Bearing Fault Severity Classification Based on EMD-KLD : A Comparative Study for Incipient Ball Faults

Mezni, Zahra; Delpha, Claude; Braham, Ahmed

doi:10.1109/phm-besancon49106.2020.00050

Cited by 6 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A kernel-based technique with a higher dimensional projection space such as a SVM could be an option. However, in previous works [ 57 , 64 ], it was shown that a SVM combined with different approaches (one-against-all, one-against-one and directed acyclic graph SVM) could provide satisfactory classification results. However, these results were obtained to the detriment of high training times ( 132.1 s for the polynomial kernel and 14.05 s for the Gaussian one).…”

Section: Results and Discussionmentioning

confidence: 99%

“…The technique is based on distance estimation and the tuning of a predefined number of nearest neighbours denoted K [ 56 ]. In [ 57 ], the determination of K as well as the pertinent metrics were analysed. In the following, K was selected in the range of

and Euclidean and city block distances were evaluated.…”

Section: The Fault Diagnosis Methodologymentioning

confidence: 99%

“…The maximum deviance reduction (MDR), twoing rule (TR) and Gini’s split criteria were adopted in our approach to tune the Spc. More details can be found in [ 57 ].…”

Section: The Fault Diagnosis Methodologymentioning

confidence: 99%

See 2 more Smart Citations

Performance of Bearing Ball Defect Classification Based on the Fusion of Selected Statistical Features

Mezni

Delpha

Braham

2022

Entropy

Self Cite

View full text Add to dashboard Cite

Among the existing bearing faults, ball ones are known to be the most difficult to detect and classify. In this work, we propose a diagnosis methodology for these incipient faults’ classification using time series of vibration signals and their decomposition. Firstly, the vibration signals were decomposed using empirical mode decomposition (EMD). Time series of intrinsic mode functions (IMFs) were then obtained. Through analysing the energy content and the components’ sensitivity to the operating point variation, only the most relevant IMFs were retained. Secondly, a statistical analysis based on statistical moments and the Kullback–Leibler divergence (KLD) was computed allowing the extraction of the most relevant and sensitive features for the fault information. Thirdly, these features were used as inputs for the statistical clustering techniques to perform the classification. In the framework of this paper, the efficiency of several family of techniques were investigated and compared including linear, kernel-based nonlinear, systematic deterministic tree-based, and probabilistic techniques. The methodology’s performance was evaluated through the training accuracy rate (TrA), testing accuracy rate (TsA), training time (Trt) and testing time (Tst). The diagnosis methodology has been applied to the Case Western Reserve University (CWRU) dataset. Using our proposed method, the initial EMD decomposition into eighteen IMFs was reduced to four and the most relevant features identified via the IMFs’ variance and the KLD were extracted. Classification results showed that the linear classifiers were inefficient, and that kernel or data-mining classifiers achieved 100% classification rates through the feature fusion. For comparison purposes, our proposed method demonstrated a certain superiority over the multiscale permutation entropy. Finally, the results also showed that the training and testing times for all the classifiers were lower than 2 s, and 0.2 s, respectively, and thus compatible with real-time applications.

show abstract

Section: Results and Discussionmentioning

confidence: 99%

and Euclidean and city block distances were evaluated.…”

Section: The Fault Diagnosis Methodologymentioning

confidence: 99%