Fault feature extraction method of low-speed heavy-load parts of portal crane

Abstract

The invention relates to a fault feature extraction method of low-speed heavy-load parts of a portal crane. The method includes: decomposing acquired vibration signals of the low-speed heavy-load parts of the portal crane into components of different frequency bands through a dual-tree complex wavelet packet, and obtaining dual-tree complex wavelet packet decomposition coefficients corresponding to different frequency bands; performing singular value decomposition on the dual-tree complex wavelet packet decomposition coefficients of the frequency bands including fault feature information, calculating a singular value differential spectrum curve, and selecting the number of singular value reconstruction components according to a maximum catastrophe point thereof; then performing singular value reconstruction thereon, and performing noise reduction on the components; and performing dual-tree complex wavelet reconstruction thereon, calculating a Hilbert enveloping spectrum of reconstruction signals, and determining whether the low-speed heavy-load parts have faults according to the enveloping spectrum. According to the method, the problem of failure of noise reduction and fault feature extraction through singular value decomposition for the low-speed heavy-load parts of the conventional portal crane by the vibration signals for an ultra-long time is solved.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis of port equipment portal cranes, in particular to a method for extracting fault features of low-speed and heavy-duty components of portal cranes. Background technique [0002] In portal cranes, there are many low-speed and heavy-duty key components, such as hoisting mechanism, luffing mechanism and rotating mechanism, etc. When these low-speed and heavy-load key components fail early, the active components in their vibration signals are extremely weak , often submerged in strong background noise and irregular vibration of cranes. Therefore, how to detect effective weak signal feature information from the strong noise background has become one of the key issues in the fault diagnosis of low-speed and heavy-duty components of portal cranes. [0003] However, the low-speed and heavy-duty equipment's unique low rotation frequency characteristics make it more difficult to extract impact fault ...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a method for extracting fault features of low-speed and heavy-duty parts of portal cranes. When the method utilizes dual-tree complex wavelet packets to decompose vibration signals, the decomposition coefficients of each frequency band will decrease with the increase of the number of decomposition layers. Therefore, it is possible to use the singular value decomposition to denoise the ultra-long time vibration signal, and overcome the defect that the existing singular value decomposition method cannot detect the weak feature information in the ultra-long time signal

Images