Verifying hyperparameter sensitivities of optimal filter design methods for fault signature enhancement

Recent advances in vibrations, noise, and their use for machine monitoring

Open Access

Table 4

The hyperparameter sensitivity verification analysis for the ICS2 is presented. Table 1 is used as the basis, with a check mark (✓) indicating alignment between the results in Figure 4 and Table 1.

#	HP	Measurements			Comment
		1	2	3
1	D	✓	✓	✓	${\tilde{M}}_{i c s 2}$ increases with D; similar results obtained for both D = 64 and D = 256. The fault components are generally weaker in the SES when D = 16.
2	CORF	✓	✓	✓	The results are mostly unaffected by the CORF. ${\tilde{M}}_{i c s 2}$ is expected to decrease or stay the same with larger CORF and vice versa. The results highlight that baseline Δα is already a good choice for the dataset.
3	Δα_b	✓	✓	✓	${\tilde{M}}_{i c s 2}$ is unaffected by perturbations in Δα_b, since the only maxima in the bands are the component-of-interest. Only the damaged components are prominent in the SES.
4	TEP	✓	✓	✓	Except for TEP = 1 in measurement 1, ${\tilde{M}}_{i c s 2}$ decreased with an increase in TEP. Meas. 1 does not have fault information in the corresponding SES, so it does not reflect enhanced fault signatures.
5	N_h	✓	✓	✓	${\tilde{M}}_{i c s 2}$ decreased when N_h is decreased and ${\tilde{M}}_{i c s 2}$ increased or remained the same with an increase in N_h, which is expected.

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.