Research On Detection Method For Contact Distribution Of Pressure Coupled Surface Based On Reflective Ultrasonography

With the development of machine tools and other equipment manufacturing industry,detection of the contact distribution for mechanical joints is received more and more attention.During mechanical surface machining process,there are concave and convex surfaces of different ranges due to different methods of machining.Contact of mechanical joints is actually the contact of series of asperities on both roughness surfaces.It will be different from the actual situation if the influence of mechanical bonding state is ignored when research on mechanical system characteristics and optimization design of the system.Therefore,nondestructive testing of contact distribution of mechanical bonding surfaces will be significant in protecting coordination,sealing,drive properties,accuracy and other performance of mechanical systems,reducing the friction wear,also in prolonging service life of the machine and improving productivity.Contact of mechanical joints is characterized by the contact and change of microcosmic profile,not material inner defect.So the traditional material defect detection method is not suitable for detecting the coupling distribution of joint surfaces.This thesis is mainly studying on the contact distribution and gives corresponding evaluation of joint surfaces by the ultrasonic detecting,ultrasonic signal processing and ultrasound image analysis and modeling technologyFirstly,this paper describes the basic theory of ultrasound detection and analyzes the principle of detecting to joint faces.Ultrasonic reflection detection method is used for air coupled contact joint faces.A platform for joint surfaces detecting system was built,the noise suppression for ultrasonic signal was completed,the characteristic value extraction methods included peak intensity,integral,and frequency spectrum was analyzed and frequency spectrum method was selected to extract characteristic value of echo.The image modeling of the contact distribution of joint surfaces is built.The image super-resolution reconstruction technique is used for reducing the accuracy limit caused by finite step distance of stepper motor and size of ultrasound beams cannot be equivalent to a point.Combining with super-resolution reconstruction theory,the mathematical model of ultrasonic inspection imaging was analyzed and ultrasonic image sequences super-resolution model was analyzed.Image registration algorithm combining Fourier-Mellin with Taylor series and super-resolution reconstruction method based on projection onto convex sets(POCS)was studied.Simulation results and performance analysis show that the algorithm can be used to improve the detection accuracy of the contact distribution of the joint surface.Finally,through experimental the effect of super resolution reconstruction technique on the contact details enhancement was analyzed,non-contact region locating and size-resolving of machine joint surfaces was studied,the effects of different roughness on the contact distribution of the specimen under the same pressure were compared and analyzed.The effect of different pressure on the contact distribution of the joint surface of the same specimen was explored.To a certain extent,the different experimental results show that the ultrasonic testing method for contact distribution of mechanical contact surface proposed by this thesis is effective and feasible.