Study On Automatic Extraction And Evaluation Technology Of Ultrasonic Flaw Signal

Ultrasonic non-destructive inspection methods are widely used fordetecting internal flaws in materials. It’s a frontal research to realize automatedultrasonic inspection for forging workpiece in recent years. Focusing on the actualrequirement in the automatic ultrasonic inspection for forging workpiece, some keytechniques and algorithms in relation to extraction of ultrasonic flaw signal, edgedetection of ultrasonic C-scan image and flaw recognition and evaluation ofultrasonic C-scan image are systematically studied in this dissertation.In chapter1, the current research of some crucial techniques which were relatedto the dissertation was firstly reviewed. Then the background and the significance ofthe research were discussed. Finally, the detailed contents of research and chaptersarrangement were given.In chapter2, ultrasonic flaw signal extraction in ultrasonic inspection wasstudied. Firstly, this method which obtained a RPS matrix used reconstructed phasespace to signals of A-wave ultrasonic. Secondly, ICA was applied to the RPS matrixbased on FastICA algorithm. Thirdly, the available ultrasonic flaw signals wereobtained. Lots of simulation experiments showed the validity of the method isapparent.In chapter3, the flaw edge detection in ultrasonic C-scan image was discussed.In order to improve the anti-noise ability and efficiency of the traditional algorithmsof detection of edge, a new algorithm based on multiscale morphology which canbetter retain the image detail was proposed and lots of experiments approved it.In chapter4, the defect recognition and evaluation in ultrasonic C-scan imagefor forging workpiece was discussed. A method on flaw recognition and evaluationof ultrasonic image was proposed. Meanwhile, this chapter introduced severalintegrant factor of the defect evaluation process.In chapter5, the components of hardware and the implementation of software inforging workpiece automated ultrasonic inspection system were firstly introduced. Then the experimental procedure by detecting instance was provided.Finally, the sixth chapter summarized some key issues related to the full textand put forward some ideas of lacks in the paper.