A Sparse Maximum Harmonic-to-Noise Ratio Deconvolution Method

Abstract

A sparse maximum harmonic-to-noise ratio deconvolution method, firstly truncate and remove the mean value of the collected signal, then perform an estimation period operation on the situation where the precise period is not given in advance, and then deconvolute the signal to The harmonic-to-noise ratio is used as the objective function to derive the filter coefficients to obtain an iterative expression, and perform sparse processing on the filtered signal in each iterative filtering process. The threshold and period of the sparse processing will follow the filtered signal. Update, and finally perform envelope analysis on the deconvoluted signal, and the fault characteristic frequency can be extracted from the envelope spectrum. The present invention does not require human participation in the extraction process of the characteristic frequency, which is beneficial to the realization of fault characteristic extraction and diagnosis and monitoring. Automate, save time and increase efficiency.

Description

technical field [0001] The present invention relates to the technical field of mechanical equipment fault diagnosis, in particular to a sparse maximum harmonics-to-noise ratio deconvolution (Sparse Maximum Harmonics-to-noise ratio Deconvolution, SMHD) method. Background technique [0002] Vibration analysis is one of the most effective methods for fault diagnosis of mechanical equipment at this stage, and the state degradation of mechanical equipment often manifests as changes or abnormalities in vibration information. At present, signal processing methods based on vibration information, such as time domain method, frequency domain method and time frequency domain method, have been successfully applied to bearing fault diagnosis and have produced very good results. However, there are still many challenges in the field of rolling bearing fault diagnosis, and the extraction of bearing faults still has many difficulties. 1. The lengthy and complicated transmission path between...

AI Technical Summary

Problems solved by technology

However, there are still many challenges in the field of rolling bearing fault diagnosis, and the extraction of bearing faults still has many difficulties

1. The lengthy and complicated transmission path between the test sensor and the fault source can seriously affect the transfer function, thereby reducing the amplitude of the impact signal and prolonging the time, so that the pulse caused by the fault is easily covered by noise

2. The random fluctuation of the rollers in the bearing will further blur the spectral envelope spectrum of the original quasi-periodic fault shock

3. The influence of aperiodic noise and periodic interference from the mechanical system adds more challenges to extracting the impact of bearing faults

However, the MCKD method has a great dependence on the accurate estimation of the fault period in advance. In engineering practice, due to the limitation of the speed measuring equipment, it is difficult to install the speed measuring device or the cost is high, and the complex working conditions make it impossible to keep the speed of the equipment constant.

These will affect the accuracy of speed measurement and lead to period estimation errors. At the same time, MCKD has poor tolerance to period fluctuations caused by random sliding. These shortcomings have brought a lot of inconvenience to the application and promotion of MCKD.

Images