Mathematical Geosciences is fortunate to publish many outstanding research papers each year, which makes the selection of the Best Paper Award recipient a challenging undertaking. This award recognizes research that has made an especially significant contribution toward expanding and enriching knowledge in our field. We are pleased to announce that the winner of the 2018 Best Paper Award is:
“Porous Structure Reconstruction Using Convolutional Neural Networks,”
Volume 50(7): 781–799
By Yuzhu Wang, Christoph H. Arns, Sheik S. Rahman, Ji-Youn Arns
2 Congratulations to the Authors!
Yuzhu Wang obtained his B.E and M.E in geology and geophysics from China University of Geosciences (Wuhan). He has been the engineer of SINOPEC (China) during the period from 2010 to 2014. He earned his Ph.D. in petroleum engineering from University of New South Wales (UNSW), Sydney, Australia in June 2018, followed by a research assistant in UNSW until the April of 2019. His research covers multi-scale reservoir characterization (ranges from reservoir scale to pore scale), image processing and machine learning. Recent theoretical contributions relate to the reconstruction of the high-resolution porous structure of reservoir samples using local similarity phenomenon.
Christoph H. Arns is a Professor at the School of Petroleum Engineering, U. of New South Wales. He is a pioneer in the area of Digital Rock Physics (20 years) and specializes in the area of pore-scale petrophysics on (micro-) tomographic images including the integration of digital and conventional core analysis. His primary interest lies in the combination of 3D tomographic imaging technology and NMR spectral techniques for petrophysical applications with a focus on core-scale heterogeneity. He was an integral part of a team commercializing digital core analysis through the ANU/UNSW spin-off Digital Core Pty Ltd, formed in 2009 and a research scientist at Applied Mathematics, ANU from 2001 to 2008 before returning to UNSW in 2008. Prof. Arns is the recipient of numerous awards, including a Distinguished Technical Achievement Award of SPWLA and the 2015 Premier’s Prizes for Science and Engineering: Leadership in Innovation in NSW as well as a postdoctoral fellowship, a research fellowship, and future fellowship of the Australian Research Council. He holds a Diploma degree in physics from the U. of Technology Aachen, Germany and a Ph.D. degree in petroleum engineering from the U. of New South Wales, Australia.
Professor Sheikh S Rahman is the discipline leader in drilling and production engineering for both conventional and unconventional reservoirs at the School of Minerals and Energy Resources Engineering. He joined the School in 1989 and has held different academic positions. Prior to joining UNSW, he served the petroleum industry for 12 years. He pioneered the application geomechanics in well design and hydraulic fracture treatment design and next generation fractured reservoir engineering. He engineered the multiple horizontal wells construction from a mother well to reduce foot print (in coal seam gas development next to the farming land) and steer through ultrathin coal seams in NSW, Australia. His innovative drilling fluids, cement, casing, drill string design have been serving as the vital resource material for major operating companies. He also developed the first fully coupled geomechanical simulator for fractured reservoir based on flow through discrete fractures in hydro-, thermo- and -chemical frame work. Several operators including ONGC, Petrovietnam have been using it to evaluate production potential from their fractured basement and limestone reservoirs. His other interests include ion tuned water flooding, application of AI in screening EOR/IOR candidate reservoirs, coal seam gas development by biogenesis and developing bacteria phage for controlling sulphate reducing bacteria. His vast publications and industry consortia served as testimonies of his career achievement.
Ji-Youn Arns currently works as Digital Imaging Scientist and Senior Advisor at CJEL Imaging Technology Pty. Ltd. She has more than 15 years working experience in Digital Imaging including (a) X-ray CT imaging acquisition as an instrumental scientist, (b) digital core analysis including pore network extraction and pore-scale network modelling for multiphase flow, and direct image-based calculations for resistivity index and relative permeability, (c) radiation safety and equipment managements, and (d) large data management. With a BE from Seoul National University and a Ph.D. from UNSW (2002), Ji-Youn was postdoctoral researcher at ANU from 2006–2009, UNSW 2009–2013, Soil scientist at the Helmholtz Centre for Environmental Research (UFZ) in Halle, Germany, 2012–2014, and Instrumental Scientist and Tyree micro-CT laboratory manager 2014–2017.
This year’s awarded paper introduces a convolutional neural network reconstruction method to reconstruct high-resolution porous structures based on low-resolution µ-CT images and high-resolution scanning electron microscope (SEM) images. The proposed method involves four steps. First, a three-dimensional low-resolution tomographic image of a rock sample is obtained by µ-CT scanning. Next, one or more sections in the rock sample are selected for scanning by SEM to obtain high-resolution two-dimensional images. The high-resolution segmented SEM images and their corresponding low-resolution µ-CT slices are then applied to train a convolutional neural network (CNN) model. Finally, the trained CNN model is used to reconstruct the entire low-resolution three-dimensional µ-CT image. Because the SEM images are segmented and have a higher resolution than the µ-CT image, this algorithm integrates the super-resolution and segmentation processes. The input data are low-resolution µ-CT images, and the output data are high-resolution segmented porous structures. The experimental results show that the proposed method can achieve state-of-the-art performance.
3 In Closing
The Best Paper Award honors the efforts of authors who seek to achieve excellence in their research. Congratulations to the 2018 winners and a most sincere thanks for their contribution not only to Mathematical Geosciences, but also to the profession.