Font Size: a A A

The Research Of Quality Real-time Detection Based On Acoustic And Image Signal In Laser Welding

Posted on:2013-01-14Degree:MasterType:Thesis
Country:ChinaCandidate:J LiuFull Text:PDF
GTID:2251330392970417Subject:Materials Processing Engineering
Abstract/Summary:PDF Full Text Request
Laser welding technology has been used more and more widely in various fieldsdue to it has the advantages of high production efficiency, easy to automate, weldingdeformation is small, stable welding process and so on. Therefore, in order to be moreintuitive understanding the quality problems in the laser welding process, based on theacoustic signal detection the quality of laser welding process, in this paper add theimage signal of the molten pool and keyhole, calculate the acoustic signal radiationsound pressure level, then form it out as image and superimpose it to the molten pooland keyhole image, in order to be more intuitive to local the welding defects.Based on the understanding of the laser acoustic signal sound mechanism and themolten pool and keyhole image signal characteristics, at first, this paper establish a setof molten pool and keyhole image monitoring system. Through the research, the bestparameters and the suitable filter models for shooting laser welding pool and keyholehas been determined, on the basis of taking clear molten pool and keyhole, to join themicrophone array, and collect the acoustic signal and keyhole image signal, laid thefoundation for the next step in the signal processing.Independent component analysis method is used in this paper, after the signals areanalyzed with zero mean and pretreatment, the feature acoustic signal of the laserwelding is extracted. At the same time, the beamforming algorithm of the microphonearray is used to do delay compensation and the weighted sum, then the intensitydistribution of the space radiation sound pressure level has been obtained.Finally, in order to validate the correctness of the algorithm, the normal weldingand defect welding of the sound pressure level and keyhole image are used to contrastrespectively. The test results demonstrate, using this method can better judge thedefect location, this also provides a new idea for the real-time detection technology ofthe laser welding processing quality.At present, the acoustic imaging system is still not perfect in domestic andoverseas. The superposition of the acoustic signal and image signal in weldingprocess is also in the preliminary stage of the research. Now we can only work out acertain range of pressure radiation field distribution, some program is still need to develop. Therefore, the laser welding process of acoustic imaging system is stillneeded further research.
Keywords/Search Tags:laser welding, acoustic signal, molten pool and keyhole imagesignal, real-time monitoring, acoustic imaging
PDF Full Text Request
Related items