Pub Date : 1900-01-01DOI: 10.23967/wccm-eccomas.2020.137
F. Bertrand, H. Schneider
{"title":"Least-Squares Methods for Linear Elasticity: Refined Error Estimates","authors":"F. Bertrand, H. Schneider","doi":"10.23967/wccm-eccomas.2020.137","DOIUrl":"https://doi.org/10.23967/wccm-eccomas.2020.137","url":null,"abstract":"","PeriodicalId":148883,"journal":{"name":"14th WCCM-ECCOMAS Congress","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124977697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.23967/WCCM-ECCOMAS.2020.292
D. Magisano, F. Liguori, L. Leonetti, A. Madeo, G. Garcea, J. Kiendl, A. Reali
. Isogeometric Kirchhoff-Love elements have received an increasing attention in geometrically nonlinear analysis of thin walled structures. They make it possible to meet the C 1 requirement in the interior of surface patches, to avoid the use of finite rotations and to reduce the number of unknowns compared to shear flexible models. Locking elimination, patch coupling and iterative solution are crucial points for a robust and efficient nonlinear analysis and represent the main focus of this work. Patch-wise reduced integrations are investigated to deal with locking in large deformation problems discretized via a standard displacement-based formulation. An optimal integration scheme for third order C 2 NURBS, in terms of accuracy and efficiency, is identified, allowing to avoid locking without resorting to a mixed formulation. The Newton method with mixed integration points (MIP) is used for the solution of the discrete nonlinear equations with a great reduction of the iterative burden and a superior robustness with respect to the standard Newton scheme. A simple penalty approach for coupling adjacent patches, applicable to either smooth or non-smooth interfaces, is proposed. An accurate coupling, also for a non-matching discretization, is obtained using an interface-wise reduced integration while the MIP iterative scheme allows for a robust and efficient
{"title":"Robust and Efficient Large Deformation Analysis of Kirchhoff–love Shells: Locking, Patch Coupling and Iterative Solution","authors":"D. Magisano, F. Liguori, L. Leonetti, A. Madeo, G. Garcea, J. Kiendl, A. Reali","doi":"10.23967/WCCM-ECCOMAS.2020.292","DOIUrl":"https://doi.org/10.23967/WCCM-ECCOMAS.2020.292","url":null,"abstract":". Isogeometric Kirchhoff-Love elements have received an increasing attention in geometrically nonlinear analysis of thin walled structures. They make it possible to meet the C 1 requirement in the interior of surface patches, to avoid the use of finite rotations and to reduce the number of unknowns compared to shear flexible models. Locking elimination, patch coupling and iterative solution are crucial points for a robust and efficient nonlinear analysis and represent the main focus of this work. Patch-wise reduced integrations are investigated to deal with locking in large deformation problems discretized via a standard displacement-based formulation. An optimal integration scheme for third order C 2 NURBS, in terms of accuracy and efficiency, is identified, allowing to avoid locking without resorting to a mixed formulation. The Newton method with mixed integration points (MIP) is used for the solution of the discrete nonlinear equations with a great reduction of the iterative burden and a superior robustness with respect to the standard Newton scheme. A simple penalty approach for coupling adjacent patches, applicable to either smooth or non-smooth interfaces, is proposed. An accurate coupling, also for a non-matching discretization, is obtained using an interface-wise reduced integration while the MIP iterative scheme allows for a robust and efficient","PeriodicalId":148883,"journal":{"name":"14th WCCM-ECCOMAS Congress","volume":"75 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129244597","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.23967/WCCM-ECCOMAS.2020.283
C. Díaz, O. González-Estrada, M. Cely
In this work, the application of an artificial neural network (ANN) is proposed to develop a predicting model for the holdup of a two-phase flow composed of water and mineral oil in a horizontal pipe. For this, the surface velocities of each fluid and the differential pressure in the pipeline are used as input parameters of the multilayer artificial neural network with backpropagation, while the holdup of the fluids is used as the output parameter for the training. A set of 56 experimental data was obtained in the LabPetroCEPETRO-UNICAMP laboratory. The best performing results for the predictive model show a mean absolute error (AAPE) of 3.01% and a coefficient of determination R2 of 0.9964 using 15 neurons in the hidden layer of the network and the TanSig transfer function.
{"title":"Predictive Modeling of Holdup in Horizontal Wateroil Flow Using a Neural Network Approach","authors":"C. Díaz, O. González-Estrada, M. Cely","doi":"10.23967/WCCM-ECCOMAS.2020.283","DOIUrl":"https://doi.org/10.23967/WCCM-ECCOMAS.2020.283","url":null,"abstract":"In this work, the application of an artificial neural network (ANN) is proposed to develop a predicting model for the holdup of a two-phase flow composed of water and mineral oil in a horizontal pipe. For this, the surface velocities of each fluid and the differential pressure in the pipeline are used as input parameters of the multilayer artificial neural network with backpropagation, while the holdup of the fluids is used as the output parameter for the training. A set of 56 experimental data was obtained in the LabPetroCEPETRO-UNICAMP laboratory. The best performing results for the predictive model show a mean absolute error (AAPE) of 3.01% and a coefficient of determination R2 of 0.9964 using 15 neurons in the hidden layer of the network and the TanSig transfer function.","PeriodicalId":148883,"journal":{"name":"14th WCCM-ECCOMAS Congress","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115834452","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.23967/WCCM-ECCOMAS.2020.332
E. Elyasi, A. Perrier, M. Bailet, Y. Payan
Modeling the lower limb muscles using the Finite Element (FE) method is required for various applications including injury mechanisms or when stress/strain distribution in the muscle is of interest. When the muscles are represented with 3D FE models, the interaction between the muscles can be taken into account that has an effect on their force production. However, the computational cost of such a method is considerably high. Besides, in many cases, a major part of this computational cost is committed to gain unnecessary information. For instance, when having two FE muscles in contact, both muscles need to get finely meshed to conserve their surface details even if having the stress/strain distribution inside one of the two muscles is not required. As a result, the current study aims to explore a model reduction technique based on mesh embedding to decrease the computational cost of such models. A combination of Computerized Tomography and Magnetic Resonance Imaging (MRI) data obtained from a volunteer subject was used to generate a musculoskeletal model of the quadriceps muscle group. The modeling process was performed in ArtiSynth which is an opensource 3D modeling platform supporting combined simulation of multibody and finite element models. This platform allows the attachment of a passive embedded mesh to a FE body so that it deforms in accordance with the motion of the FE body. Considering that the external forces applied to the passive mesh can be propagated back to the FE body attached to it, contact can be defined between the embedded mesh and any other structure. A full and a reduced model are generated and are used to simulate a passive deep knee flexion to test the reliability of the method. The kinematic outcomes are compared against data points obtained from MRI at different flexion angles. The results show that the proposed methodology can be considered as a substitute to fully FE models without a substantial sacrifice on the outcomes despite having lower computational cost.
{"title":"Techniques to Reduce The Computational Cost of Finite Element Models of the Lower Limb Muscles","authors":"E. Elyasi, A. Perrier, M. Bailet, Y. Payan","doi":"10.23967/WCCM-ECCOMAS.2020.332","DOIUrl":"https://doi.org/10.23967/WCCM-ECCOMAS.2020.332","url":null,"abstract":"Modeling the lower limb muscles using the Finite Element (FE) method is required for various applications including injury mechanisms or when stress/strain distribution in the muscle is of interest. When the muscles are represented with 3D FE models, the interaction between the muscles can be taken into account that has an effect on their force production. However, the computational cost of such a method is considerably high. Besides, in many cases, a major part of this computational cost is committed to gain unnecessary information. For instance, when having two FE muscles in contact, both muscles need to get finely meshed to conserve their surface details even if having the stress/strain distribution inside one of the two muscles is not required. As a result, the current study aims to explore a model reduction technique based on mesh embedding to decrease the computational cost of such models. A combination of Computerized Tomography and Magnetic Resonance Imaging (MRI) data obtained from a volunteer subject was used to generate a musculoskeletal model of the quadriceps muscle group. The modeling process was performed in ArtiSynth which is an opensource 3D modeling platform supporting combined simulation of multibody and finite element models. This platform allows the attachment of a passive embedded mesh to a FE body so that it deforms in accordance with the motion of the FE body. Considering that the external forces applied to the passive mesh can be propagated back to the FE body attached to it, contact can be defined between the embedded mesh and any other structure. A full and a reduced model are generated and are used to simulate a passive deep knee flexion to test the reliability of the method. The kinematic outcomes are compared against data points obtained from MRI at different flexion angles. The results show that the proposed methodology can be considered as a substitute to fully FE models without a substantial sacrifice on the outcomes despite having lower computational cost.","PeriodicalId":148883,"journal":{"name":"14th WCCM-ECCOMAS Congress","volume":"60 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128997205","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.23967/WCCM-ECCOMAS.2020.302
A. Mehraban, Jeremy L. Thompson, Jed Brown, R. Regueiro, Valeria, Barra, H. Tufo
. We examine a residual and matrix-free Jacobian formulation of compressible and nearly incompressible ( ν → 0 . 5) displacement-only linear isotropic elasticity with high-order hexahedral finite elements. A matrix-free p -multigrid method is combined with algebraic multigrid on the assembled sparse coarse grid matrix to provide an effective preconditioner. The software is verified with the method of manufactured solutions. We explore convergence to a predetermined L 2 error of 10 − 4 , 10 − 5 and 10 − 6 for the compressible case and 10 − 4 , 10 − 5 for the nearly-incompressible cases, as the Poisson’s ratio approaches 0.5, based upon grid resolution and polynomial order. We compare our results against results obtained from C3D20H mixed/hybrid element available in the commercial finite element software ABAQUS that is quadratic in displacement and linear in pressure. We determine, for the same problem size, that our matrix-free approach for displacement-only implementation is faster and more efficient
{"title":"Simulating Compressible and Nearly-Incompressible Linear Elasticity Using an Efficient Parallel Scalable Matrix-Free High-Order Finite Element Method","authors":"A. Mehraban, Jeremy L. Thompson, Jed Brown, R. Regueiro, Valeria, Barra, H. Tufo","doi":"10.23967/WCCM-ECCOMAS.2020.302","DOIUrl":"https://doi.org/10.23967/WCCM-ECCOMAS.2020.302","url":null,"abstract":". We examine a residual and matrix-free Jacobian formulation of compressible and nearly incompressible ( ν → 0 . 5) displacement-only linear isotropic elasticity with high-order hexahedral finite elements. A matrix-free p -multigrid method is combined with algebraic multigrid on the assembled sparse coarse grid matrix to provide an effective preconditioner. The software is verified with the method of manufactured solutions. We explore convergence to a predetermined L 2 error of 10 − 4 , 10 − 5 and 10 − 6 for the compressible case and 10 − 4 , 10 − 5 for the nearly-incompressible cases, as the Poisson’s ratio approaches 0.5, based upon grid resolution and polynomial order. We compare our results against results obtained from C3D20H mixed/hybrid element available in the commercial finite element software ABAQUS that is quadratic in displacement and linear in pressure. We determine, for the same problem size, that our matrix-free approach for displacement-only implementation is faster and more efficient","PeriodicalId":148883,"journal":{"name":"14th WCCM-ECCOMAS Congress","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123843704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1900-01-01DOI: 10.23967/WCCM-ECCOMAS.2020.070
N. Silva, B. Loiola, J. Costa, H. Orlande
. Approximate Bayesian Computation is used in this work for the selection and calibration of cell proliferation models. Four competing models based on ordinary differential equations are analyzed, by using the measurements of the proliferation of DU-145 prostate cancer viable cells during seven days. The selection criterion of the ABC algorithm is based on the Euclidean distance between the model prediction and the experimental observations. The Richards Model and the Generalized Logistic Model were selected by the ABC algorithm used in this work, providing accurate estimates of the evolution of the number of viable cells.
{"title":"Approximate Bayesian Computation Applied to Model Selection and Parameter Calibration in Cell Proliferation","authors":"N. Silva, B. Loiola, J. Costa, H. Orlande","doi":"10.23967/WCCM-ECCOMAS.2020.070","DOIUrl":"https://doi.org/10.23967/WCCM-ECCOMAS.2020.070","url":null,"abstract":". Approximate Bayesian Computation is used in this work for the selection and calibration of cell proliferation models. Four competing models based on ordinary differential equations are analyzed, by using the measurements of the proliferation of DU-145 prostate cancer viable cells during seven days. The selection criterion of the ABC algorithm is based on the Euclidean distance between the model prediction and the experimental observations. The Richards Model and the Generalized Logistic Model were selected by the ABC algorithm used in this work, providing accurate estimates of the evolution of the number of viable cells.","PeriodicalId":148883,"journal":{"name":"14th WCCM-ECCOMAS Congress","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127466087","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：