SolarisAI, a Queensland tech start-up affiliated with the University of Queensland, claims to have developed technology that could save the renewable energy industry $200 million annually. The innovation, led by Associate Professor Rahul Sharma from UQ’s School of Electrical Engineering and Computer Science, utilizes machine learning algorithms to analyze data and identify faulty or underperforming solar panels without the need for additional hardware installation.
Large solar farms face challenges in identifying issues within millions of panels, making manual inspection impractical. SolarisAI's technology operates at the array and string panel level, extracting vital information to monitor for degradation, soiling, wiring faults, and tracker problems. The system aims to automate the detection process and recommend targeted maintenance.
Dr. Sharma mentioned that underperformance in Australian solar farms currently costs the industry approximately $400 million annually. SolarisAI aims to reduce these losses by half and potentially increase revenue by up to eight percent. Discussions are underway to implement the technology at Edify Energy’s Hamilton solar farms and Genex Power’s Kidston solar farm in North Queensland.
Edify Energy CEO and founder John Cole expressed excitement about the project, emphasizing the importance of effective asset management for maintaining grid reliability and addressing the global energy crisis. SolarisAI collaborated with German-based electronics company Weidmueller to develop early prototypes. The project received support from UniQuest, UQ’s commercialization company, and investments from Uniseed and the UniQuest Investment Fund.
UniQuest CEO Dean Moss highlighted Australia's solar-generation capacity's potential for "stellar growth" and emphasized the commercial opportunity backed by top-notch research, with significant economic and environmental benefits expected.
