Machine learning approaches for uncertainty assessment in reservoir production forecasting

A PhD studentship is available in the Institute of Petroleum Engineering at Heriot-Watt University, to work on an industry funded Uncertainty Quantification project.

The work will involve application of data-driven statistical learning methods such as Support Vector Machines and Artificial Neural Networks to address diversity of geological reservoir models. The challenge of the project is to link contemporary machine learning algorithms with the state-of-the-art Bayesian framework for uncertainty quantification developed in the group.

The successful candidate should possess strong numerical and analytical skills and an interest in data driven approaches. A good knowledge of high level programming is desired. Experience in geomodelling / geostatistics or reservoir engineering is beneficial.

The candidate will join a dynamically developed team of postdocs and PhDs lead by Prof. Mike Christie. The research carried out by the team addresses aspects of uncertainty quantification including stochastic optimization methods, solution error models, neural nets, and employs high-level scientific computation (including access to a recently commissioned 84 node linux cluster). The research is funded by a consortium of oil companies and also by UK EPSRC, and the skills acquired in while studying for the PhD are likely to be applicable to a wide range of areas, including the oil industry.

To apply send a CV to Dr V Demyanov, vasily.demyanov@pet.hw.ac.uk, by 31 October 2005