Table 14.

Acoustic FD methods applied to various industrial systems.

Methods	Advantages	Disadvantages
MSAF	This method can effectively detect faults in electric motors by analyzing acoustic signals.	MSAF methods require the selection of parameters and the number of groups to be determined in advance.
SMOFS	This method helps diagnose faults accurately by using an iterative process to identify relevant frequency components.	As with the MSAF, SMOFS methods require a prior selection of parameters and groupings.
MSAF-17-MULTIEXPANDED-FILTER-14	More accurate results with greater resolution and detail in the signal by using 14 bandwidth and 17 frequency components.	The results of the recognition process depend on the training samples.
SMOFS-22-MULTIEXPANDED	Useful for early fault diagnosis in rotating machines, both electrical and mechanical.	The acoustic signals in this method may overlap and merge, causing problems in analysis, such as reflections and overlapping waves.
MSAF-RATIO-24-MULTIEXPANDED-FILTER-8	High recognition results in diagnosing electrical motors.	Dependence on a lot of training samples and spectral leakage errors in computed frequency bandwidth.
MSAF-RATIO-27-MULTIEXPANDED-4-GROUPS	Implementation of this method is inexpensive and has the potential to be used for a wider range of purposes than just fault detection.	Signals using this method are affected by background noise and reflected sounds.