Assessment of the technical condition of an electromechanical system based on vibration indicators in start-up mode

Object: This research aims to develop and validate a practical, low-cost methodology for the prompt technical condition assessment of electromechanical systems (EMS) during the startup transient. The primary goal is to enable binary classification ("satisfactory" or "unsatisfactory") of an EMS's state using vibration signals acquired solely via a smartphone's built-in accelerometer, facilitating widespread on-site screening without specialized equipment.

Methods: The diagnostic approach is based on recording triaxial vibration acceleration time histories during the startup of industrial asynchronous motors driving screw compressors and exhaust fans. Using a dedicated mobile application (AccelerometerMeter), measurements were taken in three orthogonal directions (axial, horizontal, vertical) relative to the motor housing. The methodology relies on relative and mutual comparison of the recorded signals, focusing on transient characteristics rather than absolute metric values. Results were benchmarked against conventional vibration analysis conducted with professional equipment according to ISO standards and post-maintenance teardown inspections.

Results: The analysis of startup vibration patterns successfully identified distinctive fault signatures. Key diagnostic indicators included anomalous peak amplitudes, sustained instability, and the presence of characteristic beats and shock impulses, particularly during star-delta switching. Empirical correlations were established between the spatial dominance of vibration (e.g., predominant axial vibration indicating misalignment, high vertical components suggesting foundation issues) and specific mechanical or electromagnetic faults. The smartphone-based method effectively differentiated between properly functioning systems and those with confirmed defects, such as bearing degradation or rotor imbalance, with conclusions corroborated by standard vibration severity assessments.

Conclusions: Utilizing a smartphone for vibration analysis during the startup transient proves to be a highly effective and accessible tool for preliminary condition monitoring and screening of electromechanical assets. It provides a viable means for early fault detection in field conditions, allowing maintenance personnel to prioritize equipment for further detailed investigation using advanced diagnostic techniques. The study confirms the high informational content of transient processes for diagnostics and establishes a foundation for a structured, two-level monitoring strategy: initial rapid screening with mobile devices followed by targeted expert analysis, thereby optimizing maintenance resources and enhancing operational reliability.

1. Bannov D. M. Metod diagnostiki obry`vov sterzhnej rotora v asinxronny`x e`lektrodvigatelyax na osnove regressionnogo analiza modulya rezul`tiruyushhego vektora toka statora // Izvestiya Tomskogo politexnicheskogo universiteta. Inzhiniring georesursov. 2022. № 333(5). S.196-208 (In Russ).

2. Abdelhak G, Sid Ahmed B, Djekidel R. Fault diagnosis of induction motors rotor using current signature with different signal processing techniques.// Diagnostyka. 2022. Vol. 23(2). pp. 1 - 9.

3. Sudip Halder, Sunil Bhat, Daria Zychma, Pawel Sowa Broken Rotor Bar Fault Diagnosis Techniques Based on Motor Current Signature Analysis for Induction Motor—A Review.// Energies. 2022. Vol. 15. pp. 8569

4. Zafer Doğan Broken Rotor Bar Fault Detection in Induction Motor Based on Spectral Analysis // Balkan Journal of Electrical and Computer Engineering. 2025. Vol.12(4). pp.357363

5. Kazy`xanov R. R. Povy`shenie nadezhnosti metodom diagnostiki podshipnikovogo uzla e`lektricheskix mashin po spektru potreblyaemogo toka // Novaya nauka: ot idei k rezul`tatu. 2025. № 5. S. 186-198 (In Russ).

6. K. Yatsugi, S. E. M. P. Kone and Y. Mizuno, "Faulty Class Diagnosis of Three Phase Induction Motor Bearing Using Stator Current Spectral Features and Machine Learning Algorithms," 2022 9th International Conference on Condition Monitoring and Diagnosis (CMD), Kitakyushu, Japan. 2022. pp. 112-117.

7. Safiullin R. A., Yangirov I. F. Issledovanie vibracii asinxronnogo e`lektrodvigatelya // E`lektrotexnicheskie i informacionny`e kompleksy` i sistemy`. 2021. T. 17, № 2. S. 41-54 (In Russ).

8. M. Marczak, K. Górny and W. Pietrowski, "Wavelet Based Vibration Analysis for Detection of Inter-Turn Faults in Induction Motors Using Pearson Correlation," 2025 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA), Poznan. Poland. 2025. pp. 22-27.

9. Vibrodiagnostika nasosnogo oborudovaniya teplovy`x e`lektrostancij / A. V. Luk`yanov, D. P. Alejnikov, A. P. Xomenko, V. A. Naletov // Sovremenny`e texnologii. Sistemny`j analiz. Modelirovanie. 2024. № 4(84). S. 103-116 (In Russ).

10. Czvyak V. A., Zhmankov V. M., Shtel`mashenko O. S Vibracionnaya diagnostika podshipnikov i centrobezhnogo nasosa neftegazovogo oborudovaniya// Rezul`taty` sovremenny`x nauchny`x issledovanij i razrabotok : sbornik statej XVII Vserossijskoj nauchnoprakticheskoj konferencii, Penza, 17 iyunya 2022 goda. – Penza: Nauka i Prosveshhenie (IP Gulyaev G.Yu.), 2022. S. 33-35 (In Russ).

11. E. Irgat, E. Çinar and A. Ünsal, "The detection of bearing faults for induction motors by using vibration signals and machine learning," 2021 IEEE 13th International Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives (SDEMPED), Dallas. TX. USA, 2021. pp. 447-453

12. N. Rajapaksha, S. Jayasinghe, H. Enshaei and N. Jayarathne, "Acoustic Analysis Based Condition Monitoring of Induction Motors: A Review," 2021 IEEE Southern Power Electronics Conference (SPEC). Kigali. Rwanda. 2021. pp. 1-10.

13. E. M. Mahani, A. Mirabadi, A. Rezazadeh and A. H. Karamali, "Acoustic Fault Diagnosis of Rolling Bearings in Induction Motors Using Time-Frequency Image Analysis," 2024 4th International Conference on Electrical Machines and Drives (ICEMD), Tehran, Iran, Islamic Republic. 2024. pp. 1-6

14. N. Yordanov, M. Zhilevski and M. Mikhov, "Fault Detection in Electric Motors using Acoustic Signals and Image Classification," 2024 International Conference on Applied and Theoretical Electricity (ICATE), Craiova. Romania. 2024. pp. 1-6

15. E. Resendiz-Ochoa, J. M. Enriquez-Ugalde, J. J. Saucedo-Dorantes and L. A. Morales-Hernandez, "Broken Rotor Bar Failures Diagnosis with Supervised Learning and Infrared Thermography," 2021 IEEE 13th International Symposium on Diagnostics for Electrical Machines, Power Electronics and Drives (SDEMPED). Dallas. TX. USA. 2021. pp. 499-504.

16. S. Kilickaya, C. Celebioglu, L. Eren and M. Askar, "Thermal Image-Based Fault Diagnosis in Induction Machines via Self-Organized Operational Neural Networks," 2025 IEEE Symposium on Computational Intelligence on Engineering/Cyber Physical Systems (CIES), Trondheim, Norway. 2025. pp. 1-7.

17. Ali A., El-Serafi K., A. K. Mostafa S., El-Sheimy N. Frequency Features Based Fuzzy System for Rotating Machinery Vibration Analysis Using Smartphones Low-Cost MEMS Sensors. Journal of Sensor Technology. 2016. Vol. 6 (3).pp. 56-74.

18. Hafizh H., Ali M.N.N., Abdul Majeed A.P.P. Vibration Condition Monitoring of Rotating Machinery with IoT and Smartphone Sensors. In: Tan, A., et al. Advances in Intelligent Manufacturing and Robotics. ICIMR 2023. Lecture Notes in Networks and Systems. 2024. Vol 845. pp. 421-431.