The oil & gas sector has massive data sets in the form of seismic data that it needs to analyse in order to identify the location of hydrocarbons and also to assess the risk for drilling operations in a particular area.
Integrating recently developed technology in the form of 3D visualisation tools (developed at the University of Aberdeen) with optimisation techniques would result in an integrated and interactive tool that would massively speeds up the understanding of the huge amounts of geophys data that oil & gas companies hold. The software tool would allow a geophys analyst to review the results of analysis in real-time by manipulating the 3D environment using intuitive hand gestures.
The software would therefore be in the form of a fully interactive 3D analysis tool that would automatically identify geological features of interest. The use of the 3D engine would allow commodity hardware such as Oculus Rift or other 3D viewers to be used, which would greatly enhance the user experience. This will be similar to current seismic analysis tools, with the main difference being that the analysis will be truly in 3D rather than 2D slices like other currently available tools.
In essence, the project would look at developing new technologies in the areas of:
• Optimising and analysing seismic data in 3D (not 2D slices)
• A modern 3D volume visualizer capable of real-time animation and interaction.
• Integration into commodity VR such as Oculus Rift or any other 3D hardware.
The aims of this project are to fine tune existing automated methods based on image and signal processing to recognize, track, visualize and manipulate specific geobodies using and transferring alternative optimisation methods and visualisation technologies originally developed through the DynamO project at the University of Aberdeen which has been optimised for the real time visualisation and simulation of 3D data, originally for Chemical Engineering problems. Our target is to specifically improve the now well established geobody innovative rendering technology by allowing a direct interactive manipulation of the geobodies and mapping and to allow a user to manipulate their related seismic attributes properties in a more efficient way. The methods will rely on novel data-visualisation techniques by borrowing techniques from the game industry and protein industry to render large data-sets interactively and in real-time. These techniques can be used to reveal hidden relationships between data-sets as well as providing visually pleasing renderings of simulations. Another issue concerning interactive facies analysis is that it generally relies on specific mathematical techniques based on using crossplot methods to manually develop patterns of classification and build up what people call self-organized map. We will also explore different automated data optimisation methods to perform a better and more complete pattern of classification, based on work undertaken at the University of Aberdeen in a number of fields.
This proposal will represent an improvement of the existing image processing software available in the market (Petrel (Schlumberger), Geoteric (ffA), GeoProbe (Halliburton)) that in fact already allows us to performs sophisticated operations such as attribute mapping, editing, draping and to interpolate information by creating blended volume attributes properties using integrated facies analysis, draping image processed data into the mapped horizons as well to perform a crossplot analysis of the various properties from specific chosen area. However most of the current operation and tools available are produced through a complex workflow and partial interactive analysis of the attributes properties (often not in real time). Finally most of the visualization and image processes that rely on commercial software do not integrate well the petrophysical information nor incorporate directly the well log data with the rendered geobodies attributes obtained through image processing. This limits the direct comparison of petrophysics values with obtained seismic facies attributes map created through image and signal processing filters.
The successful candidate should have (or expect to achieve) a minimum of a UK Honours degree at 2.1 or above (or equivalent) in Engineering, Geosciences, Physics or Computational Science.
Knowledge of: computer programming, data analysis, algorithms.
Formal applications can be completed online: http://www.abdn.ac.uk/postgraduate/apply
. You should apply for Degree of Doctor of Philosophy in Engineering, to ensure that your application is passed to the correct person for processing.
NOTE CLEARLY THE NAME OF THE SUPERVISOR AND EXACT PROJECT TITLE YOU WISH TO BE CONSIDERED FOR ON THE APPLICATION FORM.
Informal inquiries can be made to Dr A Starkey ([email protected]
) with a copy of your curriculum vitae and cover letter. All general enquiries should be directed to the Postgraduate Research School ([email protected]