Speaking at the second virtual International Rock Imaging Summit (iRIS), Raymond Pols showed how Craytive Technologies is using augmented reality to make core images more accessible. Use cases in geothermal, CCS, and gas storage are driving demand for access to core data. Virtual reality for examining cores has been mooted for some time. But earlier efforts for 3D data rooms did not really work as they were too costly to set up and maintain. Today, accessible technology such as headsets and tablets is changing the equation. Craytive has worked with Core Technical Services to establish BaselineZ, an immersive remote collaboration platform. BaselineZ provides access to cloud-based holographic core data accessible from anywhere. Cores can be viewed with the Microsoft Hololens and there are plug-ins for Petrel, JewelSuite and more. Interested parties can set up their own Azure cloud-based core data rooms. VR functionality extend to a ‘virtual core sheds’ allowing a user to pick a core from the table and hand it to another remote colleague. BaselineZ also provides point and click access to CT scan data and other imagery.
Rail Kadyrov from Kazan Federal University has also been working on augmented reality, leveraging the ‘universal scene description zipped’ (USDZ) format from Apple and Pixar that is supported by many mobile devices. The resulting AR models can be shared as 3D visualization for integration into websites or use in scientific publications and conference presentations.
Eric Piard from the University of Liege, Belgium presented RockePedia, a.k.a a ‘shared intelligence platform for computer-assisted rock identification.’ Piard believes that scanning needs to move on from the acquisition of pixels to acquiring ‘roxels’. Such ‘rock elements’ will be the building blocks of a future digital twin of the earth’s crust. Piard envisages that in the future, researchers will be able to ‘pool all the acquired images from geological surveys’ into a large collective visual intelligence database ‘tentatively named’ RockePedia. RockePedia will be developed atop a sharing platform similar to the Cytomine database. Read Piard’s RockePedia pitch.
Francois Ducet presented Elemission’s laser-induced breakdown spectroscopy. LIBS analyzers can identify and classify the chemical composition of any material, regardless of state. LIBS probes were deployed on the Curiosity and Perseverance Mars rovers. Elemission has leveraged machine learning to build a library of mineral fingerprints. Scanning is fast with up to 1,000 spectra per second. A new Ecore system is available for scanning drill cores.
While the CSIRO work on applying machine learning models to analyze hyperspectral data as presented by Fang Huang targeted the mining industry, results from the Rosetta Framework may be of interest to the oil and gas community. CSIRO’s supervised machine learning models predict geochemistry and lithology labels using Corescan hyperspectral data. Preliminary results show that the models have a pretty good accuracy in predicting lithological labels and geochemical element concentrations, ‘significantly improving the efficiency of utilizing raw hyperspectral images’.
Henk Kombrink introduced the North Sea Core CIC (a community interest company), a volunteer-run core store that started life in 2018 and is now supported by OGA and sponsored by Netherlands-based E&P One-Dyas. The idea is to give cores a second life by ‘upcycling’ information for distribution to amateur and professional geologists. The company has been granted a license to DUG Insight software. The shoestring operation is run from a garage by enthusiasts keen to preserve their ‘geo-heritage’ with training and energy transition research. One tangible deliverable is ‘Geoart’, a framed North Sea core slab for the office wall.
Adrian Neal (Badley Ashton) provided a summary of the British Geological Survey’s ‘Role of cores in 21st century reservoir characterization’ conference. This kicked-off by questioning the argument whereby advances in seismic and borehole imaging mean that cores are now surplus to requirements. The core community had no trouble in destroying this strawman. A close examination of cores in Permian ‘shale’ shows that there is huge lithological variability. In fact much is not shale at all (as Oil IT Journal surmised in 2017). Elsewhere cores allow sedimentology to be tied to shale carbon content. New technologies such as QemScan and portable XRF are allowing researchers to re-assess cores to improve reservoir understanding. Shell’s work with Imaged Reality got a shout-out. VR can ‘bridge the scale gap’ between core and outcrop interpretation. Others warned that while new uses, notably carbon storage site characterization, are on the horizon, core repositories need funding if they are to be around in 10 – 20 years’ time. Proselytizing for the core store is important. ‘Make sure folks know about them and appreciate their value, cite them in research and, if possible, quantify their value in monetary terms’.
More from iRIS.
© Oil IT Journal - all rights reserved.