drilling mechanics data allows pressure prediction in non-shale lithologies, which has been a major challenge in the oil industry and deep geothermal wells. The combined innovative approach of this study aims to reduce the current uncertainty in pore pressure and geomechanical models, with the objective to reduce borehole instability and well control incidents. The purpose of the project is that the research and developed algorithms will be applicable for any drilling application including geothermal energy.

Despite significant progress, borehole instabilities and well control incidents are still major challenges in drilling wells across the globe. On the Norwegian continental shelf about 40% of the wells encountered issues related to borehole stability or overpressure, leading to costly delays and the need for contingency measures affecting safety. Most of these problems can be avoided with a better understanding of the conditions in the subsurface and adjusted drilling practices, leading to more efficient and safer drilling. By introducing drilling mechanics data, geomechanical modelling, and advanced data analytics, more reliable predictions of stress and pore-pressure can be obtained. Applying data analytics and machine learning solves problems with data irregularity and allows joint analysis of a higher number of wells, allowing the datasets to “learn from each other”. Combining geophysical logs and drilling mechanics data allows for a more precise determination of the top and base limits of overpressured zones in the subsurface. Furthermore, a quantified analysis of drilling mechanics data allows pressure prediction in non-shale lithologies, which has been a major challenge in the industry. In the project, the geomechanical methodology developed by Geoprovider will be scientifically verified and improved, resulting in higher cost efficiency and improved safety for drilling operations. Using the wealth of data available on the Norwegian continental shelf, a number of case studies will be presented demonstrating the progress and applicability of the methods. The algorithms can be used globally, and the results of this project will be specifically tested for application in geothermal energy using data from high pressure high temperature wells.

Funding scheme:

NAERINGSPH-Nærings-phd