13 October 2008 Real-time condition monitoring based on analysis of vibration signal with wavelet transform
Author Affiliations +
By combining wavelet analysis and neural network, a new approach for condition monitoring is presented for rotating machine fault. The wavelet analysis can accurately localize the features of transient signal in time-frequency domains. The wavelet transform technology is appropriate for processing of fault signals consisting of short-lived, high-frequency components closely located in time as well as long duration components closely spaced in frequency. In a view of the inter relationship of wavelet decomposition theory, the crucial components as features are inputted into radial basis function for fault pattern recognition. In order to acquire the network parameters, the improved Levenberg-Marquardt optimization technique is used for training process. By choosing enough samples to train wavelet network, the fault pattern can be determined according to the output results. Also, the robustness of wavelet network for fault diagnosis is discussed. The applied results show that the proposed method can improve the performance for real-time monitoring of vibration fault.
© (2008) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hua Liu, Hua Liu, Hong Zhang, Hong Zhang, Aili Qin, Aili Qin, } "Real-time condition monitoring based on analysis of vibration signal with wavelet transform", Proc. SPIE 7129, Seventh International Symposium on Instrumentation and Control Technology: Optoelectronic Technology and Instruments, Control Theory and Automation, and Space Exploration, 712913 (13 October 2008); doi: 10.1117/12.807383; https://doi.org/10.1117/12.807383

Back to Top