MUMBAI, India, Jan. 23 -- Intellectual Property India has published a patent application (202641001905 A) filed by Muthayammal Engineering College, Namakkal, Tamil Nadu, on Jan. 7, for 'method for predicting cascading failures in complex networks using a spatiotemporal graph convolutional network.'
Inventor(s) include Dr. G. Kavitha; S. Uma Maheswari; Sompalli Yugesh; Thammineni Vamsi Krishna; Thanappa Surya; and Thonnati Damaraselvam.
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: "The invention suggests a process and a mechanism of early detection and prediction of the cascading failure processes in complex systems such as power systems, telecommunication and financial systems. The invention involves Spatiotemporal Graph Convolutional Network (ST-GCN) to acquire the information about intricate relationships between network topology (spatial) and the way states of nodes evolve through the course of time (temporal). The scheme is aware of nodes that may be at risk of failure before a chain reaction has started, and its mitigation is automatic through load-redistribution or isolation. This approach is far more efficient in the context of increasing the resilience of the network when compared to the traditional reactive models since it provides a proactive vulnerability map of the entire system."
Disclaimer: Curated by HT Syndication.
