MUMBAI, India, Feb. 13 -- Intellectual Property India has published a patent application (202641009448 A) filed by Mr. Sreehari TM; Dr. Gomathy M; Mrs. Geetanjali R; Dr. Vignesh V; Ms. Jayashree M. Kudari; Ms. S. Susila Sakthy; Dr. N. P. Gopinath; Dr. B. Thyla; Ms. Ramya M; and Dr. Jose Anand A, Bangalore, Karnataka, on Jan. 29, for 'sustainable machine-learning-based demand forecasting and energy optimization system.'
Inventor(s) include Mr. Sreehari TM; Dr. Gomathy M; Mrs. Geetanjali R; Dr. Vignesh V; Ms. Jayashree M. Kudari; Ms. S. Susila Sakthy; Dr. N. P. Gopinath; Dr. B. Thyla; Ms. Ramya M; and Dr. Jose Anand A.
The application for the patent was published on Feb. 13, under issue no. 07/2026.
According to the abstract released by the Intellectual Property India: "The proposed machine-learning framework, using Random Forest, Gradient Boosting, and LSTM models, accurately predicts short-term energy demand, effectively capturing nonlinear and temporal patterns in electricity consumption. 2. Integration of ML-based forecasts into the optimization module reduces total energy costs by up to 12% compared to conventional baseline approaches. 3. The system increases the share of renewable energy in the overall supply mix by 8%, promoting cleaner energy consumption and reducing reliance on conventional grid power. 4. Optimized energy dispatch decreases carbon emissions by 7.3%, demonstrating environmental sustainability benefits alongside economic gains. 5. Sensitivity and trade-off analyses show that the framework maintains reliable performance under fluctuating renewable generation and demand uncertainties, ensuring operational stability. 6. The system balances computational efficiency and accuracy-Random Forest supports real-time deployment, while LSTM captures long-term sequential patterns-making the framework practical and scalable for modern smart grid applications."
Disclaimer: Curated by HT Syndication.
