Improved algorithms for on-line oil analysis have been suggested. They utilize measurements of multiple oil properties of interest and detect faults induced by transients in the acquired signals. Transient detection is based on the cumulative sum of errors (CUSUM) technique. Performance capabilities are tested on laboratory conditions.
COBISS.SI-ID: 13160219
Premature bearing failures can be caused by a large number of factors were one of the most common causes is an inadequate lubrication. Improperly lubricated bearings detection from vibration patterns is yet a difficult task, especially when records from short operating periods are available. This problem has been addressed by applying recently introduced wavelet bispectral analysis, a technique for revealing time-phase relationships. The experimental results reveal that bispectral absolute value is not enough sensitive to reveal the lubricant deficiency fault. However by further extraction and treatment of the biphase phase information one can gain better insight into the fault bearing state. Improper lubrication is expressed in different phase coupling length and bifrequency region in the bispectrum domain.
COBISS.SI-ID: 10015817
Almost all known methods for vibration-based diagnosis of bearings stand on some sort of analysis of amplitude spectra of the acquired signal. This analysis can be agravated by presence of fluctuations in the operating conditions. We developed an alternative approach which models the occurrences of localized bearing fault patterns as a realization of random point process whose inter-event time intervals are governed by inverse Gaussian mixture. Hence the approach turns much more robust to fluctuations in operating conditions. The applicability of the model was evaluated on vibrational signals generated by bearing models with localized surface fault.
COBISS.SI-ID: 27178535
In the paper we propose a novel approach to the diagnosis of gearboxes in presumably nonstationary and unknown operating conditions. The approach makes use of information indices based on Rényi entropy derived from coefficients of the wavelet packet transform of measured vibration records. These indices quantify some statistical properties of instantaneous power of the generated vibration that are largely unaffected by changes in the operating conditions. The analysis is based on probability density of the envelope of a sum of sinusoidal signals with random amplitude and phase. Such an approach requires no a priori information about the operating conditions and no prior data describing physical characteristics of the monitored drive. The fault detection capabilities of the proposed feature set are demonstrated on a two-stage gearbox operating under different rotational speeds and loads with various seeded mechanical faults.
COBISS.SI-ID: 25765159
Standard bearing fault detection features are shown to be ineffective for estimating bearings remaining useful life. Addressing this issue, in this paper we propose an approach for bearing fault prognostics based on features describing the statistical complexity of the envelope of the generated vibrations and a set of Gaussian process models. The proposed features are sufficiently sensitive to the changes in the bearing condition and in the same time are sufficiently robust to variations in the operating conditions. Gaussian process models are nonparametric black-box models which differ from most other frequently used black-box identification approaches as they do not try to approximate the modeled system by fitting the parameters of the selected basis functions, but rather search for the relationships among measured data. In this paper the GP models are used for filtering noisy features and estimating the RUL based on filtered features.
COBISS.SI-ID: 27855399