With cutting-edge, high-performance computing (HPC), efficient data center operation, and state-of-the-art data visualization, NREL played a critical role advancing energy efficiency and renewable energy science in Fiscal Year (FY) 2022.
NREL's HPC resources provided an essential foundation for scientists and engineers to rapidly assess and integrate data into their clean energy research.
Significant upgrades to NREL advanced computing capabilities include these featured projects. For more details and all upgrades, download the full report.
Kestrel Landing Soon
In summer 2023, NREL will install its third-generation HPC system—Kestrel—to enable emerging workflows for rapidly advancing artificial intelligence (AI) applications, driving complementary physics and data-driven approaches by fusing simulation with new sensor data sources.
New Visualization Capabilities Drive Discovery, Decision-Making, and Reasoning
NREL's Insight Center now features a multi-surface, multi-user space along with a 100-megapixel display for 2D interaction with data, and updates to our 3D immersive room more than doubled the resolution for enabling multiple users to explore virtual reality simultaneously from their own perspectives.
Building Trust in Artificial Intelligence To Ensure Equitable Solutions
To help deliver clean energy solutions to communities across the country and around the world, NREL's computational scientists are addressing potential bias in AI early in the research process, and incorporating equity into practices across research, development, demonstration, and deployment.
NREL's advanced computing research collaborations resulted in more than 300 completed projects, spanning across multiple disciplines and technologies as these featured highlights show. For more details, download the full report.
NREL's advanced computing staff recorded the following software developments and inventions in FY 2022.
RLC4CLR: Reinforcement Learning Controller for Critical Load Restoration Problems (SWR-22-27)
Hybrid-RL-MPC4CLR: Hybrid Reinforcement Learning Model Predictive Control for Reserve Policy-Assisted Critical Load Restoration in Distribution Grids (SWR-22-25)
MPC4CLR: Model Predictive Control for Critical Load Restoration in Power Distribution Systems (SWR-22-24)
Geothermal_osr: Energy Predictor for Geothermal Open-Source Reservoir (SWR-22-23)
VE: Virtual Engineering of Low-Temperature Conversion (SWR-22-19)
GANISP: A GAN-Assisted Importance Splitting Probability Estimator (SWR-22-10)
PowerGridworld: A Framework for Multi-Agent Reinforcement Learning in Power Systems (SWR-22-07)
Graph-Env: Graph Search as a Reinforcement Learning Problem (SWR-22-37)
JobQueue-PG: A Task Queue for Coordinating Varied Tasks Across Multiple HPC Resources and HPC Jobs (SWR-22-41)
BUTTER: An Empirical Deep-Learning Experimental Framework (SWR-22-42)
G2Aero: Separable Shape Tensors for Aerodynamic Applications (SWR-22-44)
PeleLMeX: An Adaptive-Mesh Low Mach Number Hydrodynamics Code for Reacting Flows (SWR-22-44)
MPRL: Multi-Pulse Reinforcement Learning (SWR-22-49)
L-Marshal: Python Lightweight Marshaler (SWR-22-55)
Mesoflow: A Mesoscale Modeling Tool for Heterogenous Gas-Solid Reacting Flows (SWR-22-56)
BDEM: Discrete-Element Simulator for High-Solids Granular Flows (SWR-22-72)
NOODLES: NREL Object-Oriented Data Layout and Exploration System (SWR-22-78)
LBC: Learning Building Control (SWR 22-21)
dsgrid: Demand-Side Grid Model (SWR 22-21)
ParaEMT: Parallelizable Large-Scale Power System Electro-Magnetic Transient Simulator (SWR-22-16)
ML4PD: Machine Learning for High-Throughput Process Design (SWR-22-62)