Baixuan Yang, A. Irastorza-Landa, P. Heuberger, H. Ploeg
Sufficient anchorage of dental implants, defined as mechanical engagement between implant and bone at the time of insertion, has been recommended for positive clinical outcomes, particularly in immediate loading protocols. Accurate measuring and analysis of stress and strain in the bone are imperative to understand anchorage from the biomechanics perspective. However, the stress and strain distributions at the bone-implant interface are impossible to measure in vivo. Therefore, the aim of this study was to develop and validate an explicit continuum finite element analysis (FEA) to investigate the stress and strain in a bone surrogate during the insertion of a dental implant: with cutting flute (CF) and without (NCF). Ten dental implants (five with CF and five with NCF) were inserted into ten rigid polyurethane (PU) foam blocks with a prepared pilot hole. During the insertion, a stereo digital image correlation (DIC) system was used to record in-plane deformation on the PU foam surface at a frequency of 1.0 Hz; and, surface von Mises strain, εv_DIC, was calculated (VIC-3D, Correlated Solutions Inc). In parallel, the insertion was simulated using FEA with explicit solver (Abaqus Explicit version 2017, Simulia). The PU foam was defined as an elastic-plastic material with a progressive crushable foam failure behaviour. The surface von Mises strain predicted from FEA, εv_FEA, was compared against εv_DIC. Uncertainty of DIC displacement measurement was 0.6 μm; and the static noise floor for the strain measurement was 500 microstrain (με). Coefficient of determination for εv_DIC and εv_FEA along a horizontal line for the CF and NCF implants were 0.80 and 0.78, respectively, which suggested the overall FEA performance was good. FEA results indicated that the cutting flute reduced the maximum shear stress in the PU foam and axial force which facilitated the insertion with less effort. This study demonstrated the successful combination of mechanical testing and FEA to better understand the mechanics of dental implant insertion.
牙种植体的充分锚固，定义为植入时种植体与骨之间的机械接合，已被推荐用于积极的临床结果，特别是在立即加载方案中。准确测量和分析骨内的应力和应变是从生物力学角度理解锚定的必要条件。然而，骨-种植体界面的应力和应变分布是不可能在体内测量的。因此，本研究的目的是开发并验证一种明确的连续体有限元分析(FEA)，以研究在牙种植体插入期间骨替代物的应力和应变:有切割凹槽(CF)和没有切割凹槽(NCF)。将10个种植体(5个CF和5个NCF)植入10个硬质聚氨酯(PU)泡沫块中，并准备好导孔。在插入过程中，使用立体数字图像相关(DIC)系统以1.0 Hz的频率记录PU泡沫表面的面内变形;计算表面von Mises应变εv_DIC (VIC-3D, related Solutions Inc .)。同时，使用显式求解器(Abaqus explicit version 2017, simula)对插入进行了有限元模拟。聚氨酯泡沫被定义为一种具有渐进可破碎泡沫破坏行为的弹塑性材料。用有限元法预测的表面von Mises应变εv_FEA与εv_DIC进行了比较。DIC位移测量的不确定度为0.6 μm;应变测量的静态噪声底限为500微应变(με)。CF和NCF植入体在水平方向上的εv_DIC和εv_FEA的决定系数分别为0.80和0.78，表明整体FEA性能良好。有限元分析结果表明，切割槽降低了PU泡沫中的最大剪切应力和轴向力，有利于以较小的力插入。本研究成功地将力学测试与有限元分析相结合，以更好地了解种植体插入的力学特性。
{"title":"Digital Image Correlation Validation of Finite Element Strain Analysis of Dental Implant Insertion for Two Implant Designs","authors":"Baixuan Yang, A. Irastorza-Landa, P. Heuberger, H. Ploeg","doi":"10.1115/vvuq2023-107659","DOIUrl":"https://doi.org/10.1115/vvuq2023-107659","url":null,"abstract":"\u0000 Sufficient anchorage of dental implants, defined as mechanical engagement between implant and bone at the time of insertion, has been recommended for positive clinical outcomes, particularly in immediate loading protocols. Accurate measuring and analysis of stress and strain in the bone are imperative to understand anchorage from the biomechanics perspective. However, the stress and strain distributions at the bone-implant interface are impossible to measure in vivo. Therefore, the aim of this study was to develop and validate an explicit continuum finite element analysis (FEA) to investigate the stress and strain in a bone surrogate during the insertion of a dental implant: with cutting flute (CF) and without (NCF). Ten dental implants (five with CF and five with NCF) were inserted into ten rigid polyurethane (PU) foam blocks with a prepared pilot hole. During the insertion, a stereo digital image correlation (DIC) system was used to record in-plane deformation on the PU foam surface at a frequency of 1.0 Hz; and, surface von Mises strain, εv_DIC, was calculated (VIC-3D, Correlated Solutions Inc). In parallel, the insertion was simulated using FEA with explicit solver (Abaqus Explicit version 2017, Simulia). The PU foam was defined as an elastic-plastic material with a progressive crushable foam failure behaviour. The surface von Mises strain predicted from FEA, εv_FEA, was compared against εv_DIC. Uncertainty of DIC displacement measurement was 0.6 μm; and the static noise floor for the strain measurement was 500 microstrain (με). Coefficient of determination for εv_DIC and εv_FEA along a horizontal line for the CF and NCF implants were 0.80 and 0.78, respectively, which suggested the overall FEA performance was good. FEA results indicated that the cutting flute reduced the maximum shear stress in the PU foam and axial force which facilitated the insertion with less effort. This study demonstrated the successful combination of mechanical testing and FEA to better understand the mechanics of dental implant insertion.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124360038","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}
The development of the systems analysis codes in use today was a very challenging task, stemming from the interplay of multiple physical phenomena, special components and control systems, and particularly the wide thermodynamic state envelope for a typical design basis accident scenario that includes single and two-phase behavior of the water working fluid within both the reactor vessel and the steam generator for the indirect cycle pressurized water reactor systems and also for the direct cycle boiling water reactor systems. The major developmental work leading to the current systems analysis codes was performed between the 1970s through the 1990s—and today these analysis tools are used throughout the world by organizations that design, submit their designs for licensing reviews, build, and operate light water reactor nuclear power plants. Differences in form of the discretized equations and closure relationships used within the systems analysis codes versus those in higher-fidelity computational fluid dynamics (CFD) codes lead to correspondingly different techniques to verify and validate (V&V) the equations in these two classes of codes. Systems analysis codes use a fundamental approach which has been developed over the years and which has been approved by the regulatory agencies whereas the CFD codes use high-fidelity V&V techniques as described in the ASME V&V standards for computational fluid mechanics and heat transfer codes. Because of the wide usage of high-fidelity CFD codes together with systems analysis codes, it is advisable to normalize the techniques for verifying, validating, and performing code adequacy assessments of these tools within the methodology that is presently available in the U.S. Nuclear Regulatory Commission’s Regulatory Guide 1.203. A strategy to begin closing the gap between the fundamental approach used to V&V systems analysis codes and the high-fidelity techniques used for modern CFD codes is outlined. It is postulated that the gap can be closed to the extent that some of the “high-fidelity” techniques may be used for systems analysis codes and thus enhance the quality of the code adequacy determination process for systems analysis codes.
{"title":"Using Code Adequacy Methodologies in Confomance with ASME Standards for Nuclear Power Plant Analysis Evaluation Models","authors":"R. Schultz, G. Mesina","doi":"10.1115/vvuq2023-108796","DOIUrl":"https://doi.org/10.1115/vvuq2023-108796","url":null,"abstract":"\u0000 The development of the systems analysis codes in use today was a very challenging task, stemming from the interplay of multiple physical phenomena, special components and control systems, and particularly the wide thermodynamic state envelope for a typical design basis accident scenario that includes single and two-phase behavior of the water working fluid within both the reactor vessel and the steam generator for the indirect cycle pressurized water reactor systems and also for the direct cycle boiling water reactor systems. The major developmental work leading to the current systems analysis codes was performed between the 1970s through the 1990s—and today these analysis tools are used throughout the world by organizations that design, submit their designs for licensing reviews, build, and operate light water reactor nuclear power plants.\u0000 Differences in form of the discretized equations and closure relationships used within the systems analysis codes versus those in higher-fidelity computational fluid dynamics (CFD) codes lead to correspondingly different techniques to verify and validate (V&V) the equations in these two classes of codes. Systems analysis codes use a fundamental approach which has been developed over the years and which has been approved by the regulatory agencies whereas the CFD codes use high-fidelity V&V techniques as described in the ASME V&V standards for computational fluid mechanics and heat transfer codes.\u0000 Because of the wide usage of high-fidelity CFD codes together with systems analysis codes, it is advisable to normalize the techniques for verifying, validating, and performing code adequacy assessments of these tools within the methodology that is presently available in the U.S. Nuclear Regulatory Commission’s Regulatory Guide 1.203.\u0000 A strategy to begin closing the gap between the fundamental approach used to V&V systems analysis codes and the high-fidelity techniques used for modern CFD codes is outlined. It is postulated that the gap can be closed to the extent that some of the “high-fidelity” techniques may be used for systems analysis codes and thus enhance the quality of the code adequacy determination process for systems analysis codes.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115883244","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}
Rotating bending fatigue (RBF) life is a critical performance index for oil drilling equipment. It can be measured via cantilever-type RBF testing, with a sideload applied on the specimen to generate the required bending moment at the point of interest. Multiple factors cause runout or eccentricity of the test specimen at the sideload point, leading to deviations in the effective sideload acting on the specimen and, consequently, deviations in the bending moment at the point of interest. Runout generates greater uncertainty in fatigue life test results. A Universal Runout Compensator (URC) was developed to reduce this uncertainty by mitigating the test specimen runout. It consists of two eccentric bushings, one assembled inside the other. Depending on the relative orientation of the two bushings, the total eccentricity of the URC can be adjusted. With the properly set URC installed on the specimen, the final runout at the URC where the sideload is applied becomes negligible. Finite element analysis (FEA) was used to confirm the URC structural integrity. Full-scale RBF tests were conducted to validate the URC design and FEA studies. With the URC used in the tests, bending moment variation decreased by up to 15%, reducing life cycle uncertainty by up to 44%.
{"title":"Uncertainty Reduction in Fatigue Life Validation Testing for Drilling Tools with a Universal Runout Compensator","authors":"M. Du, F. Song, Ke Li","doi":"10.1115/vvuq2023-107686","DOIUrl":"https://doi.org/10.1115/vvuq2023-107686","url":null,"abstract":"\u0000 Rotating bending fatigue (RBF) life is a critical performance index for oil drilling equipment. It can be measured via cantilever-type RBF testing, with a sideload applied on the specimen to generate the required bending moment at the point of interest. Multiple factors cause runout or eccentricity of the test specimen at the sideload point, leading to deviations in the effective sideload acting on the specimen and, consequently, deviations in the bending moment at the point of interest. Runout generates greater uncertainty in fatigue life test results.\u0000 A Universal Runout Compensator (URC) was developed to reduce this uncertainty by mitigating the test specimen runout. It consists of two eccentric bushings, one assembled inside the other. Depending on the relative orientation of the two bushings, the total eccentricity of the URC can be adjusted. With the properly set URC installed on the specimen, the final runout at the URC where the sideload is applied becomes negligible.\u0000 Finite element analysis (FEA) was used to confirm the URC structural integrity. Full-scale RBF tests were conducted to validate the URC design and FEA studies. With the URC used in the tests, bending moment variation decreased by up to 15%, reducing life cycle uncertainty by up to 44%.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127765735","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}
As analysis utilizing Finite Element Method (FEM) has become widely adopted in engineering practices and incorporated into governing standards, physical validation of these analyses is often forgone. Physical validation gives insight into the validity of assumptions and simplifications commonly used to efficiently process FEM simulations. This paper proposes that one reason physical validation is commonly forgone is a lack of knowledge of a general end to end methodology for the physical measurement, processing, and comparison of data. This paper presents such a methodology for the comparison of structural mechanical finite element analysis against strain gauge measurements utilizing the test case of a pressure vessel. Rectangular, three-axis, 45° strain gauge rosettes have been used to obtain normal strain inputs. The limitations and pitfalls of employing strain gauges with less than three measuring directions are briefly discussed. A procedure is provided for converting the three measured normal strains into three principal strains, von Mises equivalent strain and maximum shear strain. The principal directions, as well as an algorithm needed to resolve the ambiguity of the angle between the principal directions and gauge axes, are provided as well. Then, the strains are converted into principal stresses, von Mises equivalent stress and maximum shear stress. The post-processed strain gauge readings are visualized by employing 3D Mohr’s Circle for stress and strain. The visualization provides clear proof that the maximum shear lies on a plane different from the one on which the gauge has been attached. Using the described methodology, comparison shows that the difference between the FEA results and the post-processed strain gauge readings is less than 5%. The magnitudes of principal stresses and strains, the equivalent stress and strain, as well as the maximum shear stress and strain are compared. Besides the magnitudes of stresses and strains, the principal directions are compared and scrutinized, revealing the corroboration between the FEA and the physical measurements. This corroboration gives validity to both the methodology and assumptions, such as plane stress, used.
{"title":"Methodology for Validation of Finite Element Analysis Utilizing Strain Gauge Measurements","authors":"Rafal Sulwinski, Rusty Johnston","doi":"10.1115/vvuq2023-108749","DOIUrl":"https://doi.org/10.1115/vvuq2023-108749","url":null,"abstract":"\u0000 As analysis utilizing Finite Element Method (FEM) has become widely adopted in engineering practices and incorporated into governing standards, physical validation of these analyses is often forgone. Physical validation gives insight into the validity of assumptions and simplifications commonly used to efficiently process FEM simulations. This paper proposes that one reason physical validation is commonly forgone is a lack of knowledge of a general end to end methodology for the physical measurement, processing, and comparison of data. This paper presents such a methodology for the comparison of structural mechanical finite element analysis against strain gauge measurements utilizing the test case of a pressure vessel.\u0000 Rectangular, three-axis, 45° strain gauge rosettes have been used to obtain normal strain inputs. The limitations and pitfalls of employing strain gauges with less than three measuring directions are briefly discussed.\u0000 A procedure is provided for converting the three measured normal strains into three principal strains, von Mises equivalent strain and maximum shear strain. The principal directions, as well as an algorithm needed to resolve the ambiguity of the angle between the principal directions and gauge axes, are provided as well.\u0000 Then, the strains are converted into principal stresses, von Mises equivalent stress and maximum shear stress. The post-processed strain gauge readings are visualized by employing 3D Mohr’s Circle for stress and strain. The visualization provides clear proof that the maximum shear lies on a plane different from the one on which the gauge has been attached.\u0000 Using the described methodology, comparison shows that the difference between the FEA results and the post-processed strain gauge readings is less than 5%. The magnitudes of principal stresses and strains, the equivalent stress and strain, as well as the maximum shear stress and strain are compared.\u0000 Besides the magnitudes of stresses and strains, the principal directions are compared and scrutinized, revealing the corroboration between the FEA and the physical measurements. This corroboration gives validity to both the methodology and assumptions, such as plane stress, used.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128101873","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}
In engineering simulations involving turbulent fluid flows, the Reynolds-averaged Navier-Stokes (RANS) equations are still the most common mathematical model. The RANS equations require the use of a turbulence model to calculate the Reynolds stresses generated by the averaging of the momentum equations. Nowadays, the most popular turbulence models require the solution of additional transport equations that can range from one to seven equations. In this paper we illustrate the difficulties in attaining and identifying the so-called asymptotic range in grid refinement studies performed for the numerical solution of the RANS equations in the flow over a flat plate. Three turbulence models are tested: two-equation, eddy-viscosity, k—ω SST and k-kL turbulence models and the seven-equation Reynolds stress model SSG/LRR—ω. The three turbulence models are tested with second and first-order upwind schemes applied to the convective terms of the turbulence models transport equations. The results show that even in this simple flow, attaining the asymptotic order of grid convergence requires unreasonable levels of grid refinement. Furthermore, even in strictly geometrical similar grids, the observed order of grid refinement can be extremely sensitive to the discretization schemes used in the turbulence model and to any disturbances in the data. An alternative and more efficient way to address the quality of an error estimation based on a single term expansion is to determine the change of the estimate of the exact solution with grid refinement.
{"title":"Attaining the Asymptotic Range in Rans Simulations","authors":"L. Eça, M. Kerkvliet, S. Toxopeus","doi":"10.1115/vvuq2023-108745","DOIUrl":"https://doi.org/10.1115/vvuq2023-108745","url":null,"abstract":"\u0000 In engineering simulations involving turbulent fluid flows, the Reynolds-averaged Navier-Stokes (RANS) equations are still the most common mathematical model. The RANS equations require the use of a turbulence model to calculate the Reynolds stresses generated by the averaging of the momentum equations. Nowadays, the most popular turbulence models require the solution of additional transport equations that can range from one to seven equations.\u0000 In this paper we illustrate the difficulties in attaining and identifying the so-called asymptotic range in grid refinement studies performed for the numerical solution of the RANS equations in the flow over a flat plate. Three turbulence models are tested: two-equation, eddy-viscosity, k—ω SST and k-kL turbulence models and the seven-equation Reynolds stress model SSG/LRR—ω. The three turbulence models are tested with second and first-order upwind schemes applied to the convective terms of the turbulence models transport equations.\u0000 The results show that even in this simple flow, attaining the asymptotic order of grid convergence requires unreasonable levels of grid refinement. Furthermore, even in strictly geometrical similar grids, the observed order of grid refinement can be extremely sensitive to the discretization schemes used in the turbulence model and to any disturbances in the data. An alternative and more efficient way to address the quality of an error estimation based on a single term expansion is to determine the change of the estimate of the exact solution with grid refinement.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117029110","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}
The front matter for this proceedings is available by clicking on the PDF icon.
通过点击PDF图标可获得本次会议的主题。
{"title":"VVUQ2023 Front Matter","authors":"","doi":"10.1115/vvuq2023-fm1","DOIUrl":"https://doi.org/10.1115/vvuq2023-fm1","url":null,"abstract":"\u0000 The front matter for this proceedings is available by clicking on the PDF icon.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128236800","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}
VVUQ techniques were developed to address simulation credibility when a device or manufacturing process is analyzed through computational means. If the device or manufacturing process analyzed fails, the consequences could be potentially hazardous, if not catastrophic, to the health and welfare of the public. VVUQ procedures provide a needed framework to guide the engineer in the design of a device, system, or process along with the supporting calculations so that there is transparency in presenting the simulation results. While major research facilities, medical device developers, and cutting-edge technology companies have led the development of specific VVUQ techniques, these principles are appropriate for simulation-informed decision making regardless of the industry, but the degree of detail is driven by the risk and uncertainty. This paper will present how VVUQ applies to established industries, using the ASME pressure vessel standards as an example. In traditional “by rules” pressure vessel design, uncertainty has been reduced by decades of testing and development. Similarly, explicit use of VVUQ is not typically needed for numerical modeling such as Finite Element Modeling (FEM) in “design by analysis” when applied within the confines of ASME pressure vessel standards because the testing and development to develop those engineering standards reduced the uncertainty. By showing where VVUQ principles have been implicitly applied, the paper will then show why explicit VVUQ requirements and constraints are required for the standard under development, “Design By Analysis for Glassy Polymers,” which does not have the benefit of pre-qualified material data or simplifying assumptions such as thin wall pressure theory. Identifying the implicit VVUQ methods in current pressure vessel standards will help ensure that the simulation and experimentation used for glassy polymers will meet or exceed the reliability established by those standards. These VVUQ methods will also provide guidance for novel applications of pressure vessel technology and other structural applications outside the scope of established engineering codes.
{"title":"Application of VVUQ Concepts to ASME Codes and Standards for Pressure Vessels","authors":"Bart Kemper","doi":"10.1115/vvuq2023-108506","DOIUrl":"https://doi.org/10.1115/vvuq2023-108506","url":null,"abstract":"\u0000 VVUQ techniques were developed to address simulation credibility when a device or manufacturing process is analyzed through computational means. If the device or manufacturing process analyzed fails, the consequences could be potentially hazardous, if not catastrophic, to the health and welfare of the public. VVUQ procedures provide a needed framework to guide the engineer in the design of a device, system, or process along with the supporting calculations so that there is transparency in presenting the simulation results. While major research facilities, medical device developers, and cutting-edge technology companies have led the development of specific VVUQ techniques, these principles are appropriate for simulation-informed decision making regardless of the industry, but the degree of detail is driven by the risk and uncertainty. This paper will present how VVUQ applies to established industries, using the ASME pressure vessel standards as an example. In traditional “by rules” pressure vessel design, uncertainty has been reduced by decades of testing and development. Similarly, explicit use of VVUQ is not typically needed for numerical modeling such as Finite Element Modeling (FEM) in “design by analysis” when applied within the confines of ASME pressure vessel standards because the testing and development to develop those engineering standards reduced the uncertainty. By showing where VVUQ principles have been implicitly applied, the paper will then show why explicit VVUQ requirements and constraints are required for the standard under development, “Design By Analysis for Glassy Polymers,” which does not have the benefit of pre-qualified material data or simplifying assumptions such as thin wall pressure theory. Identifying the implicit VVUQ methods in current pressure vessel standards will help ensure that the simulation and experimentation used for glassy polymers will meet or exceed the reliability established by those standards. These VVUQ methods will also provide guidance for novel applications of pressure vessel technology and other structural applications outside the scope of established engineering codes.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128939646","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}
The strength of materials at high strain rates is a challenging problem for model development and calibration. Such models can span a regime in strain rate from 1 × 10−3 s−1 to 1 × 1012 s−1 and a regime in temperature of 0K to up to the material’s melting temperature. The limits of these regimes can be difficult and expensive to access experimentally. There is interest in understanding how well calibrations made at moderate strain rates and temperature can perform when applied to more extreme regimes. Variational Bayesian techniques have been shown to be computationally inexpensive methods to both calibrate a model and understand the uncertainties in model parameters. This investigation will calibrate the parameters of a Peston-Tonks-Wallace (PTW) material strength model to low and moderate strain rate experiments from quasi-static, and Hopkinson bar experiments performed on fully annealed Oxygen Free High Conductivity (OFHC) copper. Bayesian methods will be used to quantify the correlated uncertainty in these parameters. These uncertainties will propagated forward to a simulation of a Richtmyer-Meshkov instability experiment which exercise a higher strain rate regime. The effects of the model uncertainties on the predictive ability of the simulation will be observed. This will demonstrate a strategy for Bayesian model calibration and uncertainty quantification for parametric models with applications to physics processes outside high strain rate plastic deformation.
{"title":"Variational Bayesian Calibration of a PTW Material Strength Model for OFHC Copper","authors":"Stephen A. Andrews, B. Wilson","doi":"10.1115/vvuq2023-108829","DOIUrl":"https://doi.org/10.1115/vvuq2023-108829","url":null,"abstract":"\u0000 The strength of materials at high strain rates is a challenging problem for model development and calibration. Such models can span a regime in strain rate from 1 × 10−3 s−1 to 1 × 1012 s−1 and a regime in temperature of 0K to up to the material’s melting temperature. The limits of these regimes can be difficult and expensive to access experimentally. There is interest in understanding how well calibrations made at moderate strain rates and temperature can perform when applied to more extreme regimes. Variational Bayesian techniques have been shown to be computationally inexpensive methods to both calibrate a model and understand the uncertainties in model parameters. This investigation will calibrate the parameters of a Peston-Tonks-Wallace (PTW) material strength model to low and moderate strain rate experiments from quasi-static, and Hopkinson bar experiments performed on fully annealed Oxygen Free High Conductivity (OFHC) copper. Bayesian methods will be used to quantify the correlated uncertainty in these parameters. These uncertainties will propagated forward to a simulation of a Richtmyer-Meshkov instability experiment which exercise a higher strain rate regime. The effects of the model uncertainties on the predictive ability of the simulation will be observed. This will demonstrate a strategy for Bayesian model calibration and uncertainty quantification for parametric models with applications to physics processes outside high strain rate plastic deformation.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115811838","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}
We present an application of convolutional neural networks for calibration of a tensile plasticity (TePla) damage model simulating the spallation in copper under high-explosive shock loading. Using a high-fidelity, multi-physics simulation developed by the Advanced Simulation and Computing program at Los Alamos National Laboratory (LANL), we simulate hundreds of variations of a high-explosive shock experiment involving a copper coupon. From this synthetic data, we train neural networks to learn the inverse mapping between the coupon’s late-time density field, or an associated synthetic radiograph, and the simulation’s TePla damage parameters. It is demonstrated that, using a simple convolutional architecture, we can train networks to infer damage parameters from density fields accurately. Neural network inference directly from synthetic radiographs is significantly more challenging. Application of machine-learning methods must be accompanied by an analysis of how they are making inferences in order to build confidence in predictions and to identify likely shortcomings of the technique. To understand what the model is learning, individual layer outputs are extracted and examined. Each layer in the network identifies multiple features. However, each of these features are not necessarily of equal importance in the network’s final prediction of a given damage parameter. By examining the features overlaid on the input hydrodynamic fields, we assess the question of whether or not the model’s accuracy can be attributed to human-recognizable characteristics. In this work we give a detailed description of our data-generation methods and the learning problem we address. We then outline our neural architecture trained for damage calibration and discuss considerations made during training and evaluation of accuracy. Methods for human interpretation of the network’s inference process are then put forward, including extraction of learned features from the trained network and techniques to assess sensitivity of inferences to the learned features.
{"title":"Training and Interpretability of Deep-Neural Methods for Damage Calibration in Copper.","authors":"K. Hickmann, Skylar Callis, Stephen Andrews","doi":"10.1115/vvuq2023-108759","DOIUrl":"https://doi.org/10.1115/vvuq2023-108759","url":null,"abstract":"\u0000 We present an application of convolutional neural networks for calibration of a tensile plasticity (TePla) damage model simulating the spallation in copper under high-explosive shock loading. Using a high-fidelity, multi-physics simulation developed by the Advanced Simulation and Computing program at Los Alamos National Laboratory (LANL), we simulate hundreds of variations of a high-explosive shock experiment involving a copper coupon. From this synthetic data, we train neural networks to learn the inverse mapping between the coupon’s late-time density field, or an associated synthetic radiograph, and the simulation’s TePla damage parameters. It is demonstrated that, using a simple convolutional architecture, we can train networks to infer damage parameters from density fields accurately. Neural network inference directly from synthetic radiographs is significantly more challenging. Application of machine-learning methods must be accompanied by an analysis of how they are making inferences in order to build confidence in predictions and to identify likely shortcomings of the technique. To understand what the model is learning, individual layer outputs are extracted and examined. Each layer in the network identifies multiple features. However, each of these features are not necessarily of equal importance in the network’s final prediction of a given damage parameter. By examining the features overlaid on the input hydrodynamic fields, we assess the question of whether or not the model’s accuracy can be attributed to human-recognizable characteristics. In this work we give a detailed description of our data-generation methods and the learning problem we address. We then outline our neural architecture trained for damage calibration and discuss considerations made during training and evaluation of accuracy. Methods for human interpretation of the network’s inference process are then put forward, including extraction of learned features from the trained network and techniques to assess sensitivity of inferences to the learned features.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128684328","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}
Impact limiters are often used to protect equipment by minimizing the load inflicted to the equipment due to an impact or fall. This paper presents a computer simulation of a simple and effective impact limiter used in a storage cask carrying a Multi-Purpose Canister (MPC) containing nuclear spent fuel assemblies and compares the analysis results with the actual drop test performed for the prototype of the impact limiter. The impact limiter consists of an array of stainless-steel tubes with small holes in each to define and accelerate the collapse of the tubes following an impact. The small holes drilled at the strategically picked location on the tube ensures a very uniform tube collapse pattern and thus a well-controlled overall impact limiter behavior. The numerical simulation is conducted with computer modeling in LS-DYNA with appropriate geometric parameters and material properties. The behavior of the impact limiter tubes is captured by the true stress true strain curve of the material. The numerical analysis reveals how the tubes collapse due to an impact from a drop accident and what the collapse pattern looks like. The prototype test is conducted to verify the accuracy of the computer model, and the collapse of the tubes is observed and recorded using a high-speed camera. Both the measured impact limiter deformation and impact acceleration match well with the predictions by the computer model. This simple impact limiter device is extremely effective in absorbing energy and the required design objective can be reliably confirmed by computer simulation.
{"title":"Impact Limiter Computer Simulation and Verification by Drop Tests","authors":"K. K. Niyogi, X. Zhai","doi":"10.1115/vvuq2023-108557","DOIUrl":"https://doi.org/10.1115/vvuq2023-108557","url":null,"abstract":"\u0000 Impact limiters are often used to protect equipment by minimizing the load inflicted to the equipment due to an impact or fall. This paper presents a computer simulation of a simple and effective impact limiter used in a storage cask carrying a Multi-Purpose Canister (MPC) containing nuclear spent fuel assemblies and compares the analysis results with the actual drop test performed for the prototype of the impact limiter. The impact limiter consists of an array of stainless-steel tubes with small holes in each to define and accelerate the collapse of the tubes following an impact. The small holes drilled at the strategically picked location on the tube ensures a very uniform tube collapse pattern and thus a well-controlled overall impact limiter behavior. The numerical simulation is conducted with computer modeling in LS-DYNA with appropriate geometric parameters and material properties. The behavior of the impact limiter tubes is captured by the true stress true strain curve of the material. The numerical analysis reveals how the tubes collapse due to an impact from a drop accident and what the collapse pattern looks like. The prototype test is conducted to verify the accuracy of the computer model, and the collapse of the tubes is observed and recorded using a high-speed camera. Both the measured impact limiter deformation and impact acceleration match well with the predictions by the computer model. This simple impact limiter device is extremely effective in absorbing energy and the required design objective can be reliably confirmed by computer simulation.","PeriodicalId":387733,"journal":{"name":"ASME 2023 Verification, Validation, and Uncertainty Quantification Symposium","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2023-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129549892","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：