Fault detection and identification using real-time wavelet feature extraction

Lopez, Jean-Marc; Tenney, Robert R.; Deckert, J. C.

doi:10.1109/tfsa.1994.467254

Cited by 11 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The application of real-time wavelet feature extraction to various classes of fault data, e.g. helicopter and shipboard pump data, was reported to have a perfect detection, no false alarms with only modest deferral rate [3]. Santoso et al [4; 5] also concluded that wavelet transform analysis was very powerful in detecting and localizing various types of electric power quality disturbances.…”

Section: Time-frequency and Time-scale Analysesmentioning

confidence: 99%

Application of wavelet theory to identify yielding in seismic response of bi‐linear structures

Pan

Lee

2001

Earthq Engng Struct Dyn

View full text Add to dashboard Cite

SUMMARYThe time-frequency and the time-scale analysis methods are used in this paper to identify the dynamic characteristics of non-linear seismic response of structural systems with single degree of freedom (SDOF) and multiple degrees of freedom (MDOF). Based on the oor acceleration response time histories of bi-linear SDOF and MDOF structures, the current study compares the results of system identiÿcation using the short-time Fourier transform (STFT), continuous wavelet transform (CWT) and discrete wavelet transform (DWT) methods. The aim is to identify the frequency variations and the time at on-set of yielding and unloading of a bi-linear structural system during seismic response. The results demonstrate that the CWT method is better than the STFT method in both time and frequency resolutions, and that the DWT method is the best at detecting the time at on-set of yielding and unloading. Combining the results of CWT and DWT methods therefore provides accurate information of both frequency variations and yielding time in non-linear seismic response. To alleviate the problems associated with noise-contaminated signals, e.g. seismic response data recorded on site, the study suggests that low-pass ÿltering be carried out before applying the DWT method to decompose the signals into multiple levels of details.

show abstract

Section: Time-frequency and Time-scale Analysesmentioning

confidence: 99%

Application of wavelet theory to identify yielding in seismic response of bi‐linear structures

Pan

Lee

2001

Earthq Engng Struct Dyn

View full text Add to dashboard Cite

show abstract

“…The choice criterion is based on informative considerations (minimum entropy). The wave-packets transformation has the best time-resolution [8,9] among the variations of wavelets transformations.…”

Section: Wavelets Transformmentioning

confidence: 99%

Non-Stationary Signals: Phase-Energy Approach—theory and Simulations

Klein

Ingman

Braun

2001

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

“…The inability of spectral domain analysis to interpret non-stationary waveforms may to some extent be overcome by using time-frequency transforms (including the short-time Fourier transform, Wigner-Ville, and Choi-Williams distributions [3,4]) or time-scale transforms (including Wavelets and Wavelet packet analysis [5,6,7] )to extract features which could be used to detect gear damage in the presence of fluctuating loading conditions. However manual interpretation of these transforms tends to be difficult and requires the expertise of highly trained and experienced personnel.…”

Section: Introductionmentioning

confidence: 99%

“…Automatic pattern recognition algorithms, such as Neural networks [5,6,7], Support vector machines [8], and Fuzzy logic systems [9], avoid the need for manual interpretation of the data. However due their dependence on expensive training data -which need to be representative of various fault and operating conditions for the specific machine -it is often not possible to implement these supervised learning techniques in practice [10].…”

Section: Introductionmentioning

confidence: 99%

Statistical gear health analysis which is robust to fluctuating loads and operating speeds

Heyns

Godsill²,

Villiers

et al. 2012

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

Condition-based maintenance is concerned with the collection and interpretation of data to support maintenance decisions. The non-intrusive nature of vibration data enables the monitoring of enclosed systems such as gearboxes. It remains a significant challenge to analyze vibration data that are generated under fluctuating operating conditions. This is especially true for situations where relatively little prior knowledge regarding the specific gearbox is available. It is therefore investigated how an adaptive time series model, which is based on Bayesian model selection, may be used to remove the non fault related components in the structural response of a gear assembly to obtain a residual signal which is robust to fluctuating operating conditions. A statistical framework is subsequently proposed which may be used to interpret the structure of the residual signal in order to facilitate an intuitive understanding of the condition of the gear system. The proposed methodology is investigated on both simulated and experimental data from a single stage gearbox.

show abstract

Fault detection and identification using real-time wavelet feature extraction

Cited by 11 publications

References 2 publications

Application of wavelet theory to identify yielding in seismic response of bi‐linear structures

Application of wavelet theory to identify yielding in seismic response of bi‐linear structures

Non-Stationary Signals: Phase-Energy Approach—theory and Simulations

Statistical gear health analysis which is robust to fluctuating loads and operating speeds

Contact Info

Product

Resources

About