Intellectual Property India Publishes Patent Application for Federated Learning-Based Decentralized Control System For Smart Grid Stability

MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202641100795 A) filed by B. L. Melvyn Rodger; Dr. K. Kalai Selvi; Dr. K. Senthil Kumar; Dr. Sivaranjanee R; Dr. Kvns Pavan Kumar; and Dr. Ramesh Pal on August 20, 2026, for Federated Learning-Based Decentralized Control System For Smart Grid Stability.

Inventors include B. L. Melvyn Rodger; Dr. K. Kalai Selvi; Dr. K. Senthil Kumar; Dr. Sivaranjanee R; Dr. Kvns Pavan Kumar; and Dr. Ramesh Pal.

The application for the patent was published on August 28, 2026, under issue no. 35/2026.

Abstract: The present invention relates to a Federated Learning-Based Decentralized Control System for Smart Grid Stability configured to monitor, predict, and mitigate instability conditions in a distributed electrical power network. The system comprises a plurality of distributed grid nodes, sensing and data acquisition modules, local processing units, machine-learning modules, a federated aggregation module, and decentralized control modules. Each grid node acquires electrical parameters including voltage, current, frequency, active power, reactive power, load demand, renewable-energy generation, and energy-storage parameters. The acquired data is processed locally and used to train a machine-learning model for identifying and predicting grid stability conditions without transmitting raw electrical data to a centralized processing system. Model updates generated at the distributed nodes are transmitted to the federated aggregation module, which combines the updates to generate an updated global model. The updated model is distributed to the participating nodes for improved local stability prediction. Based on the predicted stability condition, the decentralized control module generates corrective actions including active-power regulation, reactive-power control, battery charging or discharging, inverter control, renewable-generation regulation, and demand-response control. A feedback mechanism further enables the system to evaluate grid response and continuously improve the local and federated models. The system thereby provides adaptive, scalable, privacy-preserving, and resilient smart-grid stability control with reduced dependence on centralized data processing and continuous communication.

Disclaimer: Curated by HT Syndication.