| Strong scattering metal materials,represented by the 304 stainless steels,have excellent properties,such as high strength,good ductility,and strong electrical conductivity,and have been widely used in railways,aviation,and nuclear power.These key metal structures are usually subjected to high-intensity dynamic loads,high temperature,and high-pressure during production and service,thus,various kinds of tiny flaws(sub-millimeter flaws)are prone to occur.The further expansion of tiny flaws will lead to structural failure,therefore,It is essential to effectively detecte these tiny flaws to ensure the safety of structures in service.Ultrasonic nondestructive testing method has been widely used for detection of flaws in metal materials,but due to the effects of strong scattering noise from the material,missed detection and false detection may occur.Therefore,in this thesis,an ultrasonic imaging detection method based on synthetic aperture is researched to realize the detection of tiny flaws in strongly scattering metal materials.First,the wave field models for transducers are introduced to improve the accuracy of the synthetic aperture focusing technique(SAFT).By analyzing the diffusion characteristics of the sound field of the transducer and its influence on the ultrasonic imaging performance,the synthetic aperture theory was introduced into the ultrasonic field to improve the imaging resolution.Then,the realization method of time-domain twodimensional synthetic aperture C imaging(2D-SAFT)was further studied.In order to solve the limitation that the point-focused transducer cannot perform synthetic aperture processing at the focal position,the simulation and analysis of the sound field of the transducer are analyzed,and an optimized synthetic aperture processing method for the focal column area of the point-focusing probe is proposed,which considers the characteristics of the sound field.Comparative experiments show that the optimized synthetic aperture processing method can effectively identify tiny-defects,and can significantly improve the detection signal-to-noise ratio.Then,a flaw signal extraction method based on noise extreme value statistics theory is studied.In order to eliminate the influence of noise in the image before and after synthetic aperture processing on flaw detection and evaluation,a noise reduction method based on the statistical distribution of noise under the condition of low gain is proposed.When a large amount of scanned A-wave data are obtained,the noise distribution rules at different positions are counted,and based on the statistical distribution results of noise,a method to determine the optimal imaging threshold is proposed.Finally,the effects of different threshold selection methods on flaw image segmentation are compared and analyzed,and the rationality and efficiency of the optimal threshold determined based on noise statistics method in in defect detection are verified.Next,a three-dimensional imaging method based on 3D-SAFT is proposed to solve the detection problem of tiny flaws at different positions and depths.The 3D enhanced data corresponding to the depth information of the test block is obtained by performing 2D-SAFT hierarchical focusing on the single C-scan data,and then a 3D imaging method based on the 3DSAFT algorithm is proposed.In order to solve the problem of threshold selection at different focal depth positions,a spatial threshold model that depends on the changes of focal depth positions is constructed.Based on the spatial threshold model,the three-dimensional imaging experiment verification and analysis of the flat-bottom hole flaw test block were carried out.The results show that the proposed 3D imaging method can effectively detect the flaws of artificial flat-bottomed holes with a diameter of 0.2 mm.Finally,the advantages and limitations of this 3D imaging method in tiny-flaw detection are analyzed.Finally,a practical application study is carried out using the proposed method in this thesis.The 304 stainless steel test block with strong scattering effect on sound waves is used as the research object,the internal natural(unknown)flaws were detected by 3D imaging.The results of 2D imaging show that under the low system gain of 40 d B,the traditional Cscan imaging has a low signal-to-noise ratio and cannot effectively detect tiny flaws in the strongly scattering metal materials;while the processing method based on 2D-SAFT can improve the detection accuracy.The signal-to-noise ratio,combined with the optimal threshold determination method based on statistical theory,is conducive to the detection of tinyflaws in materials,but focusing waves to different depth planes is required.The results of 3D imaging show that the method based on 3D-SAFT can not only obtain 3D imaging results of natural flaw test blocks,avoid missed detection and false detection of flaws,but also have higher detection efficiency.The size of the natural tiny-flaws detected in this experiment is about 0.297×0.223 mm,which belongs to sub-millimeter flaws,which verifies the effectiveness of the method proposed in this paper in detecting natural tiny-flaws. |
PDF Full Text Request