AI-BASED PREDICTIVE MAINTENANCE AND CUTTING FORCE MONITORING SYSTEM USING MACHINE LEARNING

Author

L Rambabu, G Mohan, CH Bhanu Prakash, Dr. Vijaybabu

Abstract

 Unexpected failures of rotating machinery cause damage, lost production time and higher maintenance cost. This paper presents an AI-based predictive maintenance and cutting force monitoring system built on a laboratory-scale cutting rig with a 12 V DC motor, a rotating shaft, two pillow-block bearings and a cutting blade. Motor current, sensed by a current sensor, serves as an indirect indicator of cutting load, while a tri-axial accelerometer and K-type thermocouples on each bearing record vibration and temperature. An ESP32 microcontroller acquires all signals and passes them to a computer, where the data are stored in CSV format and analysed in Python. A balanced data set of 60,000 samples was collected for four bearing-health combinations (good/bad) at five load levels (0–100 %). Logistic Regression, Support Vector Machine (SVM), Decision Tree and Random Forest models were compared using RMSE and R². The Decision Tree and Random Forest models predicted bearing condition with testing R² of 0.93–0.99 for individual bearings and about 0.95 for the combined condition, clearly outperforming SVM and Logistic Regression. Vibration and temperature proved to be the informative signals; the motor current showed no measurable change with load or bearing condition in the collected data.

Keywords

Predictive maintenance, cutting force monitoring, machine learning, condition monitoring, current sensing, ESP32, Random Forest, bearing fault, IoT.

Full Text:

Download Paper PDF

References

[1]  Raza, A. and Aslam, M. (2025). Artificial intelligence for predictive maintenance in industrial automation.

[2]  Ali, S. and Zafar, M. (2020). Challenges of implementing artificial intelligence in Pakistani industries.

[3]  Khan, A. and Siddiqui, F. (2021). Neural-network-based anomaly detection for industrial systems.

[4]  Ahmed, S. and Javed, M. (2021). AI-based real-time fault detection in industrial environments.

[5]  Tariq, M. and Farooq, A. (2022). Data preprocessing and IoT integration for industrial predictive maintenance.

[6]  Khan, A. and Tariq, M. (2022). Feature engineering for industrial machine-condition monitoring.

[7]  Siddiqui, F. and Malik, S. (2022). Sensor fusion and data quality in industrial predictive maintenance.

[8]  Hassan, A. and Khan, M. (2023). AI-based vibration analysis for predictive maintenance in the automotive sector.

[9]  Ali, S. and Hassan, A. (2023). Integrating modern AI technologies with legacy industrial systems.

[10]  Raza, A. and Aslam, M. (2025). Review of machine-learning, deep-learning, sensor-fusion and IoT approaches for industrial fault diagnosis.

Share your valuable work from Social Media Buttons