Research On Fatigue Crack Growth Prediction Method Of Metal Components Based On Eddy Current Nondestructive Testing

The sudden fracture of metallic materials or components is a very common phenomenon in engineering applications.Fatigue defects are true defects that occur under working circumstances.Compared with artificial defects,the fatigue defects have more complicated wall surfaces,so it is very important to detect defects with non-destructive testing methods.Therefore,it is important to establish an effective combition detection appraoch by fusion the advantages of various detection technologies,and to achieve rapid localization and accurate measurement of fatigue defects on metal surfaces.Meanwhile,the existing models for the lifespan prediction of metallic components with fatigue defects have the problems of large calculation load and low prediction accuracy.This is mainly due to the lack of effective algorithms to explore the loading conditions,specimen thickness and material properties which influence the growth process of the fatigue defect on the metal surface,consequently to predict the lifetime of the metal compoment with defects.Aiming at the above problems in the fatigue defect detection,this thesis first prepared specimens containing fatigue defects,and then detect and analyze these fatigue defects by eddy current pulsed thermography and pulse eddy current technology According to the advantages and disadvantages of these two technologies,a combition non-destructive testing method was proposed by fusion pulsed eddy current thermal imaging and pulsed eddy current for fatigue defects testing.The advantages of the two detection methods were strengthened by BP neural network,which could not only quickly locate the fatigue defect,but also accurately measure the depth of the fatigue defect automatically.Considering the advantages of high compution effeciency the scaled boundary finite element method(SBFEM),SBFEM is proposed to establish the prediction model of the specimen with fatigue defect.This thesis discussed the effects of loading conditions,specimen thickness and material properties on the growth process of the fatigue defects.the growth behavior and lifespan of the fatigue defect were predicted according to the defect depth detected by the proposed combination NDT approach.therefore,a link between the nondestructive testing and fatigue life prediction was established.The main research contents of this paper are as follows:1.Preparation of metal specimens with fatigue defects by a three-point bending testBased on the simulation model,the feasibility of preparing specimens containing fatigue defects by the three-point bending test was studied;a three-point bending test platform was constructed,and stress concentration was formed at the tip of the defect through prefabricated artificial defects to allow the initiation and propagation of fatigue defects;By using high-resolution image acquisition equipment and image recognition software,the depth of fatigue defect after different cycles was measured,and the corresponding relationship between the cycle number and the defect depth was established.2.Approach to fast localization of fatigue defects based on eddy current pulsed thermographyAn eddy current pulsed thermography platform was built to capture video and image on specimens containing fatigue defects.Based on the fact that the pixel value holds strong relationship with its neighboring pixels,the recognition accuracy of the thermal image was improved;the temperature changes at the pre-defected,fatigued defect and non-defected area were investigated.Finally,an approach to fast localization the fatigue defects is proposed and the associated positioning accuracy were analyzed using different examples.3.An approach to quantitative measurement of defect depth based on pulsed eddy current testingFirstly,a numerical simulation of the pulsed eddy current testing system for metal components with defects was conducted to obtain the correlation between the Z-axis magnetic induction intensity of the testing coil and the defect depth;then a pulsed eddy current testing platform was constructed,by which the relationship between the voltage of the pulsed eddy current nondestructive test and the defect depth was obtained;Next,followed that,the effect of the detection positions of the pulsed eddy current probe on the detection voltage was also studied.Followed that,Finally,the calibration curve of the quantitative detection of fatigue defect was obtained.Finally,The depth of the defect was inversely evaluated according to the detection voltage,and the error is analyzed by comparing with the true value.4.A combination Quantitative and nondestructive detection approach for of the fatigue defect in metal by fusion pulsed eddy current thermography and pulsed eddy current testingPulsed eddy current thermography can obtain the precise positions of multiple defects through imaging,and pulsed eddy current nondestructive testing technology can accurately detect the depth of fatigue defects.therefore,the pulse eddy current thermography and pulse eddy current non-destructive testing techniques is combined to achieve automatic,fast and accurate detection of fatigue defects on the metal surface by a three-dimensional electric mobile platform control system.In order to improve the detection accuracy,an GA based BP neural network algrithm is presented to construct the the nonlinear mapping model between the defect depth and output signalbased on the experimental and simulation data obtained from the pulse eddy current detection The result indicates the proposed GA based BP neural network algrithm can improve the detection error5.Analyzation of fatigue defect growth behavior in metal and prediction of fatigue lifespanBy using the proportional boundary finite element method,the effects of loading conditions,specimen thickness and material properties on the fatigue defect growth process were discussed in detail,and the models for growth behavior and lifespan prediction of fatigue defects are proposed based on the experimental data.The effectiveness of the models was verified by using existing data.Finally,the pulse eddy current testing model and the propagation behavior prediction model were verified using the nondestructive testing data of the defect depth,and the prediction error was also analyzed.therefore,the relationship between the nondestructive testing and fatigue life prediction of metal components was linked.