Research On Ultrasonic Detection System For Axle Billet Of Heigh Speede Rail

The train has become an important means of transportation in daily life,the importance of its safety assurance is self-evident,once an accident occurs,it will cause serious consequences.With the continuous development of China's high-speed railway industry,the running speed of the rail train is increasing rapidly,and the requirements for the components of the train are also increasing.It is very important for the safety and reliability of the train to ensure the production quality of the axle.In the process of axle manufacturing quality management,the most important thing is to ensure the quality of square steel materials.At present,the ultrasonic testing of square steel is still a large range of manual detection.Using the portable ultrasonic testing instrument,the workers hold the ultrasonic probe to move on the surface of the billet to be tested.In view of the inspection requirements of the square steel billet of the train axle,the detection device adopting the ultrasonic nondestructive detection method is developed.Then the ultrasonic detection circuit of the square steel billet is designed,and the control software of the ultrasonic detection system is developed.Last wavelet based filtering method is used to the ultrasonic signal measured on the billets.There is a lot of noise in the echo ultrasonic signal.It is a difficult problem to extract the defect information which may be submerged by noise.Based on the pulse reflection ultrasonic testing method,a special testing circuit is designed.Based on the idea of constant-Q filter group,wavelet decomposition is used to extract signal features and confirm the reconstruction settings under the optimal SNR.Using the noise residual and high-scale decomposition coefficient of the reconstructed signal,the low-frequency drift component is removed.In the ultrasonic testing of square steel billet,the echo frequency will change due to the nonlinear reflection,the energy of defect echo will be distributed near the center frequency,while the frequency of noise signal is relatively scattered,and the energy will be distributed to the whole signal acquisition frequency band.Therefore,when the signal components cross the decomposition scale,they have obvious correlation,and the energy concentration,the noise does not have this feature.So based on statistical classification,local scale region features can be enhanced and defect location can be extracted.In this paper,the non-destructive testing scheme of square steel billet is studied,and the ultrasonic testing device instead of manual scanning is designed.Then the matching ultrasonic testing circuit is developed.The ultrasonic detection system for billets of train axle can reduce the labor intensity and has certain economic benefits.In view of the fact that the ultrasonic signal is easy to be submerged by the back scattered noise,a wavelet analysis based processing scheme is proposed,which can provide a reference for the follow-up study.