PSI - Issue 13

Moritz Lessmann et al. / Procedia Structural Integrity 13 (2018) 1232–1237 Author name / Structural Integrity Procedia 00 (2018) 000–000

1237

6

Figure 6 - Option 1 Failure Assessment Diagram

Figure 7 – Alternative to the FAD approach

3.2. Elasto-Plastic Cracked Body Finite Element Modelling The above outlined J opt1 approach simplifies the determination of a tolerable defect size and is recommended for load-controlled cases were an elasto-plastic analysis is not justified or possible. For displacement-controlled load cases in excess of yield, the FAD approach, informed by elastic CBFEA contains inherent conservatisms. Strain based assessment methods have been proposed as an alternative to the more conventional FAD [Budden]. However, it is also often possible to relieve significant conservatism by conducting a fully elasto-plastic analysis to determine J ep directly and compare this against the material allowable J mat . Figure 7 presents a comparison of J opt1 and J ep for a range of defect lengths under identical primary loading. The elasto-plastic analysis is shown to release significant conservatism as the defect length increases. A noticeable shift in the limiting defect size is observed as a result of the more detailed elasto-plastic analysis. Previous experience of the authors has shown that the benefits in conducting elasto-plastic analyses is greatest for components and loading scenarios where the response is strongly ductile and the failure assessment points would lie close to the L r limit. When conducting elasto-plastic CBFEA, it is further recommended that sensitivities to the material model are assessed. For load controlled cases, a lower bound stress-strain model generally provides a conservative result. How ever, for displacement controlled load cases, such as thermal shock loading, where secondary stresses dominate, an upper bound material model may return higher J-Integral results.It is important to note that whilst an elasto-plastic CBFEA can reduce conservatisms and simplifies the interpretation of results, a limit load analysis remains necessary to ensure the limiting defect size is not bounded by plastic collapse. 4. Conclusions Finite element cracked-body meshing techniques have been outlined, which can significantly improve the ease of mesh generation and the mesh quality of complex three-dimensional components. The adoption of a mixed element type meshing strategy has been shown to not affect results, giving confidence in results obtained from such models. The advantage of assessing limiting defect sizes against a material allowable, as opposed through a failure assessment diagram approach, has been demonstrated. Combined, these techniques enable detailed assessments of industrial components, relieving conservatisms by permitting more accurate geometrical modelling. Acknowledgements The authors would like to thank EDF Energy for permission to present this paper. References

Anderson T. L., 2005, Fracture Mechanics, Fundamentals and Applications - Third Edition. Budden P. J., 2005. Failure assessment diagram methods for strain-based fracture, Engineering Fracture Mechanics 73.