MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641108556 A) filed by Easwari Engineering College; and Srm Institute Of Science And Technology, Ramapuram Campus on September 10, 2026, for Predictive Maintenance Of A Motor-Bearing System With Fault Detection And Rul Prediction.
Inventors include Prem Kumar R S; Sabarivasan M; Raj Kumar K; and Dr. K G Ashok.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: A machine-learning-based predictive maintenance system for monitoring motor-bearing systems, detecting faults, and estimating remaining useful life (RUL) is disclosed. The system integrates an ESP32-based data acquisition unit with an ADXL345 tri-axial accelerometer, DS18B20 temperature sensor, and Hall-effect sensor for acquiring vibration, temperature, and rotational-speed measurements from a physical motor-rig. Acquired signals are processed through window segmentation and time-domain and spectral feature extraction. An Isolation Forest model is employed for anomaly detection, while XGBoost models are used for bearing fault classification and RUL prediction using CWRU bearing and NASA C-MAPSS FD001 datasets, respectively. The system further performs physical motor-rig classification using a 32-feature representation for Normal, Imbalance, and Misalignment conditions. A Streamlit dashboard presents processed machine-learning results, historical physical-rig analysis, and live raw telemetry, providing an integrated framework for motor-bearing condition monitoring, fault diagnosis, and prognostic assessment.
Disclaimer: Curated by HT Syndication.
