AI solution optimises solar systems
SmartHelio's Autopilot analysis software, which was launched at the beginning of 2024, has optimised the performance of more than 200 solar systems over the course of the year. It is designed to extend their service life by eight years, increase production by 10 per cent and predict faults and failures.
SmartHelio has developed a predictive diagnostics solution for photovoltaic systems based on artificial intelligence (AI) and launched it on the market in early 2024. It bears the name Autopilot. Since then, this platform for controlling, operating and maintaining solar power plants has optimised the efficiency of 218 such systems with more than 2 gigawatt peak, according to the manufacturer. Their output has been increased by an average of 8 per cent.
According to the information provided, the plug-in system with more than 50 proprietary algorithms delivers 95 per cent accuracy from day one in detecting faults before they affect performance. According to SmartHelio, the one-time diagnostics process identifies remediable energy losses of up to 7 per cent on average. Data quality is improved by over 98 per cent.
The service life is increased by eight years, solar energy production by 10 per cent and the return on investment by up to 600 per cent. Above all, Autopilot significantly reduces diagnosis times, says CEO and founder Govinda Upadhyay: “From 358 hours to just 3 hours – the technology is helping us to redefine diagnosis and save solar plant operators time.” The company plans to expand the areas of application to battery storage and energy trading, thereby driving forward the energy transition worldwide.
SmartHelio is a spin-off of the Swiss Federal Institute of Technology Lausanne based in the canton of Vaud and operates a branch office in Uttar Pradesh, India. The company, which was founded in 2019, will be presenting its technology at the World Future Energy Summitfrom 14 to 16 January 2025. It will be one of eight exhibitors in the SWISS Pavilion as part of Abu Dhabi Sustainability Week.