The Research Of Acoustic Emission Signal Recovery Based On Blind Deconvolution

Refractories is a lining material，which widely used in metallurgy industry, chemicalindustry, petroleum, machinery manufacturing, etc., while in the metallurgical industry，refractories are widely used, accounting for50%to60%of total output.The reliability ofrefractories is directly related to the blast furnace, therefore, for structural safety, to avoidsecurity accident, and improve product quality, and so is of great significance andapplication value,it is necessary to conduct safety testing for refractory materials.In recentyears, acoustic emission technology and piezoelectric sensor technology in the field ofnon-destructive testing and condition monitoring has been widely used, and has madeimportant breakthroughs and scientific research,this paper,The author have done someresearch based on the research of experts and scholars in the past, using blind deconvolutionalgorithm and piezoelectric sensor technology to research the acoustic emission signaldistortion.The main contents are following:(1) For the refractory acoustic emission signal propagation processes occurring decay,scattering distortion,designed a experimental device.Assuming the propagation path is linearsystems,with a broken pencil signal to simulate refractories’ acoustic emission signals,thisarticle assumed that the propagation path is linear systems, and estimated the propagationpath of Impulse response function, designed blind deconvolution filter, restored the acousticemission source signal.(2) The impact of the acoustic emission signal propagation path response functionestimate based on blind deconvolution, found that different propagation paths’ responsecurve are different,which shows that high frequency component attenuation faster than thelow frequency component.Furthermore, since the resonance effect of acoustic emissionprobe,the frequency response curve has obvious peak at150kHz.(3) Designed experiments to verify the feasibility of blind deconvolution， by piezoelectric stimulated a determined input signal，calculation of the propagation paths’transfer function through the input signal and output signal,found that blind deconvolutionmethod has certain reliability at the time of identification of unknown linearsystem.Meanwhile,validated the effects of restored signal.(4) Using blind deconvolution algorithm to recovery acoustic emission signals,calculating the spectrum coefficient, and contrast recovery frequency of each band signalcomponents. It provides a reference for accurately determine the damage state of therefractories.