Stony Brook, New York, December 18, 2025 /PRNewswire/ — Stony Brook University researchers Yue Zhao and Kang Pu collaborate with Ecosuite's John Gorman and Philip Court to leverage historical datasets provided by Ecogy Energy —We have devised a data-driven algorithm to detect physical anomalies in solar energy systems. This research aims to reduce operation and maintenance (O&M) costs of solar power projects by understanding and predicting long-term weather-related and inverter issues.
stony brook university logo
Through this effort, researchers trained an anomaly detector through a self-supervised learning approach trained on inverter and weather data. As such, with a comprehensive data-driven pipeline, the project not only employs robust machine learning approaches to model complex system behavior, but also ensures the ability to operate across diverse data environments by choosing widely available power generation and weather data over non-standard means.
The paper also focuses further on long-term abnormalities that elude many asset managers. Applying anomaly detectors to solar power and weather data updates can help asset managers accurately predict and diagnose potential long-term physical problems weeks or even years in advance if possible.
“…More accurate and timely awareness will greatly improve the efficiency and effectiveness of O&M practices. For example, a) maintenance visits, a key component of O&M costs, can be more efficiently scheduled to address detected underlying problems. b) On-time maintenance of hardware will significantly extend lifespan and reduce the need for expensive replacements. c) Loss of energy production due to unresolved system issues will be significantly reduced.”
With this increased predictive knowledge of solar systems, these anomaly detectors can significantly reduce O&M costs, which are a key component of project economics in solar power development. There is great potential to optimize maintenance times and visits and reduce associated energy downtime.
“It makes sense to have advanced warnings about DER assets directly from the edge computing hardware you already pay for.” Researcher John Gorman commented: “Adding these software superpowers creates immediate value, but as part of a flexible ecosystem, our machine learning algorithms can also evolve. Being able to translate learning from one system to the next is our next goal, allowing us to unlock value across our portfolio.””
About Stony Brook University
Stony Brook University, the flagship of the State University of New York (SUNY) system, is a premier public research institution known for world-class faculty and groundbreaking discoveries. Professor Yue Zhao's laboratory conducts research in a wide range of areas of machine learning, optimization, and control with applications in power systems, power markets, renewable energy, energy storage, demand response, electric vehicles, and more.
About EcoSuite
Ecosuite is a DER asset management technology platform focused on unlocking grid value from distributed energy assets. Our award-winning AI edge computing solutions and open source digital infrastructure enable secure DER integration, visibility, and optimization for asset owners and utilities.
About Ecogy Energy
Ecogy Energy is a distributed energy developer and operator with a national portfolio of renewable energy assets and is committed to accelerating the energy transition through community-centered development.
contact:
stony brook university
[email protected]
[email protected]
eco suite
[email protected]
ecosuite.io
eco energy
[email protected]
ecoenergy.com
source eco suite
