MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641072339 A) filed by Bontha Susmitha; and Dr. Putti Srinivasa Rao on June 11, 2026, for Integrated Xgboost–lstm Framework For Condition-Based Maintenance Of Rotating Machinery.
Inventors include Bontha Susmitha; and Dr. Putti Srinivasa Rao.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: Abstract A hybrid artificial intelligence system is disclosed for predictive maintenance of rotating equipment including turbines, pumps, compressors, motors, and gearboxes. The invention integrates eXtreme Gradient Boosting (XGBoost) for feature selection and Long Short-Term Memory (LSTM) networks for temporal modeling of vibration and sensor data. XGBoost identifies and ranks relevant features from high-dimensional sensor inputs, reducing overfitting and improving interpretability. The selected features are processed by LSTM networks to capture sequential degradation patterns and predict Remaining Useful Life (RUL) or fault occurrence. Experimental evaluation using benchmark datasets (NASA C-MAPSS) and real vibration sensor data demonstrates superior performance compared to standalone machine learning and deep learning models, achieving improvements in accuracy, precision, recall, F1-score, and Mean Absolute Error (MAE). The disclosed framework provides a scalable, interpretable, and robust solution for condition-based maintenance, enabling earlier fault detection, reduced downtime, lower maintenance costs, and enhanced operational safety.
Disclaimer: Curated by HT Syndication.
