Test Research On Fatigue Damage Of X80 Pipeline Steel By Magnetic Memory Methods

Fatigue failure is one of the most important components in the form of a variety of mechanical failure of critical zero, according to statistics, about 70% were fatigue fracture failure, which is characterized by no plastic deformation before failure occurs, the occurrence of brittle fracture. It’s great danger. Conventional NDT methods for macroscopic defects already formed can be effectively detected, but for metal equipment and components in service early fatigue damage, hidden damage has not yet formed, especially micro-cracks, it is difficult to achieve effective assessment. However, in engineering practice we want to stress concentration component location anticipation, prevention, or to predict the fatigue life of components, rather than the destruction occurred after the analysis.Magnetic memory testing technology with its unique concept testing, is gradually being used in non-destructive testing of ferromagnetic metal components early damage stage, but for high-level pipeline fatigue testing has not been reported. The need for a large number of targeted pilot study to explore the mechanism and applications.Firstly, chemical machinery industry were the most commonly used materials Q345 R static tensile load research MMM signal under test. For the present, most scholars believe that the magnetic signal does not guarantee the existence of the defect problem component parts at zero value point, I had a thorough discussion, and proposed a new reliable eigenvalues.Next, Tension-tension fatigue tests of prefabricated notched X80 pipeline steel specimens under three different fatigue stresses, and metal magnetic memory signals were detected. The variation of magnetic memory signal during the whole fatigue test was investigated. Methods of determining the defective portion of the specimen were analyzed. Providing quantitative degree of fatigue damage of X80 quantities characteristic. The experimental analysis shows that magnetic memory technology is able to detect high-level pipeline steels which are under fatigue damage, but the accuracy of discrimination has affected loading conditions and specimen conditions and other factors. Gradient magnetic memory signal can effectively characterize the specimen stress concentration; The gradient of magnetic memory signal maxima Kmax can effectively characterize the specimen stress concentration, which is increasing with the increase of the number of fatigue and reflects the degree of fatigue damage of the specimen. Under some certain it can be analyzed quantitatively fatigue life of the specimen.In addition, the paper will be BP neural network combined with the experimental data analysis, to overcome the shortcomings of a single technology. The results show that the BP network is applied to the specimen to identify the state of fatigue damage is feasible and effective.