DETECTION OF MECHANICAL FAULTS IN ROTATING SYSTEMS USING ARTIFICIAL INTELLIGENCE: A DECISION TREE–BASED APPROACH
Abstract
The early detection of incipient faults in rotating machinery is essential for predictive
maintenance and the reduction of operational costs. This study evaluates the validity of artificial intelligence, specifically the decision tree technique, in identifying incipient unbalance and misalignment faults, in comparison with their vibration signature and normal operating condition, assuming no other types of faults are identified among those studied. For this purpose, a simulated dataset with previously known responses was used to train a supervised learning model in Python
language. The model was trained to recognize deterministic frequencies and their harmonics and to classify each type of fault. The results obtained by the artificial intelligence were compared with the known responses, verifying the accuracy of the model in classifying the faults, with an accuracy of 93.67%. The analysis demonstrated the effectiveness of the approach, highlighting the potential of
the decision tree technique for applications in condition monitoring and fault diagnosis in rotating systems, as it was able to identify patterns and classify them satisfactorily.
Author Biographies
Bacharel em Engenharia Mecânica, Universidade Federal de São João del Rei (UFSJ), Taubaté, São Paulo, Brasil.
Doutor em Engenharia Mecânica, Universidade de Campinas (Unicamp), São Tiago, Minas Gerais, Brasil.
Doutor, Universidade Federal de São João del Rei, São João del Rei, Minas Gerais, Brasil.
References
BENTLY, Donald E.; HATCH, Charles T. Fundamentals of rotating machinery diagnostics. 2. ed. Minden: Bently Pressurized Bearing Press, 2003.
BISHOP, Christopher M. Pattern recognition and machine learning. New York: Springer, 2006.
BREIMAN, Leo. Statistical modeling: the two cultures. Statistical Science, v. 16, n. 3, p. 199–231, 2001.
BREIMAN, Leo; FRIEDMAN, Jerome H.; OLSHEN, Richard A.; STONE, Charles J. Classification and regression trees. New York: Chapman & Hall/CRC, 1984.
CARVALHO, Tiago P. et al. A systematic literature review of machine learning methods applied to predictive maintenance. Computers & Industrial Engineering, v. 137, p. 106024, 2020. DOI: https://doi.org/10.1016/j.cie.2019.106024
DAS, Oguzhan; DAS, Duygu Bagci; BIRANT, Derya. Machine learning for fault analysis in rotating machinery: a comprehensive review. Heliyon, v. 9, n. 6, e17584, 2023. DOI: https://doi.org/10.1016/j.heliyon.2023.e17584
DEWESOFT. Gráfico de vibração do eixo [imagem]. 2023. Disponível em: https://dewesoft.com/pt/blog/como-interpretar-dados-de-monitoramento-de-condicao Acesso em: 17 mar. 2026.
GOODFELLOW, Ian; BENGIO, Yoshua; COURVILLE, Aaron. Deep learning. Cambridge: MIT Press, 2016.
HAO, X.; ZHANG, Guigang; SHANG, M. Deep learning. International Journal of Semantic Computing, v. 10, n. 3, p. 417–439, 2016. DOI: https://doi.org/10.1142/S1793351X16500045
HARRIS, Charles R. et al. Array programming with NumPy. Nature, v. 585, n. 7825, p. 357–362, 2020. DOI: https://doi.org/10.1038/s41586-020-2649-2
HASTIE, Trevor; TIBSHIRANI, Robert; FRIEDMAN, Jerome. The elements of statistical learning: data mining, inference, and prediction. 2. ed. New York: Springer, 2009.
KARDEC, Allan; NASCIF, Júlio. Manutenção: função estratégica. 3. ed. Rio de Janeiro: Qualitymark, 2009.
KOUTNÍK, Jan; GREFF, Klaus; FAUSTINO, G.; SCHMIDHUBER, Jürgen. A clockwork RNN. In: INTERNATIONAL CONFERENCE ON MACHINE LEARNING, 31., 2014, Beijing. Proceedings […] PMLR, 2014. p. 1863–1871.
MCKINNEY, Wes. Data structures for statistical computing in Python. In: PYTHON IN SCIENCE CONFERENCE, 9., 2010, Austin. Proceedings [...] SciPy, 2010. p. 51–56.
MOZAMMIL, S.; PANIGRAHI, S.; ISRAR, M. Vibration detection and analysis for fault prediction of electric rotating machines. Dogo Rangsang Research Journal, v. 10, n. 12, p. 442–446, 2020.
NEUPANE, Dhiraj; BOUADJENEK, Mohamed Reda; DAZELEY, Richard; ARYAL, Sunil. Data-driven machinery fault diagnosis: a comprehensive review. Neurocomputing, v. 627, p. 129588, 2025. DOI: https://doi.org/10.1016/j.neucom.2025.129588
NGUYEN, Kien; MEDJAHER, Kamal. A new dynamic predictive maintenance framework using machine learning for failure prognostics. Reliability Engineering & System Safety, v. 188, p. 251–262, 2019. DOI: https://doi.org/10.1016/j.ress.2019.03.018
PEDREGOSA, F. et al. Scikit-learn: machine learning in Python. Journal of Machine Learning Research, v. 12, p. 2825–2830, 2011.
QUINLAN, J. Ross. C4.5: programs for machine learning. Burlington: Morgan Kaufmann, 2014.
RANDALL, Robert B. Vibration-based condition monitoring: industrial, aerospace and automotive applications. Chichester: John Wiley & Sons, 2011.
RUDIN, Cynthia. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence, v. 1, p. 206–215, 2019. DOI: https://doi.org/10.1038/s42256-019-0048-x
ZHANG, Wei et al. A review on machinery fault diagnosis using machine learning for rotating equipment. Mechanical Systems and Signal Processing, v. 148, p. 107247, 2021. DOI: https://doi.org/10.1016/j.ymssp.2020.107247
ZHAO, Rui et al. Deep learning and its applications to machine health monitoring: a survey. Mechanical Systems and Signal Processing, v. 115, p. 213–237, 2019. DOI: https://doi.org/10.1016/j.ymssp.2018.05.050
License
Copyright (c) 2026 RECIMA21 - Revista Científica Multidisciplinar - ISSN 2675-6218
This work is licensed under a Creative Commons Attribution 4.0 International License.
Os direitos autorais dos artigos/resenhas/TCCs publicados pertecem à revista RECIMA21, e seguem o padrão Creative Commons (CC BY 4.0), permitindo a cópia ou reprodução, desde que cite a fonte e respeite os direitos dos autores e contenham menção aos mesmos nos créditos. Toda e qualquer obra publicada na revista, seu conteúdo é de responsabilidade dos autores, cabendo a RECIMA21 apenas ser o veículo de divulgação, seguindo os padrões nacionais e internacionais de publicação.