Font Size: a A A

Acoustic Signal Online Detection Method And System For Cracks Of Thrust Washer

Posted on:2020-01-06Degree:MasterType:Thesis
Country:ChinaCandidate:L ZhangFull Text:PDF
GTID:2381330599459288Subject:Instrument Science and Technology
Abstract/Summary:PDF Full Text Request
Thrust needle roller bearings are widely used in the axial bearing of bearings.Thrust washers are one of the key components of thrust needle roller bearings.When the thrust washer is subjected to alternating load during operation,the crack generated during the manufacturing process will suddenly cause rapid breakage of the component,causing a major hazard,so the crack of the thrust washer must be detected.At present,the factory mainly uses human eye inspection,or the method of listening to the human ear to detect cracks.Manual inspections have high labor intensity,low efficiency,and poor reliability.In order to improve the detection speed and accuracy,and realize the intelligent diagnosis and automatic on-line detection of gasket cracks,this paper studies the on-line detection technology of gasket crack listening.This paper presents a method for detecting crack defects using the natural frequency of thrust washer.The natural frequency is an inherent property of the thrust washer structure.Cracked and crack-free washers have different natural frequencies,are excited by their natural frequencies,and are detected by comparing their natural frequencies.Firstly,the vibration model of the thrust washer is established,and the influence of crack on the natural frequency of the washer is analyzed by combining the ANSYS finite element method.This proves the feasibility of using the natural frequency to detect the crack of the washer,and provides the research direction for the specific implementation of the method.Secondly,designed an excitation device that can be applied to a variety of types of washers,and optimized the parameters of the impact test;designed a fully enclosed soundproof box to reduce the interference of noise on the measurement signal;Then,a mask crack fault signal recognition algorithm based on good products learning is proposed.First,calculate the power spectrum envelope of 30 crack-free washers.By normalizing the peak of the power spectrum,the peak value of the dimensionless parameter is introduced,and the appropriate threshold is given to obtain the characteristic parameters of all the main frequency bands and the effective peak number of the crack-free washer.Finally,after a large number of experimental tests,the appropriate detection threshold is determined.Finally,the design and development of the acoustic signal on-line detection system for cracks of thrust washer has been completed.The test results show that the sound insulation of the soundproof box is about 30.2dB,which can ensure the detection result is not interfered by the surrounding environment noise.The false positive rate of the detection system is 0,and the false positive rate is about 4%,which meets the design requirements.The detection system effectively solves the problems of low efficiency and poor reliability of the current manual inspection method.
Keywords/Search Tags:Thrust washer, Audio signal, Crack detection, Intelligent learning
PDF Full Text Request
Related items