MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541122215 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Dec. 4, 2025, for 'hybrid ensemble learning system for machine failure prediction using sensor data.'
Inventor(s) include Dr. K. Ragavan; and Sarath Kumar S.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "The present disclosure provides a computer-implemented method (100) for predicting machine failures in industrial equipment using hybrid ensemble learning. The method (100) includes acquiring sensor data from industrial machinery, including operational parameters such as air temperature, process temperature, rotational speed, torque, and tool wear measurements (102). The method (100) performs data cleaning including handling missing values and encoding transformations (104, 106), detects class imbalance in target failure classifications (108), and applies Synthetic Minority Over-sampling Technique (SMOTE) when imbalance exceeds a predetermined threshold (110). The method (100) performs feature scaling (114), splits data into training and testing sets (116), applies recursive feature elimination (118), and trains multiple stacking ensemble models (120). Each model includes different combinations of base models selected from Long Short-Term Memory networks, Random Forest classifiers, Gradient Boosting machines, K-Nearest Neighbors, Support Vector Classifiers, Gaussian Naive Bayes, and Multi-Layer Perceptron, with a meta-model selected from XGBoost, Random Forest, and CatBoost. The method (100) selects the highest accuracy model for deployment."
Disclaimer: Curated by HT Syndication.
