
ASU team named winner in the American-Made 3D Solar Visibility Prize
PAL Lab develops AI-based tool to support grid reliability

A team of Arizona State University researchers from the Phasor Assisted Learning Lab, or PAL Lab, was named one of three winners and awarded a $50,000 cash prize in the American-Made Data-Driven Distributed Solar Visibility Prize.
The competition is designed to incentivize innovators to build models and algorithms that provide precise real-time insights about distributed solar generation in electric power networks. The prize aims to spark awareness and adoption of distribution system state estimation, or DSSE, algorithms and software tools to improve the visibility of distribution systems and their operating conditions.
Led by Anamitra Pal, an associate professor in the School of Electrical, Computer and Energy Engineering, part of the Ira A. Fulton Schools of Engineering at ASU, the PAL Lab performs fundamental and applied research in the areas of critical infrastructure resilience, smart grids and the application of artificial intelligence, or AI, in power systems.
“My group has been working on topics related to this competition for over five years now and has received support from multiple federal institutions while demonstrating successes using actual power system data,” Pal says. “Receiving this prize is an icing-on-the-cake moment for us and an acknowledgement of the cutting-edge research that my group has been doing at ASU.”
The team included Fulton Schools graduate students Mohammad Golgol, Shiva Moshtagh and Kaustav Chanda. Golgol and Moshtagh, both electrical engineering students, serve as research associates in the PAL Lab, while Chanda, a computer engineering student specializing in computer systems, contributed his knowledge in the area. Together, the team developed a highly accurate algorithm and accompanying tools to detect the amount of solar power in the electricity grid at given points in time. By developing these tools as part of the competition, the team is playing a vital role in catalyzing research and development in energy grid modernization. Their work will enable utility providers and grid management operators to make better-informed decisions that optimize the distribution of solar energy technologies.
“The power grid is evolving rapidly, with an increasing shift toward distributed energy resources like rooftop solar panels and electric vehicles,” Moshtagh says. “Research like this ensures that as the grid modernizes, it remains resilient, efficient and reliable for everyone.”
Eyes on the prize
The new competition, funded by the U.S. Department of Energy’s Solar Energy Technologies Office and administered by the National Renewable Energy Laboratory, or NREL, was created to better understand distributed solar energy generation data.
Competitors were tasked with developing a DSSE tool and supplied sample data to train their respective tools. After development was complete, teams submitted results obtained using their DSSE tool for two distribution system networks through the Open Energy Data Initiative Solar Systems Integration Data Modeling, or OEDI SI, platform for 14 consecutive days. The OEDI SI team then compared each DSSE tool’s performance against a set of industry-standard metrics to determine the top-performing teams.
“Our team worked vigorously in the weeks leading up to the start of the competition by creating and continuously fine-tuning our AI-based DSSE tool,” Pal says. “The students were instrumental in both training and fine-tuning the tool and ensured daily uploads were successfully completed before the cut-off time was reached during the testing phase of the competition.”
At the end of the 14-day testing period, the PAL Lab team finished atop the leaderboard, achieving the highest cumulative score of 35 competing teams.
Keeping the lights on
According to the NREL, the U.S. energy grid is very reliable. The average U.S. customer loses power fewer than two times annually for a combined downtime of less than five hours, which represents 99.95% reliability. When outages do occur, it is rarely due to insufficient power generation — most outages are related to issues in the distribution system.
Intermittent renewable energy sources like solar are dependent on natural cycles, which creates the need for supplemental monitoring and management to ensure continued grid availability. As the utilization of solar power accelerates, the need to support energy grids with effective tools to mitigate disruptions is critical to maintaining functional and reliable service to the public.
“The goal of my research group has always been to keep the lights on,” Pal says. “In order to attain this goal, we leverage AI and robust statistics to create new data analytics tools that tackle power system problems for which solutions currently do not exist.”
By utilizing data from existing infrastructure, the PAL Lab team will supply an accurate and cost-efficient monitoring tool, which aims to improve situational awareness across the distribution system. This awareness is pivotal for improving methods to ensure system resiliency.
“This type of project is crucial because it addresses real-world challenges that can directly impact people’s daily lives,” Golgol says. “The research aims to enhance the efficiency and reliability of power systems, leading to more stable electricity delivery to homes and businesses.”
Moving forward, Pal and his team will continue their research efforts in areas such as power system monitoring, protection and control, renewable generation integration, and energy modeling in smart grids.
“The success of the team proves that, when employed correctly, AI can successfully solve complex power system problems,” Pal says. “It is expected that this success will pave the way for future research, development and demonstration projects that my group will lead using both federal funding and industry sponsorship.”