Data, Analysis, and Visualization

At NREL, scientific visualization and data analysis and management capabilities help move energy efficiency and renewable energy technologies from fundamental research to real-world application.

Text version

Our research into human-computer interaction, cognition, and immersive visualization enables us to make sense of data and communicate results while exploring the workflows that leverage unique capabilities of high-performance, cloud, and data-centric computing. We thrive on the increasingly data-intensive enterprise of transforming information into an interactive visual experience and the new world of research engagement that moves the needle on high-impact projects dealing with complex data sets.

Scientific Visualization

Located within the Energy Systems Integration Facility , NREL's Insight Center offers visualization facilities—the Visualization Room and the Collaboration Room—that allow users to walk into and through simulations and examine energy systems from new angles. The Insight Center experience supports knowledge discovery through dynamic interaction and exploration of extremely large, complex, experimental- or simulation-produced data.

Our world-class visualization experts bring data to life, applying best practices for data management, resolution, and format to ensure a successful translation into visual representations. We explore and implement new interaction and display technologies to support analytical reasoning and decision making. We research and design novel methods for visually conveying uncertainty and variability in data sets, erroneous or missing data, and data validation. We develop custom software for all stages of the visualization process, including data processing, innovative visualization dashboards, and data pathways to connect to the immersive displays.

Data Analysis and Management

NREL produces and manages tens of terabytes of data—and millions of related records—through experimentation, simulation, and observation methods. These methods are leading a paradigm shift that academic and industry collaborators leverage to advance their goals.

We use next-generation database clusters and storage systems and transform, translate, and process large-scale data sets to put them into an analysis-ready format. Our advanced analytics approaches ensure the visualizations we design answer the questions we have of the data. Qualitative and quantitative approaches to understand complex scientific data include cutting-edge methods in remote high-performance computing (HPC) visualization, visual mining, and visual exploratory tools.

Human-computer interaction, cognition, and immersive visualization are our research domains that enable researchers to make sense of their data and communicate their results. We empower social computing, learning and education, emergency planning and response, and integrated systems analysis through a variety of multimodal, context-aware interaction techniques.

Our advanced study on computers and algorithmic processes includes their principles, hardware and software designs, applications, and impact on society. We create systems, software, and tools to support the capture, analysis, and management of experimental, observational, and big data sources. Our expertise in this domain includes:

  • High-throughput and combinatorial experimental data (e.g., photovoltaics, bioenergy, materials)
  • Time-series data (e.g., observational and meter data management)
  • Real-time and streaming data (e.g., energy systems hardware-in-the-loop experiments)
  • Image data (e.g., microscopy and experimental X-ray diffraction and fluorescence)
  • Statistical modeling and prediction; statistical validation of models and prediction scenarios
  • Bioinformatics modeling and workflows
  • Computational approaches to image analysis and computer vision.

Featured Projects

Identifying Next-Generation Materials for Perovskites

Data analysis and visualization nodes on NREL's high-performance computer accelerate model training to identify improved materials for manufacturing perovskites. To investigate a class of redox-active, mixed ionic-electronic conducting metal oxides, HPC-enabled quantum-mechanic calculations model oxygen vacancy in perovskites containing three or more elements. Machine learning-trained and validated models predict oxygen vacancy formation energies in quaternary and quinary perovskites. This depth of material characterization informs strategies to boost solar-to-hydrogen thermal efficiency. DOE's Hydrogen and Fuel Cell Technologies Office funded this project.

2D and 3D Simulations for LED Manufacturing

NREL's high-performance computer is simultaneously tackling a critical manufacturing challenge for thin-film wafers in LEDs and a "grand challenge" computing problem for modeling magnetron plasma discharges at relevant densities in high-powered impulse magnetron sputtering (HiPIMS). HPC-powered simulations in both 2D and 3D are modeling how plasma behaves during the HiPIMS process, helping scientists determine how reactor geometry can be tailored to reliably create thin films with favorable electrical, optical, structural, and chemical properties. On this project, NREL partners with Lawrence Livermore National Laboratory and semiconductor chip manufacturer Applied Materials, with funding from DOE's Advanced Manufacturing Office.

Climate Change and Power Sector Operations

NREL's high-performance computer generated 700 individual-year simulations that captured multiple climate-forced water availability scenarios. This power systems model evaluated the impact of water availability and grid configurations, considering regionwide impacts as well as subregional responses to capture regional capacity differences and realistic grid connectivity dynamics. Researchers believe this work represents the largest set of power system simulations under climate-forced water constraints to date. NREL's HPC capability trimmed a traditionally 2-month processing time down to 24 hours. See the Environmental Science and Technology journal article Climate-Water Adaptation for Future U.S. Electricity Infrastructure for more information.

Publications

Transforming Energy Through Computational Excellence: Advanced Scientific Visualization Reveals Energy Insights, NREL Fact Sheet (2022)

Actionable Uncertainty Visualization, NREL Presentation (2022)

Noodles: Cooking Up Collaborative Visualization, NREL Presentation (2022)

MATBOX: An Open-Source Microstructure Analysis Toolbox for Microstructure Generation, Segmentation, Characterization, Visualization, Correlation, and Meshing, SoftwareX (2022)

Multiscale and Multivariate Transportation System Visualization for Shopping District Traffic and Regional Traffic, Transportation Research Record (2021)

Visualizing the Impacts of Renewable Energy Growth in the U.S. Midcontinent, IEEE Open Access Journal of Power and Energy (2020)

Data Storage and Visualization Solutions for Real Time Simulations and Experiments, NREL Presentation (2020)

Contact

Kristi Potter

Group Manager, Data, Analysis, and Visualization

Kristi.Potter@nrel.gov
303-275-4122

Share