April 29, 2024

AI for Energy Report Features PNNL Expertise

PNNL computing experts Robert Rallo and Court Corley contributed their knowledge in AI to a recent DOE report

Headshots of Robert Rallo and Court Corley superimposed on a background image of powerlines.

Robert Rallo (left) and Court Corley (right) add their AI expertise to a recent DOE report on Advanced Research Directions in AI for Energy.

(Composite image by Shannon Colson | Pacific Northwest National Laboratory)

Artificial intelligence (AI) has the power to transform nearly all aspects of daily life—from accelerating scientific research through autonomous experimentation to increasing protection from chemical threats.

As noted in a recent Department of Energy (DOE) report, Advanced Research Directions on AI for Energy, the energy sector provides an enormous opportunity for AI-enabled advancement. Pacific Northwest National Laboratory (PNNL) researchers Court Corley and Robert Rallo contributed their AI expertise to this report. The report is an outcome from the DOE winter 2023 workshop series on AI for energy hosted by Argonne National Laboratory (ANL). Scientists from DOE and many national laboratories, including PNNL and ANL, contributed to the report.

“AI can help protect the electric grid from cybersecurity attacks and make real-time operations more efficient,” said Corley, who leads the Center for AI @PNNL. “However, we need to make sure that AI is trustworthy and safe before we deploy these technologies on a large scale.”

The development of new materials for energy applications can also be enhanced with the help of artificial intelligence.

“Many aspects of materials research can be improved with AI,” said Robert Rallo, director of the Advanced Computing, Mathematics, and Data Division at PNNL. “With the right AI models, we could cut computational costs needed for accurate models of materials, enable faster predictions for the design of new materials, and support the autonomous synthesis of these materials.”

PNNL’s historic strengths in chemistry and materials sciences empowers researchers to combine these topics with next-generation technologies. PNNL projects and initiatives, such as Adaptive Tunability for Synthesis and Control via Autonomous Learning on Edge (AT SCALE) and the Computational and Theoretical Chemistry Institute (CTCI), combine domain science, data science, and computer science to pioneer novel discoveries in energy storage materials and catalysis. These discoveries are enabled with the help of advanced technologies, such as those developed and assessed by PNNL’s Center for Advanced Technology Evaluation and Co-Design Center for Artificial Intelligence-Focused Architectures and Algorithms.

“Though AI has the potential to revolutionize energy through many different aspects, we need more research on AI itself to verify and validate the technology,” said Corley. “Advancing AI through the mathematics and algorithms that underpin these technologies is critical to developing explainable, robust, and trustworthy systems."

In addition to this report, PNNL researchers Anurag Acharya, Anastasia Bernat, Sarthak Chaturvedi, Mahantesh Halappanavar, Sameera Horawalavithana, Phan Hung, Derek Lilienthal, Ann Miracle, Sai Munikoti, Dan Nally, Gihan Panapitiya, Mike Parker, Karl Pazdernik, Shivam Sharma, and Sridevi Wagle contributed to the DOE-issued report, AI for Energy: Opportunities for a Modern Grid and Clean Energy Economy. DOE’s Office of Electricity and Yousu Chen, Xiaoyuan Fan, Renke Huang, Qiuhua Huang, Ang Li, and Kishan Guddanti from PNNL issued a report that provides a foundation for understanding the transformative role of AI and ML in power systems. 

This work supports DOE priorities in advancing in AI safety and security as well as finding novel solutions for sustainable energy. Read more in the DOE announcement on AI.

To learn more about PNNL’s research in AI, please attend the presentations by Rallo and Corley at the upcoming AI Expo 24 on May 7-8 in Washington, D.C.