MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641061668 A) filed by Sr University, Warangal, Telangana, on May 14, for 'a transformer-based deep learning system for accurate solar radiation prediction under dynamic atmospheric conditions.'
Inventor(s) include Bongoni Naresh; and Dr. Amit Kumar Yadav.
The application for the patent was published on May 29, under issue no. 22/2026.
According to the abstract released by the Intellectual Property India: "A Transformer-based deep learning system for accurate solar radiation prediction under dynamic atmospheric conditions is disclosed. The system comprises a data acquisition module configured to collect meteorological and atmospheric data, a preprocessing module configured to normalize and synchronize the collected data, and a sequence formation module configured to generate temporal input sequences. A Transformer-based deep learning engine employing multi-head attention mechanisms processes the sequences to capture long-range temporal dependencies and atmospheric feature interactions. A dynamic atmospheric adaptation module modifies prediction behaviour in response to changing weather conditions including cloud movement, aerosol variation, and humidity fluctuations. An explainability module generates attention-based interpretations identifying influential atmospheric parameters and temporal intervals contributing to prediction outputs. The disclosed system provides accurate, adaptive, and interpretable solar radiation forecasting suitable for photovoltaic optimization, renewable energy management, and smart-grid applications."
Disclaimer: Curated by HT Syndication.
