بحث بعنوان A New Statistical Features Based Approach for Bearing Fault Diagnosis Using Vibration Signals

بحث بعنوان
A New Statistical Features Based Approach for Bearing Fault Diagnosis Using Vibration Signals
Muhammad Altaf 1, Tallha Akram 1, Muhammad Attique Khan 2 , Muhammad Iqbal 1,
M Munawwar Iqbal Ch 3 and Ching-Hsien Hsu 4,5,6,*
1 Department of Electrical and Computer Engineering, COMSATS University Islamabad, Wah 47000, Pakistan;
[email protected] (M.A.); [email protected] (T.A.); [email protected] (M.I.)
2 Department of Computer Sciences, HITEC University Taxila, Taxila 47080, Pakistan;
[email protected]
3 Institute of Information Technology, Quaid-i-Azam University, Islamabad 44000, Pakistan; [email protected]
4 Department of Computer Science and Information Engineering, Asia University, Taichung 400-439, Taiwan
5 Department of Medical Research, China Medical University Hospital, China Medical University,
Taichung 400-439, Taiwan
6 Guangdong-Hong Kong-Macao Joint Laboratory for Intelligent Micro-Nano Optoelectronic Technology,
School of Mathematics and Big Data, Foshan University, Foshan 528000, China

• Correspondence: [email protected]
  † These authors contributed equally to this work.
  Abstract: In condition based maintenance, different signal processing techniques are used to sense
  the faults through the vibration and acoustic emission signals, received from the machinery. These
  signal processing approaches mostly utilise time, frequency, and time-frequency domain analysis.
  The features obtained are later integrated with the different machine learning techniques to classify
  the faults into different categories. In this work, different statistical features of vibration signals
  in time and frequency domains are studied for the detection and localisation of faults in the roller
  bearings. These are later classified into healthy, outer race fault, inner race fault, and ball fault classes.
  The statistical features including skewness, kurtosis, average and root mean square values of time
  domain vibration signals are considered. These features are extracted from the second derivative
  of the time domain vibration signals and power spectral density of vibration signals. The vibration
  signal is also converted to the frequency domain and the same features are extracted. All three feature
  sets are concatenated, creating the time, frequency and spectral power domain feature vectors. These
  feature vectors are finally fed into the K- nearest neighbour, support vector machine and kernel
  linear discriminant analysis for the detection and classification of bearing faults. With the proposed
  method, the reduction percentage of more than 95% percent is achieved, which not only reduces the
  computational burden but also the classification time. Simulation results show that the signals are
  classified to achieve an average accuracy of 99.13% using KLDA and 96.64% using KNN classifiers.
  The results are also compared with the empirical mode decomposition (EMD) features and Fourier
  transform features without extracting any statistical information, which are two of the most widely
  used approaches in the literature. To gain a certain level of confidence in the classification results, a
  detailed statistical analysis is also provided.
  Conclusions
  In this work, the vibration signals were analysed to detect and classify faults in rotating
  machinery. The signal was recorded and its statistical features, such as Average, Kurtosis,
  Skewness and RMS, were calculated in the time domain and the frequency domain. These
  features were also calculated by first finding the second derivative of the raw time domain
  signal. The features were then fed to different machine learning algorithms and were
  analysed for different patterns due to different faults and were used to train these machine
  learning models, resulting in successful detection and classification into ball, inner race
  and outer race faults. The Power Spectral Density features showed the best results for
  KLDA, followed by the statistical features using KLDA. This result was compared with
  that of the EMD, Fourier Transform and Power Spectral Density, in which the former one
  is time-frequency while the latter two are frequency domain representation. It is also
  important to note that the sizes of our proposed features are much less than those of the
  EMD, Fourier and Power Spectral Density, showing the computational efficiency of our
  proposed techniques. The proposed technique can be extended to time-frequency analyses
  like Short Term Fourier Transform and Wavelet Transform and so forth; also other bearing
  faults can be added, such as cage fault, which is not addressed here.

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