Multiscale Ordinal Complexity Analysis for Wind Turbine Bearing Fault Diagnosis Triaxial Accelerometer Feature Hierarchy and Specimen-Level Validation via GroupKFold Cross-Validation
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
https://orcid.org/0000-0002-8011-7027
Alisson V. Brito
Abel C. Lima Filho
Abstract
Permutation-entropy methods for bearing fault diagnosis in wind turbines are typically validated on laboratory test rigs without enforcing specimen-level separation in cross-validation or quantifying axis-specific diagnostic contributions. This study addresses these limitations using the Fraunhofer LBF operational wind turbine bearing dataset, applying multiscale permutation entropy (MPE) to triaxial front-bearing accelerometer signals across 44 bearing specimens (18 healthy, 10 inner race, 4 outer race, 12 roller element), with two corrupted files excluded following data quality screening. GroupKFold cross-validation with unique specimen-level group identifiers prevents the data leakage that arises when temporally correlated analysis windows are split without regard to bearing identity - a limitation present in all ten studies identified in a systematic literature search. MPE achieves 97.67% +/- 2.10% window-level and 97.73% specimen-level accuracy using 12 features across three accelerometer axes, outperforming weighted permutation entropy (WPE, 91.83% window-level, 95.45% specimen-level) and matching a physically-augmented hybrid (MPE+Physical, 21 features) that contributes no additional specimen-level accuracy despite 33.0% feature importance within the combined set. The X-axis accelerometer (brng.f.x) accounts for 40.7% of classification importance, consistent with the primary radial load direction. Permutation entropy reveals a monotonic complexity hierarchy across fault types - Healthy (mu=0.702) < Roller Element (mu=0.889) < Inner Race (mu=0.981) < Outer Race (mu=0.992) - with Cohen's d > 0.9 for all pairwise comparisons. Roller element faults exhibit bimodal permutation entropy distributions, explained by load-zone-dependent impulsive generation. These results demonstrate that multiscale ordinal pattern analysis captures physically meaningful fault signatures in operational wind turbine data when evaluated under methodologically sound cross-validation protocols.
##plugins.themes.bootstrap3.article.details##
multiscale permutation entropy, bearing fault diagnosis, wind turbine condition monitoring, GroupKFold cross-validation, triaxial accelerometer, Shapley attribution, ordinal complexity, specimen-level validation, Fraunhofer LBF dataset, prognostics and health management
Almeida, P. R. L., Lima, T. L. V., Brito, A. V., & Lima-Filho, A. C. (2025, December). Optimal feature complexity for small-sample bearing fault detection in manufacturing. The International Journal of Advanced Manufacturing Technology. doi: 10.1007/s00170-025-17296-8
Bandt, C., & Pompe, B. (2002). Permutation entropy: a natural complexity measure for time series. Physical Review Letters, 88(17), 174102. doi: 10.1103/Phys-RevLett.88.174102
Bhende, A. (2025). A robust xgboost approach for bearing fault prediction from vibration signals: Feature analysis and model evaluation. Noise & Vibration Worldwide. doi: 10.1177/09574565251390935
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum Associates. doi: 10.4324/9780203771587
Costa, M., Goldberger, A. L., & Peng, C.-K. (2002). Multiscale entropy analysis of complex physiologic time series. Physical Review Letters, 89(6), 068102. doi: 10.1103/PhysRevLett.89.068102
Dávalos, A., Jabloun, M., Ravier, P., & Buttelli, O. (2020). Improvement of statistical performance of ordinal multiscale entropy techniques using refined composite downsampling permutation entropy. Entropy. doi: 10.3390/e23010030
Fu, L., Zhu, T., Zhu, K., & Yang, Y. (2019). Condition monitoring for the roller bearings of wind turbines under variable working conditions based on the fisher score and permutation entropy. Energies, 12(16), 3085. doi: 10.3390/EN12163085
Gao, Z.,Wang, Y., Li, X., & Yao, J. (2024). Twins transformer: rolling bearing fault diagnosis based on cross-attention fusion of time and frequency domain features. doi: 10.1088/1361-6501/ad53f1
Hameed, Z., Hong, Y. S., Cho, Y. M., Ahn, S. H., & Song, C. K. (2009). Condition monitoring and fault detection of wind turbines and related algorithms: A review. Renewable and Sustainable Energy Reviews, 13(1), 1–39. doi: 10.1016/j.rser.2007.05.008
Huo, Z., Zhang, Y., Shu, L., & Gallimore, M. (2019). A new bearing fault diagnosis method based on fine-to-coarse multiscale permutation entropy, laplacian score and svm. IEEE Access. doi: 10.1109/ACCESS.2019.2893497
Khan, M., Asad, B., Naseer, M., et al. (2026). An improved hybrid lightweight approach for bearing fault detection and classification in three-phase squirrel cage induction motors. Machines. doi: 10.3390/machines14010068
Li, Y., & Ding, Q. (2026). Multivariate interpolation multiscale chebyshev permutation entropy for bearing fault diagnosis. IEEE Transactions on Instrumentation and Measurement. doi: 10.1109/TIM.2026.3657551
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30. doi: 10.48550/arXiv.1705.07874
McNemar, Q. (1947). Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika, 12(2), 153–157. doi: 10.1007/BF02295996
Minhas, A. S., & Singh, S. (2021). A new bearing fault diagnosis approach combining sensitive statistical features with improved multiscale permutation entropy method. Knowledge-Based Systems, 218, 106883. doi: 10.1016/j.knosys.2021.106883
Mostafavi, A., & Friedmann, A. (2024a). Wind turbine condition monitoring dataset of fraunhofer lbf. Scientific Data, 11, 1108. doi: 10.1038/s41597-024-03934-5
Mostafavi, A., & Friedmann, A. (2024b). Wind turbine condition monitoring dataset of Fraunhofer LBF. Zenodo. doi: 10.5281/zenodo.11820598
Nirmal, W., Lakmal, H., Arachchige, U., & Rodrigo, W. (2026). An interpretable machine learning framework for vibration-based fault detection using multi-domain features. In International conference on soft computing and software engineering. doi: 10.1109/SCSE70081.2026.11499945
Qi, J., Mauricio, A., & Gryllias, K. (2020). Comparison of blind diagnostic indicators for condition monitoring of wind turbine gearbox bearings. In Proceedings of the asme turbo expo. doi: 10.1115/GT2020-15278
Qiao, W., & Lu, D. (2015). A survey on wind turbine condition monitoring and fault diagnosis—part i: Components and subsystems. IEEE Transactions on Industrial Electronics, 62(10), 6536–6545. doi: 10.1109/TIE.2015.2422112
Sandoval, D., Leturiondo, U., Pozo, F., & Vidal, Y. (2020). Low-speed bearing fault diagnosis based on permutation and spectral entropy measures. Applied Sciences, 10(13), 4666. doi: 10.3390/app10134666
Vieira, J., Bauler, V., Rosa, R., & Silva, D. (2025). Towards a more realistic evaluation of machine learning models for bearing fault diagnosis. arXiv preprint. doi: 10.48550/arXiv.2509.22267
Wang, Z., Yao, L., Cai, Y., & Zhang, J. (2020). Mahalanobis semi-supervised mapping and beetle antennae search based support vector machine for wind turbine rolling bearings fault diagnosis. Renewable Energy, 155, 1312–1327. doi: 10.1016/j.renene.2020.04.041
Wheat, L., Mohrenschildt, M., Habibi, S., & Al-Ani, D. (2024). Impact of data leakage in vibration signals used for bearing fault diagnosis. IEEE Access. doi: 10.1109/ACCESS.2024.3497716
Ye, M., Yan, X., & Jia, M. (2021). Rolling bearing fault diagnosis based on vmd-mpe and pso-svm. Entropy, 23(6), 762. doi: 10.3390/e23060762
Zhang, H., Zhong, C., Zhang, Z., & Jiang, Y. (2023). A comparative study of rolling bearing fault diagnosis based on tree models. In 2nd international conference on digital society and intelligent systems. doi: 10.1117/12.2673391
Zheng, J., Zheng, J., Du, Z.-l., et al. (2019). Composite multi-scale weighted permutation entropy and extreme learning machine based intelligent fault diagnosis for rolling bearing. Measurement. doi: 10.1016/J.MEASUREMENT.2019.05.002
Zhou, S., Qian, S., Chang, W., Xiao, Y., & Cheng, Y. (2018). A novel bearing multi-fault diagnosis approach based on weighted permutation entropy and an improved svm ensemble classifier. Sensors, 18(6), 1934. doi: 10.3390/s18061934
Zhou, Y., Chen, J., Zhang, J., et al. (2025). Rolling bearing fault diagnosis based on emd-cnn-transformer. In Youth academic annual conference of chinese association of automation. doi: 10.1109/YAC66630.2025.11150136