Search

Realistic description of heterogeneous material behavior demands more accurate modeling at macroscopic and microscopic scales. In this frame, the multiscale techniques employing homogenization scheme offer several solutions. Most recently developed two-scale scheme employing second-order homogenization requires the nonlocal theory at the macrolevel, while the classical local continuum theory is kept at the microlevel. In this paper, a new second-order computational homogenization scheme is proposed employing the higher-order theory at both macro- and microlevel. Discretization is performed by means of the C 1 finite element developed using the strain gradient theory. The new gradient boundary conditions employed on representative volume element (RVE) are derived. The relation between the internal length scale parameter and the RVE size has been found. The new procedure is tested on a benchmark example, where the results have been compared to the solutions obtained by the usual second-order homogenization using the local concept on the RVE. [ABSTRACT FROM AUTHOR]

In the present study, a three-dimensional numerical simulation of a T-joint fillet welding of two plates is performed using simplified material properties. To investigate the influence of the temperature-dependent thermo-mechanical material properties of carbon steel on the transient temperature distributions and weld-induced residual stresses and deformations, two different cases are considered. In the first case, all the material properties, except the yield stress, are taken as constant at room temperature. In the second case, to calculate the temperature-dependent material properties, the polynomials defined by the ASCE manual are used. From the simulation results, it is concluded that, compared with the reference case that uses the true properties of the chosen material S355JR, the first case of simplification is more suitable for the prediction of plate deflections, while the second case is more favorable for the assessment of thermal fields, heat-affected zones, residual stress fields and stress intensity factors. [ABSTRACT FROM AUTHOR]

This paper presents experimental and numerical investigation of elasto-plastic-damage behaviour of aluminium alloys. The experimental procedure includes static and dynamic tensile tests at different strain rates as well as three point bending tests. Numerical modelling of deformation and failure process of the flat specimens is conducted by using non-isothermal elastoplastic damage constitutive model and two-dimensional plane stress finite elements. During the experiment the displacement and temperature distribution on the specimen's surface is measured by digital image correlation (DIC) method and infrared thermography (IR). This has enabled more precise calibration of material parameters in constitutive relations. [ABSTRACT FROM AUTHOR]

Highlights: [•] A numerical/experimental study of welding stresses and distortions is conducted. [•] A shell/3D technique for welding simulation of T-joint fillet weld is investigated. [•] Size of 3D zone in shell/3D model does not influence temperature field distribution. [•] The minimal size of the 3D zone must be chosen according to the stress criterion. [•] For T-joint welded plates 3D zone size must be minimum 3 times the wall thickness size. [ABSTRACT FROM AUTHOR]

• Experimental and numerical research of mechanical behavior of bone was conducted. • Multiple creep/recovery tests at different stress levels were performed. • Threshold stress values for viscoplastic strain and damage accumulation were estimated. • One-dimensional viscoelastic/viscoplastic/damaging constitutive model was developed. • Numerical results at most stress levels are in good agreement with experimental ones. An experimental and numerical study of the creep behavior of the bovine metatarsal cortical bone is presented. The experimental procedure included the multiple cycle tensile creep-recovery tests at eleven creep stress levels between 20 and 120 MPa. The threshold values of stress above which viscoplastic strain and damage start to accumulate are determined. Furthermore, based on the experimental data, a one-dimensional constitutive model has been developed. We demonstrate that Abdel-Tawab's viscoelasticity / damage model combined with the viscoplastic model proposed by Zapas and Crissman can predict the nonlinear behavior of cortical bone tissue subjected to creep-recovery loading conditions. Within the framework of numerical investigations, an efficient algorithm for the integration of the proposed constitutive model at the material point level is derived and implemented into the finite element software ABAQUS using user subroutines. The computational algorithm shows a good capability to describe the tensile behavior of bovine cortical bone for the specific mechanical conditions analyzed. [ABSTRACT FROM AUTHOR]

Highlights • Recently developed phase-field staggered algorithm for fracture modeling is used. • Simplified geometry of a nodular cast iron microstructure is analyzed. • Certain size-effect is observed on various sample sizes with average nodule content. • Capabilities of the algorithm to predict complex crack paths are demonstrated. • Stopping criterion is essential for objective, increment-size independent results. Abstract The numerical simulation of fracture phenomena occurring in real microstructures of heterogeneous materials is a particularly complicated problem, involving complex cracking processes. On the basis of recent investigations, it is obvious that the phase-field approach has a strong potential to model these processes. However, it requires fine spatial discretization to resolve the smooth transition of the diffusive crack representation regulated by a small length scale parameter. Thus, it tends to be computationally intensive when combined with an inefficient solution scheme. In this paper, recently developed staggered solution procedure based on the residual norm control has been employed for the fracture analysis of heterogeneous microstructure exhibiting crack initiation and complex crack paths. Four different sample sizes have been analyzed, chosen from the simplified geometry of a nodular cast iron microstructure where the size-effect has been observed. The detailed discussions regarding the accuracy and CPU time usage have been given. An improvement in computational efficiency is demonstrated in comparison to the common single iteration staggered algorithm. [ABSTRACT FROM AUTHOR]

Highlights • The phase-field staggered algorithm with a residual norm based stopping criterion. • An arbitrary porous steel microstructure exhibiting complex crack paths is analyzed. • The stopping criterion is highly important for achieving accuracy and efficiency. • Verification by numerical and experimental results available in literature. • Numerical implementation takes advantage of Abaqus FE software features. Abstract The phase-field approach to fracture modeling circumvents the crack-surface tracking problem by introducing the phase-field variable which separates the broken and unbroken material states through a smooth transition. However, very fine spatial discretization is required to resolve the smooth distribution of the phase-field regulated by a small length scale parameter. Thus, it can be computationally rather expensive when paired with an inefficient solution scheme. This contribution presents a comprehensive discussion on the use of a stopping criterion within the broadly used staggered algorithm. The stopping criterion of the iterative scheme based on the control of the residual norm is introduced and implemented in the finite element software Abaqus. It alleviates the problem of the common single iteration staggered algorithm, which requires fine loading incrementation to produce an accurate solution. The model verification is conducted on several standard benchmark tests and a porous microstructure exhibiting complex crack phenomena. The detailed discussions regarding the proposed implementation's accuracy and CPU time usage are provided, demonstrating an improvement in computational efficiency with accurate results no longer dependent on the careful selection of loading incrementation. The algorithm codes in the form of the Abaqus user subroutines UMAT and UEL are publicly available on Mendeley data repository linked to this work. [ABSTRACT FROM AUTHOR]

This paper presents an efficient finite element procedure for the prediction of welding-induced residual stresses and distortions in large structures. It is based on a prescribed temperature approach using some features of an Abaqus extension called Abaqus Welding Interface to significantly improve the computational efficiency and speed up the normally time-consuming and cumbersome welding analysis setup performed by the user. To validate the temperature and residual stress solutions obtained by the presented method, two numerical examples are analyzed. Comparison is made with the experimental measurements and the results obtained by the heat generation rate approach using the element birth and death technique. The first example is a butt-welding of two plates, while the second is a T-joint fillet welding of two plates. The results obtained by the proposed procedure demonstrate a good agreement in comparison with the heat generation rate approach as well as the experimental measurements. Furthermore, the computational efficiency is remarkably improved compared to the heat generation rate approach as the CPU time is reduced up to ~70% in both examples. [ABSTRACT FROM AUTHOR]

The paper presents an example of the possibility of modifying a heat source model, which resulted from the differences between the shape of a standard model in the shape of a double ellipsoid and the geometry of a liquid metal pool in the metal-cored arc welding process. Due to the fact that the heat source models implemented in the software do not always allow to reproduce the actual bead geometry, it was necessary to modify these models and signal the possibilities with which the results of the analyses can be improved. The results of modeling with a standard model in the shape of a double ellipsoid as well as a model consisting of two two-ellipsoidal models and modification of heat input by heating the selected elements of the model mesh are presented. The use of the proposed solutions of heat source models, which allow to reproduce the shape of the modeled molten pool, also allows to increase the precision of the calculated maximum values of the heat cycle temperature, reducing the difference by an order of magnitude from 18 to 22 % in relation to the values measured for the standard model, to several percent (1.2–3.3 %) in the case of combined models. • A new approach to defining the heat source model is proposed. • An improvement in the quality of mapping the shape of the molten metal pool was achieved. • The simplicity of application of the proposed method of heat source model facilitates its use in industrial applications. [ABSTRACT FROM AUTHOR]

• The fatigue damage behaviour of AA 5083-H111 alloy is investigated experimentally and computationally. • Material damage parameters are obtained for the inelastic energy approach. • A computational model is developed for the damage evolution period. • The proposed model is validated by the comparison of the computational and experimental results. • The proposed model may be used to simulate the LCF-behaviour of engineering components made of AA 5083-H111 alloy. This study presents the experimental and computational investigation of a damage evaluation of the ductile aluminium alloy AA 5083-H111. The Compact Tensile (CT) specimen was selected for experimental testing and computational analysis. The proposed computational model is based on the inelastic energy approach, including the damage initiation and damage evolution criteria. In the computational model, the direct cyclic algorithm in the framework of Abaqus software was used to obtain the stabilised cyclic response of the analysed specimen subjected to the cyclic loading. Furthermore, the research presented in this study was focused mainly on the damage evolution period and subsequent determination of the appropriate maximum damage variable D max , which defines the stage of the stiffness degradation in the analysed structure, and, therefore, influence significantly on the damage evolution process. The comparison between the experimentally and computationally obtained damage evolution period and also the damage path in the analysed CT specimen showed a reasonable agreement. Therefore, it can be concluded that the proposed computational model was validated, and could be used for the further low cycle fatigue analyses of structural components made of the aluminium alloy AA 5083-H111. [ABSTRACT FROM AUTHOR]

• Recently developed generalised phase-field fracture model is employed. • Realistic microstructural geometries of nodular cast iron are analysed. • Model is able to recover brittle/ductile fracture in monotonic and fatigue regime. • By changing material properties, a brittle to ductile fracture transition is observed. • Framework is able to solve contact problems coupled with fracture analysis. The aim of this work is to model complex fracture and fatigue processes in microstructural geometries of nodular cast iron. Herein, recently developed generalised phase-field formulation for modelling fracture in brittle and ductile solids subjected to both monotonic and cyclic loading is employed. Three different sized microstructural specimens are analysed using different modelling options to investigate the transition between brittle and ductile fracture material behaviour. The results demonstrate the ability of the proposed model to reproduce crack nucleation and complex crack propagation patterns. Moreover, the low- and high-cyclic fatigue regime features are presented in terms of fracture patterns and Wöhler-type curve. [ABSTRACT FROM AUTHOR]

• Fatigue analysis of valve bodies DN100 and DN50. • Crack initiation on a valve body according to Basquin-Coffin-Manson equation. • Stress intensity factor estimation with eXtended Finite Element Method (XFEM). • Cycles for crack growth across the valve body thickness. • The engineering and the true stress–strain diagrams and cycle fatigue behaviour of GP240GH steel. Global industry depends of the different type of valves, such as globe valves which are used for precise flow regulation. Since most of the power plants rely on performance of globe valves, it is necessary that valves function without interruption. In this paper, the main part of the globe valve, valve body, is investigated. Simulation of crack initiation and its growth in a wall of valve body is carried out. Required number of cycles for crack initiation is estimated according to Basquin-Coffin-Manson equation. Afterwards, stress intensity factor is estimated through eXtended Finite Element Method (XFEM) and required number of cycles for crack growth across the valve body thickness is calculated according to Paris' law. The aforementioned method is validated on the results from the available literature by employing relatively simple shapes such as steel pipe with integrated flaw. Subsequently, the method is used on a relatively complex geometry, as is the valve body. The obtained results indicate lack of brittle fracture in a valve body. In other words, the results imply occurrence of Leakage Before Brake (LBB). This is utterly important for power plant industries, since it ensures the proper reaction of the user and possibly avoids possible catastrophic consequences for personnel and the equipment located in the proximity of the valve. [ABSTRACT FROM AUTHOR]

A damage model for quasi-brittle materials embedded into the two dimensional C 1 continuity triangular finite element formulation based on the strain gradient continuum theory is considered. The isotropic damage law is applied to the higher-order stress-strain constitutive model, which enables the analysis of both homogeneous and heterogeneous materials. Such softening formulation also ensures a decrease of the intensity of the material nonlocality associated with the damage growth, which is necessary for the correct description of the narrow localized deformation. In order to obtain the required constitutive matrices, the second-order homogenization procedure is applied to the various representative volume elements in the frame of a multiscale approach. The derived finite element formulation is implemented into the finite element program ABAQUS by means of user subroutines. The superior regularization capabilities, as well as the accuracy and efficiency of the proposed higher-order gradient damage model are demonstrated by the standard benchmark examples. [ABSTRACT FROM AUTHOR]

To assess the behaviour of porous materials, it is important to consider the influence of material microstructure. In this study, a multiscale experimental analysis is conducted to investigate the influence of micromechanical parameters, such as the material properties of metallic phases and porosity, on the macromechanical properties of a porous sintered steel. At the microlevel, indentation is conducted to identify microphases and metallographic analysis has been used on mesoscale to measure porosity. Tension uniaxial tests have been applied to characterize macro properties. The results of the tests suggest that the macromechanical properties of porous sintered metallic materials might be mostly influenced by the porosity, while at the microlevel no significant variation in the mechanical properties of microphases has been detected in samples with different density, i.e., porosity. In addition, porosity is related to the sintering density experimentally and an empirical expression that relates density and porosity is proposed. [Display omitted]. [ABSTRACT FROM AUTHOR]

• A model is proposed for bending fatigue life prediction of surface hardened spur gears. • Finite element, elastic–plastic correction, and strain-life methods are used. • The multilayer method is employed to model the surface hardened layer. • Surface from subsurface bending fatigue failure is distinguished. • The proposed model is validated against experimental results. A computational model for predicting both the location and the required number of cycles for bending fatigue failure in surface hardened spur gears is proposed. Linear elastic stresses and strains in a single tooth bending fatigue test are corrected for elastic–plastic material behavior. The tooth root region of spur gear is divided into layers. Fatigue properties are assigned to each layer via the hardness method. Based on the multiaxial fatigue criteria, fatigue failure location and corresponding fatigue lives are estimated. Predicted fatigue lives, failure locations, and transition from surface to subsurface fatigue failure show good agreement with the experimental results. [ABSTRACT FROM AUTHOR]