بحث بعنوان Fault Diagnosis in Rotating Systems Based on Vibration Analysis

بحث بعنوان
Fault Diagnosis in Rotating Systems Based on Vibration Analysis
Mohamed B. Abd-Elbary, Ahmed G. Embaby and Fawkia R. Gomaa
Department of Production Engineering and Mechanical Design, Shebin Alkoum Faculty of Eng.,
Menoufia University, Egypt.
(Corresponding author: [email protected])
ABSTRACT
Vibration is one of the major parameters to consider in condition monitoring of rotating systems. If an
undetected fault is noticed in the rotating system, then, at best, the issue will not be too significant and can be
remedied cheaply and quickly; at worst case, it may result in down-time, expensive damage, injury, or even
life loss, therefore early fault identification is a critical factor in ensuring and extending the working life of
the rotating systems. By measurement and analysis of the vibration of rotating machinery, it is possible to
detect and locate important faults such as mass unbalance, misalignment, bearing failure, gear faults and rotor
cracks. This article is aimed to guide the researchers to implement identification, diagnosis and remedy
techniques of common fault types using vibration analysis and outlines many important techniques used for
condition monitoring of rotating systems such as fast Fourier transform, frequency domain decomposition
method, wavelet transform, stochastic subspace identification and deep learning.
Keywords: fault diagnosis; vibration analysis; rotating system.
Conclusion
Vibration experts and developers have done great
efforts to create functions that solve the few
limitations of vibration analysis, however, there are
still some issues that we are unable to see through
vibration analysis such as;
 Very High frequency: Common sensors have a
maximum frequency of 10 to 15 kHz. If one does
not invest in special sensors, higher frequencies
will be invisible to the equipment.
 Ultra-low frequencies: Although it is possible to
measure very low frequencies, they are often
ignored because they require long samples which
are not done in a normal route.
 Lubricant condition: This is one of the biggest
limitations of vibration analysis. The condition of
the lubricant cannot be evaluated by this
technique, you can only suspect the lack of it.
Many vibration analysis techniques are presented to
explore their capabilities, advantages, and
disadvantage in diagnosing and monitoring rotating
systems. The following points can be concluded:

1. The identification of the bearing faults by using
   frequency analysis is difficult because, it is not
   suitable for non-stationary signal analysis.
2. The identification of the bearing faults is possible
   by using envelope analysis. However, the
   envelope analysis has a major drawback
   consisting of the requirement of a preliminary
   research of the resonance frequencies.
3. The identification of the bearing faults is possible
   by using Short Time Fourier Transform (STFT).
   However, the problem with STFT is that it
   provides constant resolution for all frequencies
   since it uses the same window for the entire
   signal. Therefore, once the window function is
   chosen, the time and frequency resolution are
   fixed. So, there is a trade‐off to choose a proper
   window function between the time resolution and
   the frequency resolution: a longer window will
   lead to a higher frequency resolution with a lower
   time frequency and vice versa.
4. The identification of the bearing faults is possible
   by using Empirical Mode Decomposition (EMD).
   Unfortunately, there are two problems in EMD,
   which are the selection of the suitable
   decomposition level and its intrinsic mode
   functions (IMF) which contains the necessary
   information for faults diagnosis.
5. Deep learning for fault diagnosis had been paid
   less attention. Because of these difficulties:(1) for
   images, the characteristics of recognition objects
   are relatively fixed, but faults are changeable,
   such as patterns variability and shape
   variability;(2) as fault has no fixed pattern,
   whether deep learning can capture a useful
   “hierarchical grouping” or” part-whole
   decomposition” of the fault data is unknown; (3)
   the detection mechanism and ability based on
   deep learning is not yet well explored, especially
   for the incipient faults not any observable
   changes, which is a bottle neck that traditional
   methods suffering.

كلمة سر فك الضغط : books-world.net
The Unzip Password : books-world.net