Fault Feature Extraction of Gearbox Based on Wavelet and Wavelet Package
Zhang Qin-zhao
Abstract
Zhang Qin-zhao
Abstract
In this paper,a double filtering algorithm is proposed for de-noising of malfunction gearbox vibration signal,which is pro-duced by combining adaptive wavelet transform de-noising and wavelet De-noising by threshold.The de-noising process of this algo-rithm consist of two phases,at first phase,de-noising is implemented by adaptive wavelet transform.At the second phase,the clas-sical wavelet de -noising algorithm by threshold is used for the signal de -noising.At last,the de -noised signal is decomposed by wavelet package.The power of frequency bands is extracted as fault feature vector.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
In this paper,a double filtering algorithm is proposed for de-noising of malfunction gearbox vibration signal,which is pro-duced by combining adaptive wavelet transform de-noising and wavelet De-noising by threshold.The de-noising process of this algo-rithm consist of two phases,at first phase,de-noising is implemented by adaptive wavelet transform.At the second phase,the clas-sical wavelet de -noising algorithm by threshold is used for the signal de -noising.At last,the de -noised signal is decomposed by wavelet package.The power of frequency bands is extracted as fault feature vector.
Key concepts: Stationary wavelet transform, Computer science, Second-generation wavelet transform, Wavelet, Wavelet packet decomposition, Pattern recognition (psychology), Lifting scheme, Discrete wavelet transform