Application of Adaptive Wavelet Transform for Gear Fault Diagnosis Using Modified-LLMS Based Filtered Vibration Signal

Background: Vibration signature acquired from a gear mesh can be used to identify the defect present in a gear mesh hence can be used to diagnose the condition of a gear mesh. But the signal acquired from the subject may not be noise free and may be non stationary. Methods: Before going for the analysis of the acquired signal a preprocessing on the acquired signal is required to make it noise free. In the present work in first phase, the acquired vibration signal is filtered to reduce the noise and to improve the SNR (signal to noise ratio). The filtering is done by an Adaptive Noise Cancellation (ANC) technique. A modified Leaky Least Mean Square (LLMS) based adaptive algorithm along with a digital filter is used to achieve the ANC. The signal acquired from a healthy gear is used as the reference signal for the adaptive filter based de-noising process. In the second phase of the present work Adaptive Wavelet Transform (AWT) is used to detect the fault by extracting the features from the filtered vibration signal. From the signal pattern the adaptive wavelet is designed. The adaptive wavelet scalogram is compared with the standard wavelet scalogram. Results: The result shows that the adaptive wavelet scalogram is better in analyzing the vibration signal. In this work a gear drive experimental set-up is made. Two different types of defective gears are used for the experiment. In type-1 defective gear one tooth is broken and in type-2 defective gear two teeth are broken. Initially, the vibration signal is acquired from a healthy gear which is used as the reference signal. Then the vibration signal from type-1 defective gear and type-2 defective gear is acquired and processed for the analysis and to identify the defects. Conclusion: The present work shows that with the application of modified-LLMS algorithm and AWT the proposed technique of signal processing is more suitable for the fault identification and hence for the condition monitoring of the gear.

Identification and Representation of Multi-Pulse Near-Fault Strong Ground Motion Using Adaptive Wavelet Transform

In order to identify the horizontal seismic motion owning the largest pulse energy, and represent the dominant pulse-like component embedded in this seismic motion, we used the adaptive wavelet transform algorithm in this paper. Fifteen candidate mother wavelets were evaluated to select the optimum wavelet based on the similarities between the candidate mother wavelet and the target seismic motion, evaluated by the minimum cross variance. This adaptive choosing algorithm for the optimum mother wavelet was invoked before identifying both the horizontal direction owning the largest pulse energy and every dominant pulse, which provides the optimum mother wavelet for the continuous wavelet transform. Each dominant pulse can be represented by its adaptively selected optimum mother wavelet. The results indicate that the identified multi-pulse component fits well with the seismic motion. In most cases, mother wavelets in one multi-pulse seismic motion were different from each other. For the Chi-Chi event (1999-Sep-20 17:47:16 UTC, Mw = 7.6), 62.26% of the qualified pulse-like earthquake motions lay in the horizontal direction ranging from ±15° to ±75°. The Daubechies 6 (db6) mother wavelet was the most frequently used type for both the first and second pulse components.

Wavelet analysis methods in the comprehensive study approach of skin microhemodynamics as a cardiovascular unit

Adaptive wavelet transform techniques for studying of microcirculatory blood flow oscillations are described. It is shown that the suggested methods will be especially claimed in the analysis of low-frequency components of short-lived transient processes under various functional test conditions. In addition, the use of adaptive wavelet transform reduces the essential duration of signal registration, which can be useful in the study of the microhemodynamics in patients with heavy pathologies. Also the method for investigating the phase relationships between microvasculatory oscillations is given which based on estimating the values of wavelet phase coherence function. The proposed method makes it possible to identify frequency intervals with high and low phase correlations of peripheral blood flow oscillations.