NVIDIA Partners in Energy

Fluidized catalytic cracking units (FCCU) are at the heart of modern refineries, processing a wide variety of feedstocks into fuels and chemicals. FCCUs operate 24/7 for years on end, so small improvements in reliability and performance have a tremendous economic impact. CPFD’s Barracuda Virtual Reactor software is used to address FCCU-specific issues, such as afterburn, emissions, catalyst losses, erosion/reliability, catalyst circulation stability, gas/catalyst maldistribution, pressure balance issues, feed distribution, riser performance, throughput limitations, and the effects of co-processing renewable feedstocks. Virtual Reactor simulation reduces a refiner’s risk when making changes to the FCCU and enables technology providers to bring new solutions to market faster, at lower total cost.

Fluidized bed processes are at the heart of myriad sustainability and decarbonization applications, such as the chemical recycling of plastics, biofuels, waste-to-energy/fuels/chemicals processes, carbon capture, and the hydrogen economy.  Engineers, researchers, and innovators are urgently working to scale existing technology and develop new approaches to address some of our planet’s most pressing challenges. Barracuda Virtual Reactor simulation leverages NVIDIA GPU and multi-GPU parallelization to enable and reduce the time to market for multiple sustainability technologies that are changing our planet for the better.

Data analytics teams are developing AI models to automate power plant and oil rig inspections, optimize fuel shipping, forecast energy demand, and manage distributed energy resources. AES, a leading energy company, built and deployed over 50 AI models in 2 years to optimize liquid gas shipping and logistics, predict when power generation equipment will need maintenance, guide fintech energy trades, make hydrology predictions, inform bids on power generation facilities, provide weather forecasting for utilities, and more. The company’s cloud infrastructure supports the diverse needs of data scientists, providing access to NVIDIA GPU-accelerated compute resources and tools within a centralized Domino Data Lab platform for frictionless enterprise MLOps.

Maintaining the integrity and security of pipeline infrastructure comes at a high cost to pipeline owners and operators. LivePIPE® II delivers a fully integrated and networked sensing solution, based on the world’s most advanced fiber optic distributed acoustic sensor (DAS), machine learning (ML), and AI technology. DAS can detect and locate pipeline leaks and intrusions faster and more accurately than legacy systems, simultaneously in multiple locations. Using the NVIDIA Jetson™ edge AI platform and NVIDIA CUDA®, Fotech is able to collect, label, and process up to 600MB of raw data per second to train and deploy AI models at the edge for inference. This increases DAS sensitivity to accurately detect leaks and intrusions, reduces false positives, and auto-generates alerts.

Reservoir engineers struggle with long runtimes for their simulation models. This limitation has shaped workflows by discouraging the use of fine-scale grids, ensemble methodologies, and more advanced physical models. Stone Ridge Technology (SRT) developed ECHELON, the world’s fastest reservoir simulation software, to overcome this bottleneck by optimally harnessing the massive parallelism of NVIDIA GPUs with advanced proprietary numerical algorithms and software architecture. The exceptional performance of ECHELON gives reservoir engineers the opportunity to simulate using fine-scale geological grids, the ability to estimate uncertainty more precisely by running large ensembles of reservoir models, and to address the advanced demands of emerging green technology workflows. ECHELON is a full-featured, industrially hardened, numerical reservoir simulator,  created to make full use of GPU capabilities and resources, optimizing for memory bandwidth, storage, and floating point throughput (FLOPS). In 2018, SRT began collaborating with Eni S.p.A to further develop ECHELON and to deploy it throughout the supermajor's asset portfolio. ECHELON is commercially available through purchase, lease, or SaaS licensing options.

As renewable energy demands grow, offshore energy operators have an expanding network of subsea power cables to monitor and maintain. Preserving the integrity and security of subsea cable, inter-arrays, and cable protection systems (CPS) infrastructure comes at a high cost to cable owners and operators.

LiveDETECT® II provides real-time, automated continuous monitoring of subsea telecoms cables through a fully integrated sensing solution based on advanced fiber optic distributed acoustic sensing (DAS), machine learning, and AI technology. This reduces the downtime of cable infrastructure, lowers associated maintenance costs, and enhances business impact.

Monitoring with DAS specifically designed and optimized for wind farm operators provides early warnings of likely failures. This enables customers to monitor, detect, and predict cable faults and threats at precise locations to preserve uptime and maintain production assurance and supply.