MUMBAI, India, July 7 -- Intellectual Property India has published a patent application (202641076510 A) filed by Muthayammal Engineering College Autonomous on June 20, 2026, for Self-Adaptive Failure Prediction Framework Using Federated Machine Learning In Industrial Networks.

Inventors include Mrs. N. Lalithakumari; Mr. S. Ananda Kumar; Dr. M. Mala; Mr. B. Prasanna; Vishali Sy; Vinothini P; Yuvabharathi; Nandhadevi V S; Buvaneshwari A; Arthi A; and Akash A.

The application for the patent was published on July 03, 2026, under issue no. 27/2026.

Abstract: The present invention relates to a self-adaptive failure prediction framework utilizing federated machine learning in industrial networks. The framework comprises an industrial sensing layer, distributed edge intelligence architecture, local model training subsystem, federated learning coordination engine, self-adaptive model evolution engine, failure forecasting processor, remaining useful life estimation subsystem, maintenance intelligence module, and visualization platform. Operational information generated by industrial assets is processed locally to develop facility-specific predictive models. Model updates are securely aggregated through federated learning mechanisms without transferring raw operational datasets, thereby preserving data privacy and confidentiality. The self-adaptive model evolution engine continuously refines predictive intelligence according to changing equipment conditions, environmental influences, operational workloads, and maintenance interventions. The framework forecasts equipment failures, predicts degradation trajectories, estimates remaining useful life, evaluates operational risks, and generates predictive maintenance recommendations. The invention improves prediction accuracy, enhances industrial reliability, reduces downtime, optimizes maintenance planning, minimizes operational costs, and enables privacy-preserving collaborative intelligence development across distributed industrial environments.

Disclaimer: Curated by HT Syndication.