The Welded Joint Reliability Evaluation Based On Magnetic Memory Parameters

The Steel equipment with welded joint, where more or less exist different variety ofdefects and stress concentration,which will reduce the strength and service life of equipment,lead to serious consequences of equipment failure, rupture, explosion.So it is prone towelding failure.Magnetic memory testing technology, a new nondestructive testing method, can not onlydiscover the plastic deformation and the macroscopic of material, but also can assess thestress concentration districts and the microcosmic defects of metal endangered state beforethe injury to early diagnosis. However, due to some randomness, there is some dispersion anduncertainty of magnetic memory signals and deviation of the objective phenomena andsubjective description or prediction of phenomena, while these deviations show a statisticalregularity, therefore, introducing statistical theory and methods of the approximateprobability reliability,which is a better way to solve the dispersion and uncertainty of themagnetic memory parameters in the quantitative evaluation.Combining the magnetic memory testing technology and the theory of reliability, andusing the advantages of the magnetic memory technology in the early damage detection.Thepaper carrys out the weld reliability evaluation based on the magnetic memory parameters.Firstly, extracting the magnetic memory feature signals of welded joint cumulative damageparts on the basis of weld static and fatigue experimental data, and analysing feature ofmagnetic memory parameter in the welded joint critical fracture characteristics, achieving thevariation of the magnetic memory under the static and fatigue loading experiment.In order toverify characterization of the welded joint magnetic memory signal to the stress distribution,applying ANSYS software to simulate the weld stress distribution of the static and fatigueloads, comparing and researching correlation of the welded joint magnetic memory leakageplane distribution fig and the stress distributioncloud image. Lastly, analysing the weldedjoint reliability under fatigue load, making optimization algorithm on the basis of theinterference reliability model, founding the welded joint reliability model based on magneticmemory parameters, computing the welded joint reliability, and providing the theoreticalbasis and methods for the actual welded joint reliability projects of assessment. Thus, thisstudy not only has important theoretical value, but also has broad application prospects.