The SRT team along with our development partners at Eni have been very busy during the last year. Despite the pandemic, we've been focusing on advancing the industry’s most unique and powerful platform for petroleum reservoir simulation.
The SRT team along with our development partners at Eni have been very busy during the last year. Despite the pandemic, we've been focusing on advancing the industry's most unique and powerful platform for petroleum reservoir simulation. ECHELON 2.0 is the most important release of ECHELON software yet and a major advance for the industry. It offers the first full GPU solution of compositional formulation, both fully implicit and adaptive implicit options. The performance and scaling of our compositional formulation is unlike anything else available on the market. In addition to full compositional support, ECHELON software 2.0 includes support for facility network modeling, multi-reservoir coupling and advanced well management. ECHELON 2.0 is:
While many in the industry were questioning the efficacy of GPU for more advanced compositional simulation, we at SRT along with our partners at Eni have been busy implementing it. Our compositional formulation is fully GPU-enabled and has demonstrated exceptional performance on the assets we've tested. ECHELON software running on 1 GPU is consistently faster than all market leading CPU or hybrid simulators using 100's of cores. Our results on NVIDIA AMPERE are even better than we had expected by comparing hardware specifications.
After two years of development we are proud to announce FNS, ECHELON's own facility network solver. FNS brings surface network modeling capabilities to ECHELON software at performance levels you’ve come to expect from SRT.
Robust reservoir coupling strategies have been added to ECHELON software to address complex models that are efficiently coupled with minimal and intuitive user settings. A new reservoir simulation management strategy has been implemented to decouple the field management aspects of the coupling from that of the individual reservoirs.
Complex user-defined dynamic quantities can now be requested for output and for group and well management functionalities.
Significant improvements have been made to modularize the I/O operations and data structures without losing performance and scalability. These data structures streamline the input data allowing for more flexibility in addressing new features both for structured and unstructured meshes, and help with seamless handling of new input formats (for example, RESQML, user-defined scripts etc.)
Significant improvements have been made to ECHELON software's restart capabilities to improve coverage, flexibility and robustness.
New options for permuting cells for efficient preconditioning of models with very high vertical transmissibilities have been implemented. For many legacy models, high vertical transmissibilities is the norm rather than an exception. With this new option, large improvements in performance have been obtained and our performance numbers look even better compared to other simulators.
Contact us for more information about the features in this exciting new release, or learn more about ECHELON reservoir simulation software capabilities.
Dr. Mukundakrishnan is the Director of Research and Development at Stone Ridge Technology. Prior to joining Stone Ridge Technology, he was a R&D Technology Manager at Dassault Systemes Simulia (formerly ABAQUS Inc.). He has more than 15 years of experience working on HPC codes and published extensively in the areas of fluid dynamics, multiphysics coupling, and reservoir simulation. Dr. Mukundakrishnan obtained his M.S. and Ph.D. in Mechanical Engineering and Applied Mechanics from the University of Pennsylvania.
What we are doing to help improve the reservoir simulation industry.
Leveraging modern GPUs and neural computing libraries, scientists are using deep learning technology to solve differential equations that emerge from a diverse set of physics-based problems. This blog introduces the topic and explores potential applications and limitations of the technology.Read article →