Kernel PCA in nonlinear visualization of a healthy and a faulty planetary gearbox data

Anna M. Bartkowiak1, Radoslaw Zimroz2

1Wroclaw University, Institute of Computer Science, 50-383, Wroclaw, Poland, (retired)

2Wroclaw University of Technology, Faculty of GeoEngineering Mining and Geology,
50-421, Wroclaw, Poland

1Corresponding author


Received 30 August 2017; accepted 31 August 2017



Abstract. PCA (Principal Component Analysis) is a powerful method for investigating the dimensionality and extracting structure from multi-dimensional data, however it extracts only linear projections. More general projections – accounting for possible non-linearities among the observed variables – can be obtained using kPCA (Kernel PCA), that performs the same task, however working with an extended feature set. We consider planetary gearbox data given as two 15-dimensional data sets, one coming from a healthy and the other from a faulty planetary gearbox. For these data both the PCA (with 15 variables) and the kPCA (using indirectly 500 variables) is carried out. It appears that the investigated PC-s are to some extent similar; however, the first three kernel PC-s show the data structure with more details.

Keywords: visualization of MD data, principal components, PCA and kPCA, kernel trick.


[1]        Jolliffe I. T. Principal Component Analysis. 2nd Ed., Springer, New York 2002.

[2]        Zimroz R., Bartkowiak A. Two simple multivariate procedures for monitoring planetary gearboxes in non-stationary operating conditions. Mechanical Systems and Signal Processing, Vol. 38, Issue 1, 2013, p. 237‑247.

[3]        Wodecki J., Stefaniak P. K., Obuchowski J., Wylomanska A., Zimroz R. Combination of principal component analysis and time-frequency representations of multichannel vibration data for gearbox fault detection. Journal of Vibroengineering, Vol. 18, Issue 4, 2016, p. 2167‑2175.

[4]        Schölkopf B., Smola A., Müller K. R. Nonlinear component analysis as a kernel eigenvalue problem. Neural Computation, Vol. 10, Issue 5, 1998, p. 1299‑1319.

[5]        Schölkopf B., Mika S., Burges Ch J. C., Knirsch P. H., Müller, K. R., Rätsch G., Smola A. Input space versus feature space in kernel-based methods. IEEE Transactions on Neural Networks, Vol. 10, Issue 5, 1999, p. 1000‑1017.

[6]        Bartelmus W., Zimroz R. A New feature for monitoring the condition of gearboxes in non-stationary operating systems. Mechanical Systems and Signal Processing, Vol. 23, Issue 5, 2009, p. 1528‑1534.

[7]        Shawe Taylor J., Christianini N. Kernel Methods for Pattern Analysis. Cambridge University Press, UK, 2004.

[8]        Wang F., Chen X., Dun B., Wang B., Yan D., Zhu H. Rolling bearing reliability assessment via kernel principal component analysis and Weibull proportional Hazard model. Shock and Vibration, 2017,

[9]        Mika S., Rätsch G., Weston J., Schölkopf B., Müller K. R. Fisher discriminant analysis with Kernels. Proceedings of the 1999 IEEE Signal Processing Society Workshop, 1999, p. 41-48.

Cite this article

Bartkowiak Anna M., Zimroz Radoslaw Kernel PCA in nonlinear visualization of a healthy and a faulty planetary gearbox data. Vibroengineering PROCEDIA, Vol. 13, 2017, p. 62‑66.


© JVE International Ltd. Vibroengineering PROCEDIA. Sep 2017, Vol. 13. ISSN 2345-0533