PredX: An IoT and Unsupervised Machine Learning-Based Predictive Maintenance Framework for Brush DC Motor Condition Monitoring

PredX presents an end-to-end IoT-enabled condition-monitoring architecture for a high-speed brush DC motor. By combining MPU6050 vibration sensing, DS18B20 temperature measurement, INA219 electrical monitoring, Hall-based RPM measurement, ESP32 edge acquisition, MQTT telemetry, FastAPI processing, SQLite persistence, React visualization, threshold rules, and Isolation Forest anomaly detection, the prototype connects physical machine behavior with a practical software monitoring workflow. The primary research contribution at the current stage is architectural and demonstrative: a compact and reproducible pipeline from multi-sensor motor measurements to anomaly analysis and alert visualization. The scientific scope is deliberately limited to condition monitoring and unsupervised anomaly detection until controlled experiments produce ground-truth data and measured performance. Such validation will allow the study to quantify sensor stability, detection latency, false-alarm rate, anomaly-detection performance, and alert recovery behavior, and will provide a stronger basis for comparison with published predictive-maintenance systems.

[1] T. Akyaz and D. Engin, “Machine Learning-Based Predictive Maintenance System for Artificial Yarn Machines,” IEEE Access, vol. 12, pp. 125446–125461, 2024, doi: 10.1109/ACCESS.2024.3454548. [2] M. Çakır, M. A. Güvenç, and S. Mistikoglu, “The experimental application of popular machine learning algorithms on predictive maintenance and the design of IIoT based condition monitoring system,” Computers & Industrial Engineering, vol. 151, p. 106948, 2021, doi: 10.1016/j.cie.2020.106948. [3] K. S. Kiangala and Z. Wang, “An Effective Predictive Maintenance Framework for Conveyor Motors Using Dual Time-Series Imaging and Convolutional Neural Network in an Industry 4.0 Environment,” IEEE Access, vol. 8, pp. 121033–121049, 2020, doi: 10.1109/ACCESS.2020.3006788. [4] M. Calabrese, M. Cimmino, F. Fiume, M. Manfrin, L. Romeo, S. Ceccacci, M. Paolanti, G. Toscano, G. Ciandrini, A. Carrotta, M. Mengoni, E. Frontoni, and D. Kapetis, “SOPHIA: An Event-Based IoT and Machine Learning Architecture for Predictive Maintenance in Industry 4.0,” Information, vol. 11, no. 4, p. 202, 2020, doi: 10.3390/info11040202. [5] T. P. Carvalho, F. A. A. M. N. Soares, R. Vita, R. da P. Francisco, J. P. Basto, and S. G. S. Alcalá, “A systematic literature review of machine learning methods applied to predictive maintenance,” Computers & Industrial Engineering, vol. 137, p. 106024, 2019, doi: 10.1016/j.cie.2019.106024. [6] T. Zonta, C. A. da Costa, R. da Rosa Righi, M. J. de Lima, E. S. da Trindade, and G. P. Li, “Predictive maintenance in the Industry 4.0: A systematic literature review,” Computers & Industrial Engineering, vol. 150, p. 106889, 2020, doi: 10.1016/j.cie.2020.106889. [7] O. Sürücü, S. A. Gadsden, and J. Yawney, “Condition Monitoring using Machine Learning: A Review of Theory, Applications, and Recent Advances,” Expert Systems with Applications, vol. 221, p. 119738, 2023, doi: 10.1016/j.eswa.2023.119738. [8] M. Yousuf, T. Alsuwian, A. A. Amin, S. Fareed, and M. Hamza, “IoT-based health monitoring and fault detection of industrial AC induction motor for efficient predictive maintenance,” Measurement and Control, vol. 57, no. 8, pp. 1146–1160, 2024, doi: 10.1177/00202940241231473. [9] F. T. Liu, K. M. Ting, and Z.-H. Zhou, “Isolation Forest,” in Proc. 8th IEEE International Conference on Data Mining (ICDM), 2008, pp. 413–422, doi: 10.1109/ICDM.2008.17. [10] E. F. Swana, P. Bokoro, and W. Doorsamy, “Isolation Forest and Local Outlier Factor-Based Anomaly Detection on an Induction Machine,” in 2024 IEEE PES/IAS PowerAfrica, 2024, doi: 10.1109/PowerAfrica61624.2024.10759503.