Reaserch On The Key Technology Of Blind Zone Detection With Low-frequency Long Distance Ultrasonic Guided Wave In Square Steel Bar

When using ultrasonic guided wave for long-distance Nondestructive Testing,the echo wave method is always used for defect detection.But when the defect is close to the the distal end,it often occurs that the echo waves of the defect will not be observed in the normal monitoring time domain.So,such case is categorized as the so-called blind zone of defect detection within the far-field areas.However,there are more defects in these areas.Therefore,this study will use a long bar-shaped square steel as the research object to study the blind zone in the far-field areas when using ultrasonic guided wave for Nondestructive Testing.The main work and innovative research results achieved are as follows:1.To improve the simulation efficiency,it uses the two-dimensional equivalent model to simplify the model of ultrasonic guided wave transmitting in square steel bar.With simulation and experiment comparison,it proves that the equivalent model is effective.In the experiment,to inhibit signal attenuation caused by the coupling agent,it develops a lithium-based grease instead as the ultrasonic coupling agent,thus greatly improving the performance of the ultrasonic signal transmitting at the contact between the probe and the test piece.In this way,it theoretically explains the reason why the specific unexpected wave occurs in the experiment(that is due to the small angle deviation of the incident wave),and then it also analyzes the impact of the unexpected wave on the distribution of the total incident energy.2.It gives theoretical analysis and deduction to the blind zone phenomenon and its received signal characteristics as ultrasonic guided wave is transmitted in long distance when using the dual-probe reflection method.Based on the theorelical analysis,it derives the range of the blind zone of the ultrasonic guided wave,summaries five common modal conversion forms of the defect reflection waveforms,and analyzes the law that the defect waveforms overlapped in the blind zone.3.It proposes the defect location algorithm in the blind zone based upon differential envelope correlation operation within the monitoring time domain interval ? and ?.By performing correlation calculations between the receiving waveform envelope and the simulation waveform envelope within the monitoring time domain interval,it then can locate the defect.And this is verified by experiments.4.It proposes the defect shape recognition algorithm based upon the time-frequency conversion method.Taking the examples of the classical square-shaped,triangle-shaped and circle-shaped defects,it firstly uses the time-frequency conversion PWVD algorithm to obtain the time-frequency map of the receiving waveforms within the monitoring time domain interval? and ?.Then it uses the convolution neural network algorithm to recognize the three defect categories.In this study,a library containing 1200 maps of intermediate and boundary defects with three shapes is established by simulation method.And the results show that the recognition accuracy of the intermediate and boundary defects categories is 0.89 and 0.85,respectively.In conclusion,this study of the two-dimensional equivalent model of square steel bar,the theory of ultrasonic guided wave blind zone and defect detection algorithm.These relevant research results provide some reference for the theory and method of ultrasonic guided wave blind zone detection.