MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202531133708 A) filed by Kalinga Institute Of Industrial Technology, Bhubaneswar, Orissa, on Dec. 30, 2025, for 'federated learning-based fault detection system for power transmission lines using synchrophasor data.'
Inventor(s) include Jha, Amit Kumar V.; Appasani, Bhargav; Bhowmik, Wriddhi; and Mishra, Sunil Kumar.
The application for the patent was published on Jan. 23, under issue no. 04/2026.
According to the abstract released by the Intellectual Property India: "A federated learning-based fault detection system for power transmission lines comprises Phasor Measurement Units (PMUs) distributed across a power transmission network, each acquiring real-time voltage and current phasor measurements. A local processing module associated with each PMU performs preprocessing including noise filtering, normalization, and feature extraction, trains a local fault detection model, and generates encrypted model parameters. An aggregator receives encrypted model parameters from the PMUs without receiving raw data, updates a global fault detection model using an adaptive weighting mechanism based on data quality and freshness, and distributes the updated model to each PMU. Each PMU performs local inference using the updated global model to detect and classify faults, and the system supports asynchronous updates such that each PMU transmits parameters independently without requiring synchronization."
Disclaimer: Curated by HT Syndication.
