Font Size: a A A

Evaluation Of Material 16Mo3 Creep Properties By Small Punch Technology Based On K-R Damage Model

Posted on:2017-05-06Degree:MasterType:Thesis
Country:ChinaCandidate:Y H XuFull Text:PDF
GTID:2272330482998687Subject:Chemical Process Equipment
Abstract/Summary:PDF Full Text Request
Many equipments and components worked under high-temperature in power, The creep properties of these equipments and components needs to be assessed. When assess those pressure vessels in-service, the main advantages of small punch method are that only small amount of amount of material is needed and no repair is required afterwards. In the creep performance analysis and life assessment of the material, creep damage is an imp ortant question worth considering. This paper gives an overview of the research work on the determination of K-R creep damage parameters of material 16Mo3.Present work describes an approach to identify creep properties of the material with finite element simulations and neural networks. The small punch test was used to determine the material state under loading. The method could avoid the transformation error of experimental data with experience relationship.The finite element model that the study used was K-R creep damage model. Subroutines were compiled based on the properties of 16Mo3 and the damage constitutive equation of K-R, and coupled into ABAQUS-UMAT module for finite element calculation. The finite element method was used to simulate the time deflection curve depending on the k-R parameters. The time deflection curve was transferred to a neural network, which was trained using time deflection curves generated by the corresponding material parameters and finite element simulations of the small punch test. Via systematic variation of the material parameters, a data base was built up, by which the neural network was trained. The neural networks can be used to evaluate the time deflection curve of the small punch creep test for a given material parameter set, and the parameters are determined. This is an expanding of neural network in areas of small punch creep tests.This study substituted the creep performance parameters by the neural network to the uniaxial creep model. By comparing the uniaxial creep test and simulated curves, that the identified parameters were suitable for uniaxial creep specimens was proved. This illustrates the reasonability of small punch tests for evaluating material creep properties and uniaxial creep test can be replaced.
Keywords/Search Tags:small punch creep tests, K-R creep damage model, neural network, user subroutine, UMAT
PDF Full Text Request
Related items