Browse ORBi

- What it is and what it isn't
- Green Road / Gold Road?
- Ready to Publish. Now What?
- How can I support the OA movement?
- Where can I learn more?

ORBi

A unified enrichment approach addressing blending and conditioning issues in enriched finite elements ; ; Bordas, Stéphane in Computer Methods in Applied Mechanics and Engineering (2019), 349 We present a combination of techniques to improve the convergence and conditioning properties of partition of unity (PU) enriched finite element methods. By applying these techniques to different types of ... [more ▼] We present a combination of techniques to improve the convergence and conditioning properties of partition of unity (PU) enriched finite element methods. By applying these techniques to different types of enrichment functions, namely polynomial, discontinuous and singular, higher order convergence rates can be obtained while keeping condition number growth rates similar to the ones corresponding to standard finite elements. [less ▲] Detailed reference viewed: 78 (0 UL)Model order reduction accelerated Monte Carlo stochastic isogeometric method for the analysis of structures with high-dimensional and independent material uncertainties Ding, Chensen ; ; et al in Computer Methods in Applied Mechanics and Engineering (2019), 349 Structural stochastic analysis is vital to engineering. However, current material related uncertainty methods are mostly limited to low dimension, and they mostly remain unable to account for spatially ... [more ▼] Structural stochastic analysis is vital to engineering. However, current material related uncertainty methods are mostly limited to low dimension, and they mostly remain unable to account for spatially uncorrelated material uncertainties. They are not representative of realistic and practical engineering situations. In particular, it is more serious for composite structures comprised of dissimilar materials. Therefore, we propose a novel model order reduction via proper orthogonal decomposition accelerated Monte Carlo stochastic isogeometric method (IGA-POD-MCS) for stochastic analysis of exactly represented (composite) structures. This approach particularly enables high-dimensional material uncertainties wherein the characteristics of each element are independent. And the novelties include: (1) the structural geometry is exactly modeled thanks to isogeometric analysis (IGA), as well as providing more accurate deterministic and stochastic solutions, (2) we innovatively consider high-dimensional and independent material uncertainties by separating the stochastic mesh from the IGA mesh, and modeling different stochastic elements to have different (independent) uncertainty behaviors, (3) the classical Monte Carlo simulation (MCS) is employed to universally solve the high-dimensional uncertainty problem. However, to circumvent its computational expense, we employ model order reduction via proper orthogonal decomposition (POD) into the IGA coupled MCS stochastic analysis. In particular, we observe that this work decouples all IGA elements and hence permits independent uncertainty models easily, thereby the engineering problem is modeled to be more realistic and authentic. Several illustrative numerical examples verify the proposed IGA-POD-MCS approach is effective and efficient; and the larger the scale of the problem is, the more advantageous the method will become. [less ▲] Detailed reference viewed: 37 (2 UL)A simple and robust computational homogenization approach for heterogeneous particulate composites ; ; et al in Computer Methods in Applied Mechanics and Engineering (2019), 349 In this article, a computationally efficient multi-split MsXFEM is proposed to evaluate the elastic properties of heterogeneous materials. The multi-split MsXFEM is the combination of multi-split XFEM ... [more ▼] In this article, a computationally efficient multi-split MsXFEM is proposed to evaluate the elastic properties of heterogeneous materials. The multi-split MsXFEM is the combination of multi-split XFEM with multiscale finite element methods (MsFEM). The multi-split XFEM is capable to model multiple discontinuities in a single element which leads to reduction in the number of mesh elements, whereas MsFEM helps in reducing the computational time. Strain energy based homogenization has been implemented on an RVE (having volume fraction of heterogeneities up to 50%) for evaluating the elastic properties. From macro-element size analysis, we estimate that the RVE edge length must be 5 times the edge length of the macro-element. The directional analysis has been performed to verify the isotropic behavior of the material, whereas contrast analysis has been done to check the numerical accuracy of the proposed scheme. A level set correction (LSC) based on higher order shape functions has been proposed to reduce mapping errors of level set values. It is also observed that multi-split MsXFEM is about 16 times computationally more efficient than MsXFEM for 50% volume of heterogeneities. [less ▲] Detailed reference viewed: 53 (0 UL)Modelling Complex Systems: a primer - agent-based models, equation-based models, statistical models and Bayesian inference, digital twins Bordas, Stéphane Learning material (2019) Modelling Complex Systems: a primer - agent-based models, equation-based models, statistical models and Bayesian inference, digital twins Detailed reference viewed: 581 (24 UL)B-Spline FEM for Time-Harmonic Acoustic Scattering and Propagation ; ; Bordas, Stéphane in Journal of Theoretical and Computational Acoustics (2019), 27 We study the application of a B-splines Finite Element Method (FEM) to time-harmonic scattering acoustic problems. The infinite space is truncated by a fictitious boundary and second-order Absorbing ... [more ▼] We study the application of a B-splines Finite Element Method (FEM) to time-harmonic scattering acoustic problems. The infinite space is truncated by a fictitious boundary and second-order Absorbing Boundary Conditions (ABCs) are applied. The truncation error is included in the exact solution so that the reported error is an indicator of the performance of the numerical method, in particular of the size of the pollution error. Numerical results performed with high-order basis functions (third or fourth order) showed no visible pollution error even for very high frequencies. To prove the ability of the method to increase its accuracy in the high frequency regime, we show how to implement a high-order Padé-type ABC on the fictitious outer boundary. The above-mentioned properties combined with exact geometrical representation make B-Spline FEM a very promising platform to solve high-frequency acoustic problems. [less ▲] Detailed reference viewed: 62 (1 UL)A one point integration rule over star convex polytopes ; ; et al in Computers and Structures (2019), 215 In this paper, the recently proposed linearly consistent one point integration rule for the meshfree methods is extended to arbitrary polytopes. The salient feature of the proposed technique is that it ... [more ▼] In this paper, the recently proposed linearly consistent one point integration rule for the meshfree methods is extended to arbitrary polytopes. The salient feature of the proposed technique is that it requires only one integration point within each n-sided polytope as opposed to 3n in Francis et al. (2017) and 13n integration points in the conventional approach for numerically integrating the weak form in two dimensions. The essence of the proposed technique is to approximate the compatible strain by a linear smoothing function and evaluate the smoothed nodal derivatives by the discrete form of the divergence theorem at the geometric center. This is done by Taylor’s expansion of the weak form which facilitates the use of the smoothed nodal derivatives acting as the stabilization term. This translates to 50% and 30% reduction in the overall computational time in the two and three dimensions, respectively, whilst preserving the accuracy and the convergence rates. The convergence properties, the accuracy and the efficacy of the one point integration scheme are discussed by solving few benchmark problems in elastostatics. [less ▲] Detailed reference viewed: 67 (0 UL)Model I cohesive zone models of different rank coals ; Lian, Haojie ; et al in International Journal of Rock Mechanics and Mining Sciences (2019), 115 The present work develops cohesive zone models (CZM), i.e. cohesion-separation laws, for mode I fractures in different rank coals, including weakly caking coals, gas coals, fat coals, meager-lean coals ... [more ▼] The present work develops cohesive zone models (CZM), i.e. cohesion-separation laws, for mode I fractures in different rank coals, including weakly caking coals, gas coals, fat coals, meager-lean coals and anthracite, through disk-shaped compact tension tests. Firstly, the experiments show that with the coal rank rising, the critical crack separation displacements and the degrees of the nonlinearity of the softening function decline gradually. By fitting the experimental data with the four commonly used cohesive zone models including the power law, the exponential law, the bilinear law and the linear law, the best-fitted model for each rank of coals was identified and the corresponding parameters were found. Secondly, to arrive at a general CZM formulation for the different rank coals, Karihaloo’s polynomial law was employed, which also gave better fit to the experimental data compared with the aforementioned four CZMs. After obtaining the CZM for coals, fracture energy was evaluated which is equal to the area under the softening curve. With the increase of the coal rank, the fracture energy reduces but its coefficient of variation increases. Thirdly, the geometric characteristics of fractures in different rank coals are studied. The lower rank coals have more tortuous crack propagation paths and larger roughness coefficients, whereas the higher rank coals possess wider average fracture apertures. Lastly, in order to further test the applicability of the obtained cohesion-separation laws, we implemented the Karihaloo’s polynomial CZM and the bilinear CZM into the cohesive elements of ABAQUS® using the user-subroutine VUMAT, and thereby simulated the crack propagation in single-edge notched beams made of weakly caking coals, fat coals, and meager-lean coals, respectively. It is found that the numerical results based on Karihaloo’s polynomial CZM have a better agreement with the experimental data than the bilinear CZM [less ▲] Detailed reference viewed: 57 (0 UL)Meshing or not meshing - Iso/sub/super-geometric analysis (adaptive unfitted methods for real-time simulations) Immersed collocation methods
Bordas, Stéphane Scientific Conference (2019, January 04) Detailed reference viewed: 274 (41 UL)A Tutorial on Bayesian Inference to Identify Material Parameters in Solid Mechanics Rappel, Hussein ; Beex, Lars ; Hale, Jack et al in Archives of Computational Methods in Engineering (2019) The aim of this contribution is to explain in a straightforward manner how Bayesian inference can be used to identify material parameters of material models for solids. Bayesian approaches have already ... [more ▼] The aim of this contribution is to explain in a straightforward manner how Bayesian inference can be used to identify material parameters of material models for solids. Bayesian approaches have already been used for this purpose, but most of the literature is not necessarily easy to understand for those new to the field. The reason for this is that most literature focuses either on complex statistical and machine learning concepts and/or on relatively complex mechanical models. In order to introduce the approach as gently as possible, we only focus on stress–strain measurements coming from uniaxial tensile tests and we only treat elastic and elastoplastic material models. Furthermore, the stress–strain measurements are created artificially in order to allow a one-to-one comparison between the true parameter values and the identified parameter distributions. [less ▲] Detailed reference viewed: 667 (106 UL)Identifying elastoplastic parameters with Bayes' theorem considering double error sources and model uncertainty Rappel, Hussein ; Beex, Lars ; et al in Probabilistic Engineering Mechanics (2019), 55 We discuss Bayesian inference for the identi cation of elastoplastic material parameters. In addition to errors in the stress measurements, which are commonly considered, we furthermore consider errors in ... [more ▼] We discuss Bayesian inference for the identi cation of elastoplastic material parameters. In addition to errors in the stress measurements, which are commonly considered, we furthermore consider errors in the strain measurements. Since a difference between the model and the experimental data may still be present if the data is not contaminated by noise, we also incorporate the possible error of the model itself. The three formulations to describe model uncertainty in this contribution are: (1) a random variable which is taken from a normal distribution with constant parameters, (2) a random variable which is taken from a normal distribution with an input-dependent mean, and (3) a Gaussian random process with a stationary covariance function. Our results show that incorporating model uncertainty often, but not always, improves the results. If the error in the strain is considered as well, the results improve even more. [less ▲] Detailed reference viewed: 375 (67 UL)Geometrical and material uncertainties for the mechanics of composites ; Bordas, Stéphane ; et al Scientific Conference (2019) Detailed reference viewed: 129 (15 UL)h- and p-adaptivity driven by recovery and residual-based error estimators for PHT-splines applied to time-harmonic acoustics ; ; et al in Computers and Mathematics with Applications (2018), 77(9), 2369-2395 In this work, we demonstrate the application of PHT-splines for time-harmonic acoustic problems, modeled by the Helmholtz equation. Solutions of the Helmholtz equation have two features: global ... [more ▼] In this work, we demonstrate the application of PHT-splines for time-harmonic acoustic problems, modeled by the Helmholtz equation. Solutions of the Helmholtz equation have two features: global oscillations associated with the wave number and local gradients caused by geometrical irregularities. We show that after a sufficient number of degrees of freedom is used to approximate global oscillations, adaptive refinement can capture local features of the solution. We compare residual-based and recovery-based error estimators and investigate the performance of -refinement. The simulations are done in the context of recently introduced Geometry Independent Field approximaTion (GIFT), where PHT-splines are only used to approximate the solution, while the computational domain is parameterized with NURBS. This approach builds on the natural adaptation ability of PHT-splines and avoids the re-parameterization of the NURBS geometry during the solution refinement process. [less ▲] Detailed reference viewed: 83 (0 UL)The influence of fracture geometry variation on non-Darcy flow in fractures under confining stresses ; Lian, Haojie ; et al in International Journal of Rock Mechanics and Mining Sciences (2018), 113 To investigate the influence of geometric characteristics of deformable rough fractures under confining stresses on the behaviors of non-Darcy flow, four fractured sandstone specimens were used for ... [more ▼] To investigate the influence of geometric characteristics of deformable rough fractures under confining stresses on the behaviors of non-Darcy flow, four fractured sandstone specimens were used for hydraulic tests in the experiments. According to the experimental results of the relationships between the hydraulic gradient and the flow rate, it is demonstrated that the Forchheimer's equation can offer a good description of the non-Darcy flow in rough fractures. In addition, the coefficients A and B in Forchheimer's equation are sensitive to the fracture geometric characteristics, and their values also increase as the confining stress rises, mainly owing to the reduction of the hydraulic aperture and the heterogeneous distribution of the interconnected void areas with the confining stress rising. Then, the surface and interior geometric properties of rough fractures were quantitatively characterized with the peak asperity height and the box-counting fractal dimension of the heterogeneous distribution of the interconnected void areas, respectively. Furthermore, an empirical relationship between the fractal dimension D and the fracture apertures was constructed according to the experimental results. Lastly, a quantitative model was proposed to represent the relationship between the fracture geometric characteristics and the non-Darcy coefficient . This model was further used to link the non-linear coefficient of Forchheimer's equation and the critical Reynold number with the fracture geometric characteristics. The proposed models were validated by the experimental data and would be helpful to characterize the non-Darcy flow behavior in rough fractures under various confining stresses. [less ▲] Detailed reference viewed: 55 (0 UL)Linear smoothed extended finite element method for fatigue crack growth simulations ; ; et al in Engineering Fracture Mechanics (2018), 206 In this paper, the recently proposed linear smoothed extended finite element method (LSmXFEM) is employed to simulate the fatigue crack growth. Unlike the conventional extended finite element method, the ... [more ▼] In this paper, the recently proposed linear smoothed extended finite element method (LSmXFEM) is employed to simulate the fatigue crack growth. Unlike the conventional extended finite element method, the LSmXFEM does not require special numerical integration technique to integrate the terms in the stiffness matrix. The stress intensity factors (SIFs) are evaluated by using the domain form of the interaction integral technique. The fatigue crack growth rate is evaluated using the generalized Paris’ law in conjunction with the maximum hoop stress criterion. The robustness of the method is demonstrated with a few examples for which the results are available in the literature. Then, the fatigue crack growth from the numerical simulation is compared with the experimental investigations performed on CR5 grade cold formed steel. It is seen that the fatigue life and the crack path obtained from the proposed method is in close agreement with the experimental observation. [less ▲] Detailed reference viewed: 64 (0 UL)A volume-averaged nodal projection method for the Reissner-Mindlin plate model ; ; Hale, Jack et al in Computer Methods in Applied Mechanics and Engineering (2018), 341 We introduce a novel meshfree Galerkin method for the solution of Reissner-Mindlin plate problems that is written in terms of the primitive variables only (i.e., rotations and transverse displacement) and ... [more ▼] We introduce a novel meshfree Galerkin method for the solution of Reissner-Mindlin plate problems that is written in terms of the primitive variables only (i.e., rotations and transverse displacement) and is devoid of shear-locking. The proposed approach uses linear maximum-entropy approximations and is built variationally on a two-field potential energy functional wherein the shear strain, written in terms of the primitive variables, is computed via a volume-averaged nodal projection operator that is constructed from the Kirchhoff constraint of the three-field mixed weak form. The stability of the method is rendered by adding bubble-like enrichment to the rotation degrees of freedom. Some benchmark problems are presented to demonstrate the accuracy and performance of the proposed method for a wide range of plate thicknesses. [less ▲] Detailed reference viewed: 195 (21 UL)Simple and extensible plate and shell finite element models through automatic code generation tools Hale, Jack ; ; Bordas, Stéphane et al in Computers and Structures (2018), 209 A large number of advanced finite element shell formulations have been developed, but their adoption is hindered by complexities of transforming mathematical formulations into computer code. Furthermore ... [more ▼] A large number of advanced finite element shell formulations have been developed, but their adoption is hindered by complexities of transforming mathematical formulations into computer code. Furthermore, it is often not straightforward to adapt existing implementations to emerging frontier problems in thin structural mechanics including nonlinear material behaviour, complex microstructures, multi-physical couplings, or active materials. We show that by using a high-level mathematical modelling strategy and automatic code generation tools, a wide range of advanced plate and shell finite element models can be generated easily and efficiently, including: the linear and non-linear geometrically exact Naghdi shell models, the Marguerre-von K ́arm ́an shallow shell model, and the Reissner-Mindlin plate model. To solve shear and membrane-locking issues, we use: a novel re-interpretation of the Mixed Interpolation of Tensorial Component (MITC) procedure as a mixed-hybridisable finite element method, and a high polynomial order Partial Selective Reduced Integration (PSRI) method. The effectiveness of these approaches and the ease of writing solvers is illustrated through a large set of verification tests and demo codes, collected in an open-source library, FEniCS-Shells, that extends the FEniCS Project finite element problem solving environment. [less ▲] Detailed reference viewed: 533 (47 UL)Corotational cut finite element method for real-time surgical simulation: Application to needle insertion simulation Bui, Huu Phuoc ; Tomar, Satyendra ; Bordas, Stéphane in Computer Methods in Applied Mechanics and Engineering (2018), 345 We present the corotational cut Finite Element Method (FEM) for real-time surgical simulation. The only requirement of the proposed method is a background mesh, which is not necessarily conforming to the ... [more ▼] We present the corotational cut Finite Element Method (FEM) for real-time surgical simulation. The only requirement of the proposed method is a background mesh, which is not necessarily conforming to the boundaries/interfaces of the simulated object. The details of the surface, which can be directly obtained from binary images, are taken into account by a multilevel embedding algorithm which is applied to elements of the background mesh that are cut by the surface. Dirichlet boundary conditions can be implicitly imposed on the surface using Lagrange multipliers, whereas traction or Neumann boundary conditions, which is/are applied on parts of the surface, can be distributed to the background nodes using shape functions. The implementation is verified by convergences studies, of the geometry and of numerical solutions, which exhibit optimal rates. To verify the reliability of the method, it is applied to various needle insertion simulations (e.g. for biopsy or brachytherapy) into brain and liver models. The numerical results show that, while retaining the accuracy of the standard FEM, the proposed method can (1) make the discretisation independent from geometric description, (2) avoid the complexity of mesh generation for complex geometries, and (3) provide computational speed suitable for real-time simulations. Thereby, the proposed method is very suitable for patient-specific simulations as it improves the simulation accuracy by automatically, and properly, taking the simulated geometry into account, while keeping the low computational cost. [less ▲] Detailed reference viewed: 52 (0 UL)The elastic properties of composites reinforced by a transversely isotropic random fibre-network ; ; et al in Composite Structures (2018), 208 This research stems from the idea of introducing a fibre-network structure into composites aiming to enhance the stiffness and strength of the composites. A novel new type of composites reinforced by a ... [more ▼] This research stems from the idea of introducing a fibre-network structure into composites aiming to enhance the stiffness and strength of the composites. A novel new type of composites reinforced by a tranversely isotropic fibre-network in which the fibres are devided into continuous segments and randomly distributed has been proposed and found to have improved elastic properties compared to other conventional fibre or particle composites mainly due to the introduction of cross linkers among the fibres. Combining with the effects of Poisson’s ratio of the constituent materials, the fibre network composite can exhibit extraordinary stiffness. A simplified analytical model has also been proposed for comparison with the numerical results, showing close prediction of the stiffness of the fibre-network composites. Moreover, as a plate structure, the thickness of the fibre network composite is adjustable and can be tailored according to the dimensions and mechanical properties as demanded in industry. [less ▲] Detailed reference viewed: 123 (3 UL)Lack of separation of scales: A view from reduced order modelling and homogenisation
Bordas, Stéphane Speeches/Talks (2018) Detailed reference viewed: 108 (1 UL)Classification of states and model order reduction of large scale Chemical Vapor Deposition processes with solution multiplicity ; ; Beex, Lars et al in Computers and Chemical Engineering (2018), 121 This paper presents an equation-free, data-driven approach for reduced order modeling of a Chemical Vapor Deposition (CVD) process. The proposed approach is based on process information provided by ... [more ▼] This paper presents an equation-free, data-driven approach for reduced order modeling of a Chemical Vapor Deposition (CVD) process. The proposed approach is based on process information provided by detailed, high-fidelity models, but can also use spatio-temporal measurements. The Reduced Order Model (ROM) is built using the method-of-snapshots variant of the Proper Orthogonal Decomposition (POD) method and Artificial Neural Networks (ANN) for the identification of the time-dependent coefficients. The derivation of the model is completely equation-free as it circumvents the projection of the actual equations onto the POD basis. Prior to building the model, the Support Vector Machine (SVM) supervised classification algorithm is used in order to identify clusters of data corresponding to (physically) different states that may develop at the same operating conditions due to the inherent nonlinearity of the process. The different clusters are then used for ANN training and subsequent development of the ROM. The results indicate that the ROM is successful at predicting the dynamic behavior of the system in windows of operating parameters where steady states are not unique. [less ▲] Detailed reference viewed: 111 (9 UL) |
||