| This paper studies damage assessment degree of steel corrosion fatigue acoustic emission(AE)detection.Research contents included: the acoustic emission(AE)characteristics of metal corrosion fatigue at different depth of linear defects,different loading stress ratios and different loading frequencies are studied in detail.The acoustic emission signal processing method which can characterize the metal corrosion fatigue process under the three different factors mentioned above are explored.First of all,in order to make a comparison under different influencing factors,the loading scheme of experimental mechanics was formulated according to the mechanical properties of the samples.Secondly,a spray corrosion fatigue device was designed to provide a corrosive environment for the samples,the metal corrosion fatigue mechanics test system and acoustic emission on-line test system are established,acoustic emission(AE)signals during corrosion fatigue damage of metal were collected.Finally,for the acoustic emission signal processing method that can characterize the process of metal corrosion fatigue damage,the wavelet function suitable for the characteristics of metal corrosion fatigue acoustic emission signal is selected and the wavelet packet decomposition layer number of acoustic emission signal is determined,the energy variation of metal corrosion fatigue damage in the main frequency band is analyzed and characterized by the acoustic emission signal processing method of wavelet packet energy spectrum.According to the characteristics of metal corrosion fatigue acoustic emission signal,the kernel function in S transform is improved,a generalized transformation of kernel function which regulated the morphology by regulatory factors is proposed,the time-frequency graph of generalized S transform is quantized by information entropy.The acoustic emission signal processing method of generalized S transform and its information entropy was used to analyze and characterize the information change of metal corrosion fatigue damage in time-frequency diagram,the change of information in time-frequency diagram of metal corrosion fatigue damage is analyzed and characterized by generalized S transform and its information entropy of acoustic emission signal processing method.Wavelet packet energy spectrum,generalized S transform and information entropy analysis of acoustic emission signal in metal corrosion fatigue damage process under different influencing factors can be "sensitive" to the different types of acoustic emission signals that produced by interruption and continue corrosion fatigue test.In the corrosion fatigue test of all samples under three different influence factors,the influence factors are linear defect depth or higher loading stress level or lower loading frequency,the probability of occurrence of waveform characteristics such as mixed types of burst and continuous acoustic emission signals or a single continuous acoustic emission signal will increase in the time domain waveform,corrosion fatigue damage is aggravated.Compared with the wavelet packet energy spectrum,only the energy percentage change is shown in the main frequency band to show the singleness of waveform change characteristics,generalized S-transform time-frequency diagram can more comprehensively reflect the above waveform characteristics in the time domain and the frequency domain and the information entropy quantization is used to express the waveform change characteristics as a whole in the time-frequency domain.Therefore,it is feasible to evaluate the degree of metal corrosion fatigue damage by using emission detection technology. |
PDF Full Text Request