MUMBAI, India, Feb. 6 -- Intellectual Property India has published a patent application (202541126598 A) filed by Krishna Pavan Inala; and B. V. Raju Institute Of Technology, Narsapur, Telangana, on Dec. 14, 2025, for 'a federated learning approach for distributed anomaly detection in v2g communication networks.'
Inventor(s) include Krishna Pavan Inala; and Rampelli Manojkumar.
The application for the patent was published on Feb. 6, under issue no. 06/2026.
According to the abstract released by the Intellectual Property India: "This invention presents a federated learning-based distributed anomaly detection framework for secure vehicle-to-grid (V2G) communication in smart grids. The system enables electric vehicles, charging stations, and aggregators to collaboratively learn normal and abnormal operational behaviours without sharing raw data. Local anomaly detection models are trained at charging station controllers using power flow, voltage, and communication parameters. Model updates are securely transmitted to a federated learning server hosted at the utility level. The server performs global model aggregation to generate a system-wide anomaly detection model. The aggregated model is redistributed to all participating nodes for real-time deployment. The framework detects cyber-attacks, communication failures, and abnormal charging or discharging behaviour. Privacy preservation and reduced communication overhead are ensured through federated learning. The proposed approach is scalable to large EV populations. Overall, the invention enhances grid reliability, cybersecurity, and operational resilience in V2G-enabled smart grids."
Disclaimer: Curated by HT Syndication.
