Griebel / Schüller / Schweitzer Scientific Computing and Algorithms in Industrial Simulations

Projects and Products of Fraunhofer SCAI
1. Auflage 2017
ISBN: 978-3-319-62458-7
Verlag: Springer International Publishing
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 371 Seiten

The contributions gathered here provide an overview of current research projects and selected software products of the Fraunhofer Institute for Algorithms and Scientific Computing SCAI. They show the wide range of challenges that scientific computing currently faces, the solutions it offers, and its important role in developing applications for industry. Given the exciting field of applied collaborative research and development it discusses, the book will appeal to scientists, practitioners, and students alike. The Fraunhofer Institute for Algorithms and Scientific Computing SCAI combines excellent research and application-oriented development to provide added value for our partners. SCAI develops numerical techniques, parallel algorithms and specialized software tools to support and optimize industrial simulations. Moreover, it implements custom software solutions for production and logistics, and offers calculations on high-performance computers. Its services and products are based on state-of-the-art methods from applied mathematics and information technology.

Michael Griebel is Director of the Fraunhofer Institute for Algorithms and Scientific Computing SCAI in Sankt Augustin, and holds a chair for Scientific Computing and Numerical Simulation at the Institute for Numerical Simulation at the University of Bonn.Marc Alexander Schweitzer is head of the department of Numerical Software of the Fraunhofer Institute for Algorithms and Scientific Computing SCAI in Sankt Augustin. Furthermore, he is currently Director of Institute for Numerical Simulation at the University of Bonn and holds a chair for Scientific Computing there.
Anton Schüller is working at the Fraunhofer Institute for Algorithms and Scientific Computing SCAI in Sankt Augustin.

Autoren/Hrsg.

Griebel, Michael

Schüller, Anton

Schweitzer, Marc Alexander

Inhaltsverzeichnis

1;Preface;5
2;Contents;7
3;Part I Methods;9
3.1;Calculation of Chemical Equilibria in Multi-Phase: Multicomponent Systems;10
3.1.1;1 Introduction;10
3.1.2;2 Problem Formulation;12
3.1.2.1;2.1 Non-Ideal Gibbs Function;12
3.1.2.2;2.2 Stoichiometric Constraints;13
3.1.2.3;2.3 The Optimization Problem;15
3.1.3;3 Methodology for the Calculation of Chemical Equilibria;16
3.1.3.1;3.1 Reformulation of the Minimization Problem;17
3.1.3.2;3.2 Discretization of the H-Problem;18
3.1.3.3;3.3 Corrector Step;19
3.1.4;4 Automated Detection of Miscibility Gaps;20
3.1.5;5 Results;23
3.1.5.1;5.1 Gibbs Free Energy Minimization Using BePhaSys;23
3.1.5.2;5.2 Calculation of Two-Dimensional Phase Diagrams: Interpolation and Parallelization;23
3.1.6;Appendix: The Gibbs Free Energy Function;29
3.1.7;References;30
3.2;LC-GAP: Localized Coulomb Descriptors for the Gaussian Approximation Potential;32
3.2.1;1 Introduction;32
3.2.2;2 Potential Energy Prediction Through Machine Learning;33
3.2.2.1;2.1 The GAP Framework and Gaussian Process Regression;34
3.2.2.2;2.2 Localized Coulomb Matrix Descriptors;35
3.2.3;3 Results;37
3.2.3.1;3.1 Comparison of Descriptor Functions on QM7;39
3.2.3.2;3.2 Larger Datasets and Prediction of Multiple Properties;41
3.2.3.3;3.3 Distribution of Individual Atomic Contributions;44
3.2.4;4 Conclusions and Future Work;47
3.2.5;References;48
3.3;River Bed Morphodynamics: Metamodeling, Reliability Analysis, and Visualization in a Virtual Environment;50
3.3.1;1 Introduction;50
3.3.2;2 RBF Metamodel;53
3.3.3;3 Quantile Estimation;55
3.3.3.1;3.1 Sensitivity-Based Approach;55
3.3.3.1.1;3.1.1 First-Order Approximation;55
3.3.3.1.2;3.1.2 Second-Order Approximation;55
3.3.3.2;3.2 Monte Carlo;56
3.3.3.3;3.3 Weighted Monte Carlo;57
3.3.3.4;3.4 Quasi-Monte Carlo (QMC);57
3.3.3.5;3.5 Quasi-Random Splines (QRS);58
3.3.4;4 Numerical Tests;59
3.3.5;5 Visualization in Virtual Environment;62
3.3.6;6 Conclusion;65
3.3.7;References;65
3.4;Cooling Circuit Simulation I: Modeling;67
3.4.1;1 Introduction;67
3.4.2;2 Network;68
3.4.3;3 Water Pipes;69
3.4.3.1;3.1 Continuum Mechanics;69
3.4.3.2;3.2 Simplifying Assumptions;70
3.4.3.3;3.3 Discretization and Regularization;72
3.4.4;4 Further Devices;75
3.4.4.1;4.1 Resistors and Valves;75
3.4.4.2;4.2 Pumps;77
3.4.4.3;4.3 Heat Exchangers;79
3.4.5;5 Element Control;82
3.4.6;6 Conclusion;84
3.4.7;References;85
4;Part II Products;86
4.1;Algebraic Multigrid: From Academia to Industry;87
4.1.1;1 Introduction;87
4.1.2;2 From Geometric to Algebraic Multigrid;89
4.1.3;3 The Early Phase of Algebraic Multigrid (1982–1987);91
4.1.3.1;3.1 The First Documented AMG Application;92
4.1.3.2;3.2 The Basics of `Classical' AMG;93
4.1.4;4 The Renaissance of AMG (1995–2000);95
4.1.4.1;4.1 Resumption of Major Research on AMG;95
4.1.4.2;4.2 Towards Industry;96
4.1.4.2.1;4.2.1 Computational Fluid Dynamics;97
4.1.4.2.2;4.2.2 Streamline Approach in Oil Reservoir Simulation;99
4.1.5;5 The Main AMG Development Phase (2000–Today);101
4.1.5.1;5.1 The General Trend;101
4.1.5.2;5.2 Bridging the Gap;102
4.1.5.3;5.3 SAMG for Coupled PDE Systems;105
4.1.5.3.1;5.3.1 Unknown-Based Approach;105
4.1.5.3.2;5.3.2 Point-Based Approach;106
4.1.5.3.3;5.3.3 Status of the Solver Framework SAMG;106
4.1.6;6 Industry-Driven Applications;107
4.1.6.1;6.1 Semiconductor Applications;107
4.1.6.2;6.2 Multi-Ion Transport and Reaction;110
4.1.6.3;6.3 Oil Reservoir Simulation;111
4.1.6.3.1;6.3.1 The Reservoir Simulation Models;113
4.1.6.3.2;6.3.2 Fully Implicit Methods;114
4.1.7;7 Summary, Conclusions and Lessons Learned;119
4.1.8;References;121
4.2;Parallel Algebraic Multigrid;124
4.2.1;1 Introduction;124
4.2.2;2 Challenges Imposed by Parallel Computer Architectures;126
4.2.2.1;2.1 Single Core Performance: CPU Clock Speed and Memory Frequency;126
4.2.2.2;2.2 Multi-Core CPUs and Shared Memory Parallelism;127
4.2.2.2.1;2.2.1 Multi-Core and Memory Access;128
4.2.2.2.2;2.2.2 Intrinsically Serial Components;129
4.2.2.2.3;2.2.3 Race Conditions;129
4.2.2.2.4;2.2.4 Multiple Sockets;130
4.2.2.3;2.3 Distributed Memory Parallelism;130
4.2.3;3 How SAMG Counters the HPC Challenges;131
4.2.3.1;3.1 Tuning and Parallelization of Smoothing;132
4.2.3.2;3.2 Ruge-Stüben Coarsening;134
4.2.3.3;3.3 Coarse Grid Solution;135
4.2.4;4 SCAI's Parallel SAMG Solver Library;136
4.2.5;References;137
4.3;MpCCI: Neutral Interfaces for Multiphysics Simulations;138
4.3.1;1 Introduction;138
4.3.2;2 MpCCI CouplingEnvironment;139
4.3.2.1;2.1 Aero-Elasticity and Fluid-Structure-Interaction;141
4.3.2.1.1;2.1.1 Wing and Spoiler Design;141
4.3.2.1.2;2.1.2 Hydraulic Pump Layout;142
4.3.2.2;2.2 Thermal and Vibration Loads in Turbomachinery;143
4.3.2.2.1;2.2.1 Thermal Loads on Ceramic Impeller;143
4.3.2.2.2;2.2.2 Life-Time Estimation of Turbine Blades;144
4.3.2.3;2.3 Vehicle Dynamics and Nonlinear Component Behavior;144
4.3.2.3.1;2.3.1 Driving Over Obstacles;144
4.3.2.3.2;2.3.2 Wading Simulation for Off-Road Vehicles;144
4.3.2.4;2.4 Automotive Thermal Management;146
4.3.2.4.1;2.4.1 Automotive Thermal Management for Full Vehicles;146
4.3.2.4.2;2.4.2 Automotive Thermal Management for Vehicle Manifolds;146
4.3.2.5;2.5 Component Design in Electrical Engineering;147
4.3.2.5.1;2.5.1 Cooling of a 3-Phase Transformer;147
4.3.2.5.2;2.5.2 Electric Arc in Switching Devices;147
4.3.3;3 MpCCI FSIMapper;148
4.3.4;4 MpCCI Mapper Solution for Integrated Simulation Workflows;149
4.3.4.1;4.1 Passive Safety;150
4.3.4.2;4.2 Forming Tools and Material Properties;151
4.3.4.2.1;4.2.1 Lightweight Stamping Tools: Use Forming Loads in Structural Optimization;151
4.3.4.2.2;4.2.2 Validation of Material Model Parameters: Compare Forming Results and Experimental Data;151
4.3.4.3;4.3 Composite Structures and Plastic Components;152
4.3.4.3.1;4.3.1 CFRP Workflows: From Draping via Mulling and Curing to Structural Analysis;152
4.3.4.3.2;4.3.2 Structural Integrity of Blow Moulded Plastic Components;153
4.3.5;5 Conclusion;153
4.3.6;References;153
4.4;Cooling Circuit Simulation II: A Numerical Example;155
4.4.1;1 Introduction;155
4.4.2;2 Application;156
4.4.2.1;2.1 Cooling System;156
4.4.2.2;2.2 Circuit Basics and Example;156
4.4.3;3 Concept and Software;163
4.4.3.1;3.1 Framework and Components;163
4.4.3.2;3.2 Semi-Automatic Model Creation with Schemparser;165
4.4.3.3;3.3 Device Modeling and Sensor Mapping;166
4.4.3.4;3.4 Collection of Measurement Data with PowerDAM;167
4.4.3.5;3.5 Nonlinear Problem Setup and Solution with MYNTS;168
4.4.4;4 Numerical Tests;170
4.4.4.1;4.1 Simplified Heat Exchanger;170
4.4.4.2;4.2 Logarithmic Mean Temperature Difference;177
4.4.5;5 Conclusion;179
4.4.6;References;181
4.5;The LAMA Approach for Writing Portable Applications on Heterogenous Architectures;183
4.5.1;1 Introduction;183
4.5.2;2 LAMA;184
4.5.2.1;2.1 Heterogeneous Memory;186
4.5.2.2;2.2 Heterogeneous Kernel;188
4.5.2.3;2.3 Task Parallelism;188
4.5.2.4;2.4 Distributed Memory Support;190
4.5.2.5;2.5 Matrices and Vectors;191
4.5.2.6;2.6 Solver Framework;191
4.5.2.7;2.7 Extensibility and Maintainability;193
4.5.3;3 Performance Comparison;194
4.5.4;4 Summary;197
4.5.5;Appendix;198
4.5.5.1;Test Environment;198
4.5.5.2;Test Matrices;199
4.5.6;References;200
4.6;ModelCompare;201
4.6.1;1 Introduction;201
4.6.2;2 Development History;202
4.6.3;3 Capabilities;202
4.6.3.1;3.1 Detection of Geometry Changes;203
4.6.3.2;3.2 Detection of MultiParts;204
4.6.3.3;3.3 Spotwelds and Rigid Body Elements;205
4.6.3.4;3.4 Detection of Material-ID and Thickness Changes;206
4.6.4;4 Outlook;207
4.6.5;References;207
4.7;Rapid Enriched Simulation Application Development with PUMA;208
4.7.1;1 Introduction;208
4.7.2;2 Partition of Unity Methods;209
4.7.3;3 PUMA Framework Design;210
4.7.4;4 Application Examples;212
4.7.5;5 Concluding Remarks;225
4.7.6;References;226
5;Part III Applications and Show Cases;228
5.1;Applying CFD for the Design of an Air-Liquid Interface In-Vitro Testing Method for Inhalable Compounds;229
5.1.1;1 Introduction;229
5.1.2;2 In-Vitro Air-Liquid Interface;230
5.1.3;3 Simulating the Aerosol Conduction System;231
5.1.4;4 Simulating the Liquid Supply System;237
5.1.4.1;4.1 Clogging in Liquid Channels;239
5.1.5;5 Simulating an Aerosol Sampling Box;240
5.1.6;6 Conclusions;243
5.1.7;References;243
5.2;A Mapping Procedure for the Computation of Flow-Induced Vibrations in Turbomachinery;244
5.2.1;1 Introduction;244
5.2.2;2 Nonlinear Harmonic Method;245
5.2.3;3 Mapping of Pressure Excitations;246
5.2.3.1;3.1 Periodic Models and Nodal Diameters;248
5.2.3.2;3.2 Deriving Excitation and Responding Shape;251
5.2.3.3;3.3 Summary;252
5.2.4;4 Application Example;252
5.2.4.1;4.1 Harmonic CFD Simulation;254
5.2.4.2;4.2 Mapping;254
5.2.4.3;4.3 Harmonic Structural Analysis;257
5.2.5;5 Conclusion;259
5.2.6;References;261
5.3;Molecular Dynamics Simulation of Membrane Free Energy Profiles Using Accurate Force Field for Ionic Liquids;263
5.3.1;1 Introduction;263
5.3.2;2 Computational Methods;264
5.3.2.1;2.1 Simulation Details;264
5.3.2.1.1;2.1.1 Technical Details;264
5.3.2.1.2;2.1.2 Force Field Development for [C2MIM][EtSO4];265
5.3.2.2;2.2 Umbrella Sampling;267
5.3.3;3 Results and Discussion;269
5.3.3.1;3.1 Force Field Development for [C2MIM][EtSO4];269
5.3.3.1.1;3.1.1 Density;269
5.3.3.1.2;3.1.2 Self-Diffusion Coefficients;270
5.3.3.1.3;3.1.3 Heat of Vaporization;270
5.3.3.1.4;3.1.4 Shear Viscosity;271
5.3.3.2;3.2 Free Energy Profiles;271
5.3.4;4 Outlook and Conclusion;274
5.3.4.1;4.1 Outlook: Towards Fully Automated Force Field Development;274
5.3.4.1.1;4.1.1 Case Study: Automated Parameterization of Ethylene Oxide;277
5.3.4.2;4.2 Conclusion;279
5.3.5;References;279
5.4;The cloud4health Project: Secondary Use of Clinical Data with Secure Cloud-Based Text Mining Services;283
5.4.1;1 Introduction;283
5.4.2;2 Developing a Secure Cloud-Solution for Medicine;285
5.4.2.1;2.1 Existing Cloud Solutions for Medicine;285
5.4.2.2;2.2 Requirements for Cloud Infrastructures Arising from Patient Data Processing;287
5.4.2.3;2.3 Security Mechanisms;288
5.4.2.4;2.4 Secure Cloud Infrastructure;290
5.4.2.4.1;2.4.1 Secure Clinical Gateway to the Cloud;291
5.4.2.4.2;2.4.2 Data Processing Flow;292
5.4.2.4.3;2.4.3 End-to-End Encryption;293
5.4.2.4.4;2.4.4 Multi-Tenancy and No Data Persistence;295
5.4.3;3 Clinical Text Mining Solutions;297
5.4.3.1;3.1 Short Literature Overview;297
5.4.3.2;3.2 General Architecture of the Text Mining Services;299
5.4.3.3;3.3 Overview Use Cases;301
5.4.3.3.1;3.3.1 General Use Case Process Model;302
5.4.3.4;3.4 Mining Endoprosthetic Surgery Reports;303
5.4.3.5;3.5 Mining Pathology Reports;306
5.4.4;4 Discussion;309
5.4.5;References;311
5.5;Dimensionality Reduction for the Analysis of Time Series Data from Wind Turbines;314
5.5.1;1 Introduction;314
5.5.2;2 Time Series Characteristics in Wind Energy;316
5.5.2.1;2.1 Numerical Simulations of Wind Turbines;316
5.5.2.2;2.2 Condition Monitoring of Wind Turbines;318
5.5.3;3 Exploration of Time Series Data from Numerical Simulations;319
5.5.3.1;3.1 Virtual Sensor Data from Wind Turbine Simulations;319
5.5.3.2;3.2 Nonlinear Dimensionality Reduction for Time Series Analysis;321
5.5.3.3;3.3 Diffusion Maps;322
5.5.3.4;3.4 Numerical Results;323
5.5.4;4 Anomaly Detection Based on Linear Dimensionality Reduction for Condition Monitoring Sensor Data from Wind Turbines;326
5.5.4.1;4.1 Sensor Data from Rotor Blades;326
5.5.4.1.1;4.1.1 Pre-processing;327
5.5.4.2;4.2 Anomaly Detection in Sensor Data;329
5.5.4.2.1;4.2.1 Model of Undamaged State;329
5.5.4.2.2;4.2.2 Deviation from the Undamaged State;330
5.5.4.3;4.3 Methodology for Damage Detection;331
5.5.4.4;4.4 Numerical Results;332
5.5.5;5 Conclusions;335
5.5.6;References;335
5.6;Energy-Efficiency and Performance Comparison of Aerosol Optical Depth Retrieval on Distributed Embedded SoC Architectures;337
5.6.1;1 Introduction;337
5.6.2;2 Method;338
5.6.3;3 Implementation;340
5.6.4;4 Embedded Low-Energy System;342
5.6.5;5 Benchmarks;344
5.6.5.1;5.1 Benchmark Environment;345
5.6.5.2;5.2 Performance Benchmarks;345
5.6.5.3;5.3 Energy Benchmarks;347
5.6.6;6 Discussion;353
5.6.7;7 Outlook;353
5.6.8;References;354
6;Part IV A Short History;355
6.1;The Fraunhofer Institute for Algorithms and ScientificComputing SCAI;356
6.1.1;1 Foundation of the GMD, the First Decade (1968–1977);358
6.1.2;2 Numerical Simulation, Multigrid and Parallel Computing (1978–1991);359
6.1.3;3 SCAI: Algorithms and Scientific Computing (1992–2001);362
6.1.4;4 SCAI as a Fraunhofer Institute, the First Years (2001–2009);366
6.1.5;5 New Fields of Research and New Business Areas (2010–2016);368
6.1.6;References;369