The Low-Cycle Fatigue of S960 MC Direct-Quenched High-Strength Steel
Dabiri, Mohammad (2018-10-18)
Väitöskirja
Dabiri, Mohammad
18.10.2018
Lappeenranta University of Technology
Acta Universitatis Lappeenrantaensis
School of Energy Systems
School of Energy Systems, Konetekniikka
Kaikki oikeudet pidätetään.
Julkaisun pysyvä osoite on
https://urn.fi/URN:ISBN:978-952-335-258-2
https://urn.fi/URN:ISBN:978-952-335-258-2
Tiivistelmä
Demanding strength, manufacturability and critical weight limitations are generally the main criteria in the design and fabrication of structural components. The emergence and development of ultra-high and high-strength materials, especially steels benefiting from an excellent high strength to weight ratio, along with their acceptable manufacturability, led to the special role that these materials play in the industry. Although they are so promising for components that experience static loading conditions, their dominant field of application (i.e. mobile machineries and equipment) imposes fluctuating service loads on them, making these structural members susceptible to fatigue failure. In addition, due to the inevitable presence of stress raisers and notches in real components, the high-strength steels, with their high-notch sensitivity, could suffer more compared with other commercial low- and medium-strength steels.
In this study a high-strength steel S960 MC (direct-quenched) is selected for a comprehensive fatigue analysis under constant amplitude loading. The study covers the aspects of microstructural analysis, experimental tests and numerical simulations. The microstructural investigations are performed in order to first characterize the material in question and then analyse the fatigue fracture surfaces. The result of measurements is also used in numerical simulations.
Experiments, such as tensile and low-cycle fatigue tests, are conducted on plain specimens. Using the experimental strain-life curve as the reference, different methods available for the estimation of this important curve are evaluated and a new model, based on artificial neural networks, is proposed. The application of this technique is extended to the estimation of stress concentration factors in butt- and T-welded joints, resulting in models with higher accuracy compared with available parametric equations. The model is also able to explicitly take in to account the effect of axial misalignment (in butt-welded joints) and undercut (in T-welded joints).
Notched specimens made of the same material as that in question are also investigated in the present study. Common analytical approaches, such as Neuber’s rule and the strain energy density method, are used in order to compare their estimations with the experiments and elastoplastic finite element simulations. An approach based on the theory of critical distances (TCD) is also utilized, yielding the best fatigue life estimations when compared to other approaches. In order to investigate the microstructural effects on this method and mainly on its parameter (the material characteristic length), the crystal plasticity formulation is used to model the microstructural heterogeneities at the critical zone of the notch root. It was interesting to observe that the crystal plasticity finite element (CPFE) model is also able to be used coupled with the TCD concept in order to estimate the material’s characteristic length and to perform the notch fatigue analysis. This is a valuable finding, showing the possibility of simultaneously utilizing the elastoplastic TCD with CPFE models, which are so demanding and extensively used in multiscale modelling of fatigue and failure of metals.
In this study a high-strength steel S960 MC (direct-quenched) is selected for a comprehensive fatigue analysis under constant amplitude loading. The study covers the aspects of microstructural analysis, experimental tests and numerical simulations. The microstructural investigations are performed in order to first characterize the material in question and then analyse the fatigue fracture surfaces. The result of measurements is also used in numerical simulations.
Experiments, such as tensile and low-cycle fatigue tests, are conducted on plain specimens. Using the experimental strain-life curve as the reference, different methods available for the estimation of this important curve are evaluated and a new model, based on artificial neural networks, is proposed. The application of this technique is extended to the estimation of stress concentration factors in butt- and T-welded joints, resulting in models with higher accuracy compared with available parametric equations. The model is also able to explicitly take in to account the effect of axial misalignment (in butt-welded joints) and undercut (in T-welded joints).
Notched specimens made of the same material as that in question are also investigated in the present study. Common analytical approaches, such as Neuber’s rule and the strain energy density method, are used in order to compare their estimations with the experiments and elastoplastic finite element simulations. An approach based on the theory of critical distances (TCD) is also utilized, yielding the best fatigue life estimations when compared to other approaches. In order to investigate the microstructural effects on this method and mainly on its parameter (the material characteristic length), the crystal plasticity formulation is used to model the microstructural heterogeneities at the critical zone of the notch root. It was interesting to observe that the crystal plasticity finite element (CPFE) model is also able to be used coupled with the TCD concept in order to estimate the material’s characteristic length and to perform the notch fatigue analysis. This is a valuable finding, showing the possibility of simultaneously utilizing the elastoplastic TCD with CPFE models, which are so demanding and extensively used in multiscale modelling of fatigue and failure of metals.
Kokoelmat
- Väitöskirjat [1076]