3rd Workshop on Challenges in Next-Generation Analytics for the Future Power Grid
3rd Workshop on Next-Generation Analytics
for the Future Power Grid
Richland Washington, USA
July 17-18, 2014
Researchers developing new analytic methods and models for the future power grid are increasingly looking to advanced computing to enable higher resolution simulations and faster solutions for operation and planning. Pacific Northwest National Laboratory (PNNL) believes that this research—funded in large part by federal agencies—can be more productive and impactful if efforts are guided and coordinated by an accepted, grid-specific development architecture that includes a data and software structure, open-source libraries and visualization frameworks to translate data into actionable information. This is consistent with a desire on the part of the Department of Energy (DOE) to see more coordination across the community of laboratories, universities, vendors, and utilities to increase the benefit of federal funds invested in R&D.
PNNL and other national laboratories and universities are creating elements of new software to support next-generation power delivery systems. Often, these developments are pursued in isolation and are not well positioned to demonstrate the impact and value which will eventually be delivered. PNNL's Future Power Grid Initiative (https://gridoptics.org) is hosting a workshop on July 17-18 at Washington State University's Tri-Cities campus in Richland, Wash., to foster community and accelerate deployment of new solutions. All persons who are developing or are interested in the development of shared-source software for the power grid are invited to participate.
As currently planned, the workshop will focus on several key problem areas with associated objectives, and will include:
- Identification of a small number of key "challenge problems" which will demonstrate value. We expect these problems will span areas from data collection and management, mathematics to analysis, prediction and control, computational methods, and techniques for support of human decision making. We plan a panel and breakout session to propose and harden these ideas. We invite attendees to offer up challenges, and at the workshop we will self-organize to discuss some of these proposals.
- Identification of the portfolio of software components and human expertise which are available in shared-source form to enable demonstrations to be constructed for potential challenge problem solutions. We propose sessions where participants share the current software components with an emphasis on understanding the key capabilities and how that software could be integrated into a larger demonstration.
- An expert-led panel to share understanding of the uses and importance of data in the power industry, with discussion on specific topics such as types of data that is collected; sensitivity categorization for different data types; storage requirements in different environments, both in volume and speed; coordination needs during routine and emergency operations (e.g., shared visualization); data sharing needs; integrating new research tools in “engineering” environments side-by-side with vendor developed tools; anonymizing/desensitizing operations data for use by research community; generating realistic synthetic data sets; or computation on data without the need to download data for analysis
- Identification of the next steps in organizing ourselves to advance community objectives. This will include both discussions of objectives for future workshops, and also best practices for sharing source code, governance of shared-code, and how to address challenges to improving the technical readiness of research software to prepare it for commercial adoption.
The workshop is inviting keynote speakers from industry and DOE to set context from their perspectives. A workshop report will be issued that captures the challenge problems and software portfolio for broader dissemination.
Pre-meeting tutorials on the afternoon of July 16 will be available for specific PNNL software. Post-meeting tours of PNNL facilities will be available on the afternoon of July 18. Specifics for both events will be listed on the registration site as details become available. Please note: Participation in the tours will require additional registration steps to gain access to PNNL facilities.
Workshop steering committee:
- Bora Akyol, Pacific Northwest National Laboratory
- Scott Backhaus, Los Alamos National Laboratory
- Anjan Bose, Washing State University
- Russell Robertson, Grid Protection Alliance
- Gil Bindewald, Program Manager, Advanced Grid Modeling Research Program, DOE Office of Electricity Delivery & Energy Reliability
- Larry Bekkedahl, Senior Vice President, Transmission Services, Bonneville Power Administration
- Jesse Berst, Founding Editor and Chief Analyst for Smart Grid News; Chairman, Smart Cities Council
- Researchers (National Labs, Universities, etc.): The workshop will provide opportunities to use open-source software to expedite research progress; to contribute to open-source software for broader recognition of your own research; and to publicize your own open-source software.
- Software Vendors: The workshop will help vendors stay abreast of the trends and available resources of open-source software for power grid operation and planning, and identify opportunities for early adoption of high-performance open-source software functions for future market positioning.
- Consulting Companies: The workshop will facilitate learning and use of open-source software codes for the most cost-effective execution of consulting tasks.
- Utility Companies: The workshop will provide opportunities to identify potential pilot applications for the best benefit of ratepayers, and to influence the development of open-source software packages to best fit the utilities’ own software environments.
For more details on workshop registration, please visit our Workshop Registration Page.
For additional information please contact:
David Callahan, Workshop Chair
Northwest Institute for Advanced Computing