Gear Fault Detection Effectiveness as Applied to Tooth Surface Pitting Fatigue Damage

Lewicki, David G.; Dempsey, Paula J.; Heath, Gregory F.; Shanthakumaran, P.

doi:10.21236/ada498995

Cited by 17 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Earlier studies [13][14][15][16] have used cumulative particle counting with inductive particle sensors in combination with fuzzy logic post processing to detect pitting damage in gears. Results indicated that cumulative mass was a good predictor of pitting damage, and that fuzzy logic was a good technique for setting alarm thresholds.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Comparison of optical and magnetic particle detection systems for detection of pitting damage in low speed hydraulic motors

Sjödin

2013

Linköping Electronic Conference Proceedings

View full text Add to dashboard Cite

In this study two systems for detection of pitting damage in low speed hydraulic motors were compared. These systems were based on an optical particle detector and a magnetic plug respectively. The purpose of the comparison was to determine the relative effectiveness of two methods for detecting pitting damage in a low speed hydraulic motor. In order to do this extensive testing was conducted in Bosch Rexroth Mellansel AB laboratory in Mellansel, Sweden. In the course of the project two post-processing algorithms were developed to enable automatic triggering of alarms based on measurement values from the optical sensor. The optical sensor showed earlier indication of damage than the magnetic plug and the algorithms produced automatic alarms, but there was also a higher false alarm rate when using the optical sensor compared to the magnetic plug.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Dempsey et al have shown the applicability of inductive particle counting on gear box systems [13][14][15][16]. They found that cumulative particle counting can be used to indicate the onset and progression of pitting damage in gears.…”

Section: Introductionmentioning

confidence: 99%

Comparison of optical and magnetic particle detection systems for detection of pitting damage in low speed hydraulic motors

Sjödin

2013

Linköping Electronic Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…the Time Synchronous Average), and then statistics of the analysis, which define the condition indicator. Generally speaking, some papers (McFadden, 86, Ma, 1995) describe an analysis, but not specific condition indictors, whereas other papers, (Zakrajsek, 1989, and 1993, or Lewicki, 2009 refer the condition indicator (CI) which is a statistic of an analysis (such as the kurtosis of the Narrowband Analysis). Finally, Stewart (1977) describe CIs which are derived as gear mesh specific statistics, usually ratio of analysis statistics.…”

Section: Gear Fault Condition Indicatorsmentioning

confidence: 99%

A Comprehensive Analysis of the Performance of Gear Fault Detection Algorithms

Bechhoefer¹,

Butterworth

2019

PHM_CONF

View full text Add to dashboard Cite

Modern gear fault detection analysis began with algorithms based on the time synchronous average. Over the course of decades, many gear analyses have been proposed, but with no evidence that the analysis was significantly more powerful in terms of fault detection than existing algorithms. This study focuses on a comprehensive comparison of gear fault detection algorithms to evaluate their performance. Using a large, statistically significant set of data from three nominal machines and a damaged machine, the CI responses of 88 different analysis are compared in terms of their statistical significance to detect a cracked tooth. The comparison includes residual, energy operator (and its variants), the narrowband analysis (with a comparison of bandwidth requirements), the amplitude and frequency modulation analysis, an analysis of variance of the "factor” analysis: crest, shape, impulse, and margin, and other standard gear fault CIs. Further, the effect of CI selection in the establishment of gear component health is evaluated, where given a set of CIs, a gear health indicator is built, showing that CIs with high statistically separability and low correlation have improved fault detection power. This is validated on a third, dissimilar gear fault propagation test.

show abstract

“…The mean frequency of a scalogram was studied in Ozturk et al 31 to get features for gear pitting fault detection. In the work by Lewicki et al, 32 condition indicators were extracted from time-averaged vibration data. Spectral kurtosis was applied to extract features for gear pitting fault detection.…”

Section: Introductionmentioning

confidence: 99%

Gear pitting fault diagnosis using disentangled features from unsupervised deep learning

Zhang

et al. 2019

Proceedings of the Institution of Mechanical Engineers, Part O:

View full text Add to dashboard Cite

Effective feature extraction is critical for machinery fault diagnosis and prognosis. The use of time–frequency features for machinery fault diagnosis has prevailed in the last decade. However, more attentions have been drawn to machine learning–based features. While time–frequency domain features can be directly correlated to fault types and fault levels, data-driven features are typically abstract representations. Therefore, classical machine learning approaches require large amount of training data to classify these abstract features for fault diagnosis. This article proposed a fully unsupervised feature extraction method for “meaningful” feature mining, named disentangled tone mining. It is shown that disentangled tone mining can effectively extract the hidden “trend” associated with machinery health state, which can be used directly for online anomaly detection and prediction. Compared with wavelet transform and time domain statistics, disentangled tone mining can better extract fault-related features and reflect the fault degradation process. Shallow methods, such as principal component analysis, multidimensional scaling and single-layer sparse autoencoder, are shown to be inferior in terms of disentangled feature learning for machinery signals. Simulation analysis is also provided to demonstrate and explain the potential mechanism underlying the proposed method.

show abstract

Gear Fault Detection Effectiveness as Applied to Tooth Surface Pitting Fatigue Damage

Cited by 17 publications

References 11 publications

Comparison of optical and magnetic particle detection systems for detection of pitting damage in low speed hydraulic motors

Comparison of optical and magnetic particle detection systems for detection of pitting damage in low speed hydraulic motors

A Comprehensive Analysis of the Performance of Gear Fault Detection Algorithms

Gear pitting fault diagnosis using disentangled features from unsupervised deep learning

Contact Info

Product

Resources

About