MUMBAI, India, Aug. 17 -- Intellectual Property India has published a patent application (202541012102 A) filed by Greppipe Software Solutions Private Limited on February 12, 2025, for Machine Learning Based Dynamic Industrial Structural Failure Prediction System.
Inventors include Dr. Pravin Dinkar Nemade; Ms. Kaveri Ravindra Sonawane; Mr. Rahul Madhukar Jadhav; Ms. Monika B. Murkute; and Dr. Ajay Pandurangji Shelorkar.
The application for the patent was published on August 14, 2026, under issue no. 33/2026.
Abstract: A machine learning-based industrial structural failure prediction system monitors structural health in real time. It consists of a main body (101) that houses all components, with guide rails (103) ensuring linear motion of moving parts. A strain gauge (102) measures strain variations, linked to a pair of structure-holding arrangements (104) connected to the monitored structural member. The system features a wireless transmission module (105) for seamless data transfer and real-time display. A data processing module in the body employs the Hybrid LSTM-Autoencoder model to analyze strain data, detect anomalies, and predict failures. This automated monitoring system enhances safety, reduces maintenance costs, and ensures early failure detection in industrial applications, making it a reliable solution for predictive structural health assessment.
Disclaimer: Curated by HT Syndication.
