Determination Of Elastoplastic Properties Of X80 Pipeline Steel Welded Joint Based On Small Punch Test

With the rapid growth of demand for natural gas in China,it is an inevitable trend for the development of gas pipeline construction to adopt high-grade and large-diameter pipeline for high-pressure gas transmission.As the main means of pipeline connection,it is prone to cause welding defects such as porosity and incomplete penetration when welding,which can easily lead to pipeline fracture during service then cause disastrous consequences.Because the welded joints can’ t provide the samples required for the standard mechanical properties test,it is an urgent problem with evaluating the mechanical properties of the materials in different areas of the in-service pipeline welded joints,which is of great significance to ensure the long-term effective,safe and stable work of the pipeline.As a representative micro sample test method,standard samples can be taken from different zones of welded joint for test with its advantage of small sample size.This paper suggests two new methods on the mechanical properties of X80 pipeline steel welded joint combined with small punch test,the first method is empirical correlation method and the second is Back Propagation(BP)neural network.The main research work of this paper is as follows:(1)Based on the assumption of volume invariance of metal plastic deformation and Hollomon formula,the true stress and true strain of material are rewritten as the function of elastic modulus,yield strength and tensile strength,which achieve the goal of obtaining the true stress-strain of material with small punch test.And the results can also be applied to the finite element simulation to obtain other mechanical properties parameters.(2)A two-dimensional axisymmetric finite element model of small punch test with GTN damage parameters is established,and the accuracy of the model is verified by experiments.Firstly,by changing the parameter K and n in Hollomon formula Systematically,2261 groups of hypothetical materials constitutive relations with different elastic-plastic mechanical properties are obtained.Secondly,the obtained constitutive relations are substituted into the verified small punch test finite element model,and 2261 load-displacement curves of different hypothetical materials corresponding to the true stress-strain curves are obtained.Finally,the two groups of data are substituted into BP neural network and the corresponding relationship between the load-displacement curve of small punch test and the true stress-strain curve of the conventional uniaxial tensile test is established by the method proposed.This relationship can be used to obtain the true stress-strain curve of pipeline steel base on the results of small punch test,and the maximum error is only 5.7% comparing to the results of uniaxial tensile test.(3)Specimens required for uniaxial tensile test and small punch test were carried out in weld zone,heat affected zone and base metal zone respectively.The results were analyzed by empirical correlation method and BP neural network method,and the true stress-strain curves of materials in different zones were obtained,which provided accurate parameters for finite element simulation of welded joint.And reveals the distribution of elastic-plastic mechanical properties at the welded joint: weld zone is the best performance of the welded joint,the base material zone is the second,the heat-affected zone is the worst.It indicates that the welded joint heat-affected zone is the most vulnerable to damage,so that cause safety accidents,but also provide a reference for the improvement of pipeline welding technology.