Gearbox Fault Identification through Cyclostationary Analysis of Vibration Signals

Achmad Widodo; Ismoyo Haryanto; Mahendra Haryo; Toni Prahasto; Ojo Kurdi

doi:https://doi.org/10.47001/ICJES/2025.411007

Authors

Achmad Widodo

Department of Mechanical Engineering, Universitas Diponegoro, Semarang, Indonesia

Ismoyo Haryanto

Department of Mechanical Engineering, Universitas Diponegoro, Semarang, Indonesia

Mahendra Haryo

Undergraduate Program, Department of Mechanical Engineering, Universitas Diponegoro, Semarang, Indonesia

Toni Prahasto

Department of Mechanical Engineering, Universitas Diponegoro, Semarang, Indonesia

Ojo Kurdi

Department of Mechanical Engineering, Universitas Diponegoro, Semarang, Indonesia

Abstract

Gearboxes are critical components that ensure the operational continuity of industrial machinery. Gearbox failures may arise from material defects, design errors, manufacturing inconsistencies, surface wear, excessive torque loading, and fatigue. Among these, the most frequently observed failures are associated with gear damage, including abrasive wear, cracking, and tooth breakage. Consequently, reliable diagnostic methods are required, and vibration signal analysis using cyclostationary techniques has emerged as a promising approach. A cyclostationary process is a specific class of non-stationary process characterized by periodic variations in its statistical moments. This study examines the condition monitoring of a Suzlon wind turbine using secondary data consisting of 11 samples for each operating condition: normal and faulty. The FFT spectrum analysis indicates average frequencies of the GNF, 1×GMF, and 2×GMF at 473.4 Hz, 925.3 Hz, and 1851.2 Hz, respectively, accompanied by multiple sidebands. The spectral correlation density (SCD) results show corresponding average frequencies of 470.4 Hz, 920.7 Hz, and 1841.0 Hz, also exhibiting sideband structures. Based on the diagnostic outcomes, the gearbox is strongly indicated to exhibit three fault types: wear, eccentricity with backlash, and cracking.

Keywords

Machine diagnostics; Gearbox; Cyclostationary; Vibration; Spectral correlation density

Citation of this Article

Achmad Widodo, Ismoyo Haryanto, Mahendra Haryo, Toni Prahasto, & Ojo Kurdi. (2025). Gearbox Fault Identification through Cyclostationary Analysis of Vibration Signals. International Current Journal of Engineering and Science (ICJES), 4(11), 36-41. Article DOI: https://doi.org/10.47001/ICJES/2025.411007

References

T. Hasegawa et al., “Adaptive training of vibration-based anomaly detector for wind turbine condition monitoring,” Waseda University, Japan, 2017.
F. Liao, C. Sun, and B. Tang, “Wind turbine gear fault detection based on improved cyclostationary analysis with adaptive time–frequency reassignment,” Renewable Energy, vol. 195, pp. 245–258, 2023.
L. Hossain, A. Abu-Siada, and S. M. Muyeen, “Methods for advanced wind turbine condition monitoring and early diagnosis: A literature review,” Curtin Univ., Australia, 2018.
D. J. Wulpi, Understanding How Components Fail, 2nd ed. Ohio, USA: ASM International, 1999.
J. Antoni, “Cyclic spectral analysis in practice,” Mechanical Systems and Signal Processing, vol. 21, no. 2, pp. 597–630, 2007.
F. Liao, C. Sun, and B. Tang, “Wind turbine gear fault detection based on improved cyclostationary analysis with adaptive time–frequency reassignment,” Renewable Energy, vol. 195, pp. 245–258, 2023.
M. H. Zhang and T. Y. Lee, “Cyclostationary feature-enhanced transformers for intelligent gearbox fault detection,” IEEE Transactions on Instrumentation and Measurement, vol. 73, pp. 1–12, 2024.
K. R. Singh and P. K. Kankar, “Gearbox fault diagnosis using enhanced spectral kurtosis and CNN-based classification,” Mechanical Systems and Signal Processing, vol. 172, p. 109063, 2022.
E. Bechhoefer, “High Speed Gear Dataset.” Available: http://data-acoustics.com/measurements/gearfaults/gear-1/. Accessed: May 28, 2019.
W. A. Gardner, Introduction to Random Processes with Applications to Signals and Systems. New York, NY, USA: McGraw-Hill, 1986.
W. A. Gardner, A. Napolitano, and L. Paura, “Cyclostationarity: Half a century of research,” Signal Processing, vol. 86, no. 4, pp. 639–697, 2006.
P. Gidhar and C. Scheffer, Practical Machinery Vibration Analysis and Predictive Maintenance. Oxford, UK: Elsevier, 2004.
P. D. McFadden, “Detecting fatigue cracks in gears by amplitude and phase demodulation of the meshing vibration,” Journal of Vibration, Acoustics, Stress, and Reliability in Design, vol. 108, pp. 165–170, 1986.
J. Martínez, F. Pineda, and R. Ruiz, “Vibration-based predictive maintenance of wind turbine gearboxes using digital twin models and cyclostationary analysis,” Energy Reports, vol. 10, pp. 188–199, 2024.
S. S. Rao and H. Jiang, “Advanced gear health monitoring using 3D spectral correlation and machine learning,” IEEE Access, vol. 11, pp. 112345–112358, 2023.

[1] T. Hasegawa et al., “Adaptive training of vibration-based anomaly detector for wind turbine condition monitoring,” Waseda University, Japan, 2017.

[2] F. Liao, C. Sun, and B. Tang, “Wind turbine gear fault detection based on improved cyclostationary analysis with adaptive time–frequency reassignment,” Renewable Energy, vol. 195, pp. 245–258, 2023.

[3] L. Hossain, A. Abu-Siada, and S. M. Muyeen, “Methods for advanced wind turbine condition monitoring and early diagnosis: A literature review,” Curtin Univ., Australia, 2018.

[4] D. J. Wulpi, Understanding How Components Fail, 2nd ed. Ohio, USA: ASM International, 1999.

[5] J. Antoni, “Cyclic spectral analysis in practice,” Mechanical Systems and Signal Processing, vol. 21, no. 2, pp. 597–630, 2007.

[6] F. Liao, C. Sun, and B. Tang, “Wind turbine gear fault detection based on improved cyclostationary analysis with adaptive time–frequency reassignment,” Renewable Energy, vol. 195, pp. 245–258, 2023.

[7] M. H. Zhang and T. Y. Lee, “Cyclostationary feature-enhanced transformers for intelligent gearbox fault detection,” IEEE Transactions on Instrumentation and Measurement, vol. 73, pp. 1–12, 2024.

[8] K. R. Singh and P. K. Kankar, “Gearbox fault diagnosis using enhanced spectral kurtosis and CNN-based classification,” Mechanical Systems and Signal Processing, vol. 172, p. 109063, 2022.

[9] E. Bechhoefer, “High Speed Gear Dataset.” Available: http://data-acoustics.com/measurements/gearfaults/gear-1/. Accessed: May 28, 2019.

[10] W. A. Gardner, Introduction to Random Processes with Applications to Signals and Systems. New York, NY, USA: McGraw-Hill, 1986.

[11] W. A. Gardner, A. Napolitano, and L. Paura, “Cyclostationarity: Half a century of research,” Signal Processing, vol. 86, no. 4, pp. 639–697, 2006.

[12] P. Gidhar and C. Scheffer, Practical Machinery Vibration Analysis and Predictive Maintenance. Oxford, UK: Elsevier, 2004.

[13] P. D. McFadden, “Detecting fatigue cracks in gears by amplitude and phase demodulation of the meshing vibration,” Journal of Vibration, Acoustics, Stress, and Reliability in Design, vol. 108, pp. 165–170, 1986.

[14] J. Martínez, F. Pineda, and R. Ruiz, “Vibration-based predictive maintenance of wind turbine gearboxes using digital twin models and cyclostationary analysis,” Energy Reports, vol. 10, pp. 188–199, 2024.

[15] S. S. Rao and H. Jiang, “Advanced gear health monitoring using 3D spectral correlation and machine learning,” IEEE Access, vol. 11, pp. 112345–112358, 2023.