MUMBAI, India, May 29 -- Intellectual Property India has published a patent application (202641063791 A) filed by Sr University, Warangal, Telangana, on May 20, for 'an attention-based deep learning system for ultra-short-term solar irradiance forecasting.'
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: "Disclosed herein is an attention-based deep learning system (100) for ultra-short-term solar irradiance forecasting using sky imaging and meteorological data, the system (100) comprises a sky imaging module (102) comprising a ground-based sky camera. The system also includes a meteorological data acquisition module (104) configured to collect meteorological parameters. The system also includes an image processing module (106) configured to pre-process the captured sky images. The system also includes a deep learning prediction module (108) comprises a visual encoder, a temporal learning module and an attention mechanism module. The system also includes a multi-modal fusion unit (110) configured to combine extracted visual features. The system also includes a prediction module (112) configured to generate ultra-short-term solar irradiance forecasts. The system also includes an explainability module (114) configured to generate attention maps and feature-importance representations. The system also includes a computing unit (116) comprising graphical processing unit (GPU)."
Disclaimer: Curated by HT Syndication.
