MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202631069881 A) filed by Asansol Engineering College on June 04, 2026, for “ai-Based Fault Detection In Electrical Distribution Networks”.
Inventors include Dr. Apurba Chatterjee; Sumanta Karmakar; and Kaushik Neogi.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: Abstract An artificial intelligence-based fault detection, localization, and predictive maintenance system for electrical distribution networks utilizes a distributed network of intelligent sensors, smart meters, and IoT monitoring units to continuously capture high-frequency electrical telemetry parameters. Raw sensor streams are ingested by a signal preprocessing engine that executes noise filtering, signal normalization, harmonic analysis, and feature extraction to construct clean temporal- frequency feature vectors. These feature vectors are analyzed by an artificial intelligence engine combining convolutional neural networks, recurrent neural networks, support vector machines, and decision trees to identify abnormal operating conditions. A fault classification subsystem isolates the precise category, severity, and geospatial location of faults, including line-to-ground, short circuit, and transformer failures, while a predictive maintenance module projects equipment remaining useful life. The system interfaces with cloud networks and SCADA systems to automatically trigger protective relays and transmit real-time emergency notifications.
Disclaimer: Curated by HT Syndication.
