ALTAIR Info-Tag Strukturmechanik 2015-3-24

24/03/2015Info-Tag Strukturmechanik Christian Alscher Innovation Intelligence® Kristian Holm Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Agenda 9:00 Registrierung und Kaffee 9:30 Begrüßung Überblick: Aufgabenstellungen in der Strukturmechanik Kaffeepause Thermisch-mechanische Analyse am Beispiel eines Motorblocks Materialmodellierung für unterschiedliche Werkstoffe Strukturberechnung mit MKS-Lastbestimmung 13:00 Mittagessen Composites: Modellaufbau, Berechnung und Auswertung Workflow für Schwingungs- und Akustikanalysen Möglichkeiten zur Strukturoptimierung Diskussion 1 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Why Composites? • High stiffness to weight ratio! • Tailored properties! • • • Fibers can be aligned to provide directional stiffness Thermal expansion control Vibrational damping • Desirable Material Properties! • Design and manufacturing flexibility: • Concrete Carbon Fatigue and corrosion resistance fewer parts! Design Challenge: • High Number of Variables • Orthotropic Material Properties • Orientation of the Ply Angles • Stacking Sequence Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Modeling Techniques for Composites • Each layer with at least one solid • Large model, long CPU time • High accuracy solid • Mixed approach (middle layer thick) • Shells for top and bottom layer • Solid, thick shell or coehesive for middle layer • Several solid layers to provide rotation solid shell • Sandwich shell approach • One shell element through the thickness • Multiple layers, with different materials shell 2 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Modeling Intro: Zone Based • One property required for each ply drop/add (i.e. Zone) • Properties are not linked across zones • No direction relationship to manufacturing process • Example: Data duplication, No Direct Ply Shape Zone 1 Plate requires three property tables Zone 2 with repeating ply definitions Delete P4 requires three updates Zone 1 – Property Table Zone 3 – Property Table Ply Mat Thk Theta Ply Mat Thk Theta P7 M1 0.01 45 P7 M1 0.01 45 P4 M1 0.01 0 P6 M1 0.01 90 P1 M1 0.01 45 P5 M1 0.01 -45 P4 M1 0.01 0 P3 M1 0.01 -45 P2 M1 0.01 90 P1 M1 0.01 45 Zone 2 – Property Table Ply Mat Thk Theta P7 M1 0.01 45 P5 M1 0.01 -45 P4 M1 0.01 0 P3 M1 0.01 -45 P1 M1 0.01 45 Zone 3 P1 P2 P3 P4 P5 P6 P7 Zone 2 Zone 1 45 90 -45 0 -45 90 45 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Ply Based Composite Modeling • One stack (property) required per part • A stack defines zones via ply shape definitions • Direction relationship to manufacturing process • Example: Plate requires seven ply definitions, no repeating plies, each ply shape defined Stacking plies automatically defines zones Delete P4 requires only one update Zone 1 No Data Duplication Zone 2 Zone 3 P1 P2 P3 P4 P5 P6 P7 Zone 2 Zone 1 45 90 -45 0 -45 90 45 Stack Table Ply Mat Thk Theta P7 M1 0.01 45 P6 M1 0.01 90 P5 M1 0.01 -45 P4 M1 0.01 0 P3 M1 0.01 -45 P2 M1 0.01 90 P1 M1 0.01 45 3 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Ply Based Composite Modeling Integrated into HyperMesh via Model Browser Dialogs Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Visualization • Objective: Visually verify engineering data associated with math model • “Live” 3D representation of 1D/2D elements w/ or w/o layers • Color “by thickness” • Element normal • Material direction • Ply orientations • Ply expansion 4 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Ply-Based Post-Processing • Individual Ply Results • Strain Tensor Components • Stress Tensor Components • Principal Strain/Stress • Failure Theories (Maximum Strain, Hoffman, Tsai-Hill, TsaiWu) • Envelope Ply Results • Min/Max/Extreme • Sum/Average/Range (12.0) • Identify Min/Max/Extreme Layer Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. HyperWorks Solvers OptiStruct Statics NVH Thermal RADIOSS MotionSolve AcuSolve FEKO Multi-body Dynamics Thermal and CFD ElectroMagnetics Highly Nonlinear Crash Safety Forming Nonlinear Optimization Smart Multiphysics 5 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. OptiStruct: Composite Optimization Phase 1: Concept Free-Size Phase 2: Dimension Size • Objective: minimize mass (failure index, …) • Constraint: displacement (stress, …) • Manufacturing constraints: Phase 3: Sequence Shuffle • • • • • Min. and max. total laminate thickness Min. and max. ply thickness Min. and max. percentage of a fibre orientation Linkage of thicknesses of plies Constant thickness for a particular ply orientation Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. OptiStruct: Composite Optimization Phase 1: Concept Free-Size Phase 2: Dimension Size Phase 3: Sequence Shuffle Phase 1: Concept Q: Which fibre angles do I need where? A: Free-Size-Optimization creates a thickness distribution for each fibre angle. Phase 2: Dimension Q: How many plies of each shape do I need? A: Discrete Size-Optimization calculates optimal number of each ply shape of each fibre angle. Phase 3: Sequence Q: I which order do I have to stack the plies? A: Shuffle-Optimization finds optimal stacking order under consideration of ply book rules. 6 24/03/2015 Copyright © 2013 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. OptiStruct: Composite Optimization Composite Free-Size Optimization What are the most efficient ply shapes? Initial Design Ply Slicing Determines ply shapes Final Design Composite Shuffling What is the exact stacking sequence to meet manufacturing requirements? Composite Size Optimization How many plies of each ply shape to meet engineering targets? Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization of a Composite Sports Car Wing Finite element model representation Optimal shapes/patches: free size optimization Optimal number and thickness of plies: sizing optimization Peak failure index reduction: 20% Displacement constraint satisfied Minimum reserve factor met Ply stacking sequence optimization 7 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. HyperWorks Solvers OptiStruct Statics NVH Thermal RADIOSS MotionSolve AcuSolve FEKO Multi-body Dynamics Thermal and CFD ElectroMagnetics Highly Nonlinear Crash Safety Forming Nonlinear Optimization Smart Multiphysics Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. RADIOSS - DALLARA Race cars design 8 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. RADIOSS - DALLARA Race cars design Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Helicopter Survivability • • • Model: Helicopter floor structure Objective: Improve survivability of cabin crew under crash landing Loading: Crash landing Vi= 3-10 m/s 9 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Helicopter Survivability • • • Model: Helicopter floor structure Objective: Improve survivability of cabin crew under crash landing Loading: Crash landing Vi= 3-10 m/s Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Partner Caterham Composites Lotus T127 Formel 1 Simulation dynamischer Seitencrash 10 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Gegenstand der Untersuchung Paul Schischkin Seitencrash-Absorber aus Kohlenstofffaser-Kunststoff-Verbund Laminatmaterial • 2x2 Twill Gewebe Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Gegenstand der Untersuchung Paul Schischkin Realversuch Seitencrash: Vorher-Nachher-Vergleich 11 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Modellaufbau Paul Schischkin Solver: RADIOSS V12.0.210 Vernetzung der Geometrie: • Grundlage: CAD-Model • 5x5mm 4-Knoten-Schalenelemente (quads) • eine Schale pro Laminatdicke • Anzahl: insgesamt ca. 7500 Elemente Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Modellaufbau Paul Schischkin Elementeigenschaften (properties): • Definition des Laminataufbaus durch den Sandwich Schalen Ansatz • Jede property erhält Informationen zu Anzahl der Lagen, sowie ihre jeweilige Position, Dicke, Orientierung und Material 12 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Modellaufbau Paul Schischkin Materialmodell: Elasto-plastisch-orthotrop (exemplarisch) Druckrichtung Druckrichtung Zugrichtung Zugrichtung Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Radioss Composite Material Model #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| /MAT/COMPSH/10 Altair composite material # Init. dens. Ref. dens. 1.5000E-6 0 # E11 E22 NU12 IFLAG E33 42 40 .05 1 0.50 # G12 G23 G31 EPSF1 EPSF2 3.4 3 3 0 0 # ESPT1 EPSM1 EPST2 EPSM2 Dmax 0 0 0 0 .9999 # Wpmax Wpref Ioff 0 0 5 # C EPS ALPHA Icc 0 0 0 0 # sig_trac_1 B_1T N_1T SIGMA_1MAXT C_1T 0.1 25 .10 0 0 # EPS_1T1 EPS_2T1 SIGMA_RST1 Wpmax_trac_1 0 0 0 0 # sig_trac_2 B_2T N_2T SIGMA_2MAXT C_2T 0.1 20 .10 0 0 # EPS_1T2 EPS_2T2 SIGMA_RST2 Wpmax_trac_2 0 0 0 0 # sig_comp_1 B_1C N_1C SIGMA_1MAXC C_1C .0050 2000 .5 0 0 # EPS_1C1 EPS_2C1 SIGMA_RSC1 Wpmax_comp_1 0 0 0 0 # sig_comp_2 B_2C N_2C SIGMA_2MAXC C_2C .0050 2000 .5 0 0 # EPS_1C2 EPS_2C2 SIGMA_RSC2 Wpmax_comp_2 0 0 0 0 # sig_12 B_12T N_12T SIGMA_12MAXT C_12T .0040 83.0 .31 0 0 # EPS_1T12 EPS_2T12 SIGMA_RST12 Wpmax_trac_12 0.075 0.085 0.05 0 # GAMMA_INI GAMMA_MAX Dmax 1E31 1E31 .9999 # Fsmooth Fcut 0 0 #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| Linear Nonlinear 13 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Material Model: Yield Stress #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| /MAT/COMPSH/10 Altair composite material σ2 # Init. dens. Ref. dens. t σ 12 1.5000E-6 0 σ 2t E11 E22 NU12 IFLAG E33 σ1 42 40 .05 1 0.50 # G12 G23 G31 EPSF1 EPSF2 σ 1t σ 1c 3.4 3 3 0 0 c F (σ ) = 1 # ESPT1 EPSM1 EPST2 EPSM2 Dmax σ 12 σ 2c 0 0 0 0 .9999 # Wpmax Wpref Ioff 0 0 5 # C EPS ALPHA Icc 0 0 0 0 # sig_trac_1 B_1T N_1T SIGMA_1MAXT C_1T 1e+30 0 0 0 0 # EPS_1T1 EPS_2T1 SIGMA_RST1 Wpmax_trac_1 0 0 0 0 # sig_trac_2 B_2T N_2T SIGMA_2MAXT C_2T 1e+30 0 0 0 0 # EPS_1T2 EPS_2T2 SIGMA_RST2 Wpmax_trac_2 0 0 0 0 # sig_comp_1 B_1C N_1C SIGMA_1MAXC C_1C 1e+30 0 0 0 0 # EPS_1C1 EPS_2C1 SIGMA_RSC1 Wpmax_comp_1 0 0 0 0 # sig_comp_2 B_2C N_2C SIGMA_2MAXC C_2C 1e+30 0 0 0 0 # EPS_1C2 EPS_2C2 SIGMA_RSC2 Wpmax_comp_2 0 0 0 0 # sig_12 B_12T N_12T SIGMA_12MAXT C_12T .0040 67.0 .29 0 0 # EPS_1T12 EPS_2T12 SIGMA_RST12 Wpmax_trac_12 0.00 0.00 0.00 0 # GAMMA_INI GAMMA_MAX Dmax 1E31 1E31 .9999 # Fsmooth Fcut 0 0 #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| # Yield Stress Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Material Model: Plasticity #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| /MAT/COMPSH/10 Altair composite material # Init. dens. Ref. dens. 1.5000E-6 0 # E11 E22 NU12 IFLAG E33 42 40 .05 1 0.50 # G12 G23 G31 EPSF1 EPSF2 3.4 3 3 0 0 # ESPT1 EPSM1 EPST2 EPSM2 Dmax 0 0 0 0 .9999 # Wpmax Wpref Ioff 0 0 5 # C EPS ALPHA Icc 0 0 0 0 # sig_trac_1 B_1T N_1T SIGMA_1MAXT C_1T 1e+30 0 0 0 0 # EPS_1T1 EPS_2T1 SIGMA_RST1 Wpmax_trac_1 0 0 0 0 # sig_trac_2 B_2T N_2T SIGMA_2MAXT C_2T 1e+30 0 0 0 0 # EPS_1T2 EPS_2T2 SIGMA_RST2 Wpmax_trac_2 0 0 0 0 # sig_comp_1 B_1C N_1C SIGMA_1MAXC C_1C 1e+30 0 0 0 0 # EPS_1C1 EPS_2C1 SIGMA_RSC1 Wpmax_comp_1 0 0 0 0 # sig_comp_2 B_2C N_2C SIGMA_2MAXC C_2C 1e+30 0 0 0 0 # EPS_1C2 EPS_2C2 SIGMA_RSC2 Wpmax_comp_2 0 0 0 0 # sig_12 B_12T N_12T SIGMA_12MAXT C_12T .0040 67.0 .29 0 0 # EPS_1T12 EPS_2T12 SIGMA_RST12 Wpmax_trac_12 0.00 0.00 0.00 0 # GAMMA_INI GAMMA_MAX Dmax 1E31 1E31 .9999 # Fsmooth Fcut 0 0 #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| Plasticity, Strain Rates 14 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Material Model: Damage and Failure #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| /MAT/COMPSH/10 Altair composite material # Init. dens. Ref. dens. 1.5000E-6 0 # E11 E22 NU12 IFLAG E33 42 40 .05 1 0.50 # G12 G23 G31 EPSF1 plastic work EPSF2 3.4 3 3 0 0 Wp # ESPT1 EPSM1 EPST2 EPSM2 Dmax 0 0 0 0 .9999 # Wpmax Wpref Ioff 0 0 5 # C EPS ALPHA Icc 0 0 0 0 # sig_trac_1 B_1T N_1T SIGMA_1MAXT C_1T 1e+30 0 0 0 0 # EPS_1T1 EPS_2T1 SIGMA_RST1 Wpmax_trac_1 0 0 0 0 # sig_trac_2 B_2T N_2T SIGMA_2MAXT C_2T 1e+30 0 0 0 0 # EPS_1T2 EPS_2T2 SIGMA_RST2 Wpmax_trac_2 0 0 0 0 # sig_comp_1 B_1C N_1C SIGMA_1MAXC C_1C 1e+30 0 0 0 0 EPS_1C1 EPS_2C1 SIGMA_RSC1 Wpmax_comp_1 # 0 0 0 0 # sig_comp_2 B_2C N_2C SIGMA_2MAXC C_2C 1e+30 0 0 0 0 EPS_1C2 EPS_2C2 SIGMA_RSC2 Wpmax_comp_2 # 0 0 0 0 # sig_12 B_12T N_12T SIGMA_12MAXT C_12T .0040 67.0 .29 0 0 # EPS_1T12 EPS_2T12 SIGMA_RST12 Wpmax_trac_12 0.00 0.00 0.00 0 # GAMMA_INI GAMMA_MAX Dmax 1E31 1E31 .9999 # Fsmooth Fcut 0 0 #---1----|----2----|----3----|----4----|----5----|----6----|----7----|----8----|----9----|---10----| Damage & Failure Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Material Model: Advanced Failure Models Failure models can be coupled with compatible material laws 15 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Puck failure criteria Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Ergebnisse Paul Schischkin 16 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Ergebnisse Paul Schischkin Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Ergebnisse Paul Schischkin 17 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. CAE Process Chain What is known ? • Manufacturing process • Design idea • Available material (fabric, UD-layers, fiber/matrix) • Possible fiber orientations Manufacturing Simulation Phase 1: Free-Size Optimisation What is not known ? • Distribution of material and fiber orientation • Final lay-up of the laminate Phase 2: Sizing Optimisation Phase 3: Shuffle Optimisation Manufacturing and Design In addition, we can answer: • Is it possible to manufacture my design at all? • What is the influence of the manufacturing on the fiber orientation? • Do I have local thickenning or thinning? Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. HyperMesh: Drape Estimator for Composite Fibers • Calculate draping angles and thickness variation • Geometry based, very fast (seconds) • Good estimation for deviation of the fiber oriention, but feedback on manufacturability only indirect material/fiber direction 18 24/03/2015 Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Radioss: Draping Simulation Sandwich method Independant layers method 1 Component 1 Material 1 Property 1 Component for each Ply 1 Material for each Ply 1 Property for each Ply Sandwich material law Fabric material law • • All plies defined in the same property • All layers defined independently Contact interface between the layers Setup in HyperForm user process Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Radioss: Independent Layers Simulation Sliding • Sliding effect between layers can be modeled by contact interface • Each layer is free to slide over other layers • In this example, fiber orientaions are 0, +45, 90 and -45°regarding X axis • The behavior during stamping is different for each layer 19 24/03/2015 Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Composite Forming Sandwich Method • Fiber orientation • The rotation of fibers for each layer can be displayed in HV Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Radioss: Independent Layers with Resin Independant layers + resin method Independent Fiber Layers Shell elements Warm Resin Solid elements Fibers + Resin Shell + solid with coincident nodes Connect material is used to model resin : • A plasticity domain is reached after a small yield stress value • Strain rate dependancy available for user defined curves 20 24/03/2015 Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Altair Partner Alliance: Composites Offering The composite offering from the Altair Partner Alliance reaches across many applications. Working in tandem with OptiStruct and RADIOSS, material modeling, failure modeling, optimization with failure constraints, composite panel analysis and composite panel detail stress analysis can all be performed effectively. Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Agenda 9:00 Registrierung und Kaffee 9:30 Begrüßung Überblick: Aufgabenstellungen in der Strukturmechanik Kaffeepause Thermisch-mechanische Analyse am Beispiel eines Motorblocks Materialmodellierung für unterschiedliche Werkstoffe Strukturberechnung mit MKS-Lastbestimmung 13:00 Mittagessen Composites: Modellaufbau, Berechnung und Auswertung Workflow für Schwingungs- und Akustikanalysen Möglichkeiten zur Strukturoptimierung Diskussion 21 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. HyperWorks Solver Technology Workflow für Schwingungsund Akustikanalysen Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow OptiStruct Statics NVH Thermal RADIOSS MotionSolve AcuSolve FEKO Multi-body Dynamics Thermal and CFD ElectroMagnetics Highly Nonlinear Crash Safety Forming Nonlinear Optimization Smart Multiphysics 22 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow - Modular Construction - Flexible Assembly - Parametric - Model Build Analysis - - TPA Participation Force vs TFs Test forcing Root Cause Analysis Reporting Targets / Metrics Model reduction Rating / Appraisal Focused Optimisation Sensitivity - What if Sensitivity Robustness DOE Correlation Performance Issue Resolution Mass / Cost - - Common Model Parameters - Design Constraints MDO Validation Delivery Size Gauge Shape Topography ERP Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Efficiency gains through integration 61 23 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Improve simulation efficiency CMS super element method Reduction of degrees of freedom Structural/Fluid or Combined Fluid-Structural SE Both Craig-Chang and Craig-Bampton methods General Method (combined free and fixed attachment points) Participation factors (PFCMS) FRF based super element Compliments CMS based approach Faster residual run solution in the medium frequency range (up to 500 Hz) component dynamic stiffness (CDS) - OptiStruct AMSES (Automatic Multilevel Substructuring Eigen Solver) - OptiStruct Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow • CDS superelement technology in OptiStruct • Reduces runtimes for variants to seconds NVH parametric study (e.g. bushings) overnight 24 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow - Modular Construction - Flexible Assembly - Parametric - Model Build Root Cause Analysis Analysis - - TPA Participation Force vs TFs Test forcing Reporting Targets / Metrics Model reduction Rating / Appraisal Focused Optimisation Sensitivity - What if Sensitivity Robustness DOE Correlation Performance Issue Resolution Mass / Cost - - Common Model Parameters - Design Constraints MDO Validation Delivery Size Gauge Shape Topography ERP Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Problem Response Which Mode? Which Panel? What Energy? Which Paths? What’s Moving? What If…? What Power Flow? What DSA? 25 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Noise Diagnostic tools Grid participation Transfer Path Analysis N&V Design Sensitivity Analysis Modal participation Vibration Energy consideration Power Flow Low Frequency Mid Frequency Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Sensitivity filtering Optimization on all possible parameters in full vehicle model is to large Detection of sensitive components or parameters (use normalized variables) Size variables; gauge, stiffness, damping, materials 26 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow - Modular Construction - Flexible Assembly - Parametric - Model Build Analysis - - TPA Participation Force vs TFs Test forcing Root Cause Analysis Reporting Targets / Metrics Model reduction Rating / Appraisal Focused Optimisation Sensitivity - What if Sensitivity Robustness DOE Correlation Performance Issue Resolution Mass / Cost - - Common Model Parameters - Design Constraints MDO Validation Delivery Size Gauge Shape Topography ERP Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow • Frequencies • FRF • Forces, Stress, von-Mises Stress • Functions: max, average, sum, … • Normal displacement, velocity, acceleration, • Frequency sub-ranges • Mode shape • PSD and RMS of displacement, velocity, acceleration, pressure, stress, strain • Acoustic pressure • ERP (Equivalent Radiated Power) • Equations • External routines (DRESP3 -> Fortran, C, HyperMath) 27 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Min(max) Minimize : max [d i ( x)] g j ( x) ≤ 0 Subject to : i = f 0 ,..., f n j = 1, K , m Min (max) with frequency sub ranges Minimize : max[d i ( x)] Subject to : g j ( x) ≤ 0 i = f sub1 ,..., f sub 2 j = 1, K, m System identification n i =1 Subject to :  ∑  w Minimize :  i f i ( x ) − Ti   Ti  g j ( x) ≤ 0 2 j = 1,K, m or Minimize : β with wi Subject to : g j ( x) ≤ 0 f i ( x ) − Ti ≤β Ti j = 1,K, m Curve Fitting Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Trimmed body optimisation using screening techniques TPA Result (Vehicle Model) Modal Participation Panel Participation Plot at peak frequency Component Gauge + Topography optimisation + Full FE Tbdy Model Identify peaks Panel Set as design Variables 28 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Trimmed body optimisation using screening techniques Response Curves Topography Results Minimise Mass with constraint on Response (157Hz) 3dB improvement at peak of interest 40% reduction in mass of panel set Topography result to lead bead pattern design loop 35hr Run time for 30 iterations (12 cpus, AMSES) Gauge Results Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow • Use topology optimisation to design an efficient engine cover • Minimise radiated noise for a given volume of material: • • Equivalent radiated power • Σ area * normal velocity2 * constant Use CMS superelement of non-designable structure 12400 nodes included in ERP response Designable Material 29 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow • Concept design for internal reinforcement structure • Optimised structure reduces ERP by 30% • Run time 2.5hrs (laptop) • Memory 8Gb Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Noise and Vibration - workflow Model Build Analysis Root Cause Analysis Sensitivity Focused Optimisation Fast and Accurate Analysis MDO Validation Delivery OptiStruct HyperStudy Model Quality – Flexibility – Time Reduction Root Cause Understanding Problem Diagnostics – Transfer Path – Sensitivity Robust Optimization Algorithms and Processes Problem Definition – Method selection – Reduction Techniques – Multi Disciplinary 30 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Agenda 9:00 Registrierung und Kaffee 9:30 Begrüßung Überblick: Aufgabenstellungen in der Strukturmechanik Kaffeepause Thermisch-mechanische Analyse am Beispiel eines Motorblocks Materialmodellierung für unterschiedliche Werkstoffe Strukturberechnung mit MKS-Lastbestimmung 13:00 Mittagessen Composites: Modellaufbau, Berechnung und Auswertung Workflow für Schwingungs- und Akustikanalysen Möglichkeiten zur Strukturoptimierung Diskussion Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. HyperWorks Solver Technology Möglichkeiten zur Strukturoptimierung 31 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Performance / Weight / Cost Balancing Cost Weight Performance Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization enhances the comprehensible Design Space Performance Weight Physical Design Limit Optimization Compreh. Design Limit Effort / Cost Weight Current Design Point New Weight Target 32 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Structural Optimization OptiStruct Statics NVH Thermal RADIOSS MotionSolve AcuSolve FEKO Multi-body Dynamics Thermal and CFD ElectroMagnetics Highly Nonlinear Crash Safety Forming Nonlinear Optimization Smart Multiphysics Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Structural Optimization HyperStudy OptiStruct System Level Component Level INSPIRE Concept Design Detailed Design 33 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Detailed Design NVH Optimisation Methodology Size/Shape FreeSize Low number of DV Highlight reinforcement NVH responses patterns Reduced Model Size/Shape Bodynumber Gauge Medium of DV NVH responses Reduce mass Detailed Model Concept Phase Size Size/Shape Topology/Topography Topography High number of DV Medium of DV Mountnumber Locations NVH responses Mode Tracking Concept Model responses Panel NVH Acoustic Response Reduced Model Topology Topology High number of DV Concept Structure and Global stiffness responses loadpath Design definition Space Subsystem/Component Level System Level Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization – State of the Art Technology Today Calculating weight saving..... 15% ..... Calculating weight saving..... 23% Calculating weight saving..... 20% FE-Based optimization technology topology, topography, sizing, shape, staking, free-shape, multi-disciplinary... 34 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Moving in Direction of Strategic Optimization What is strategic optimization? 1. Simultaneous consideration of all important functional requirements (and mass) 2. Application on subsystems, systems and global level 3. Deployment of optimization early in development (Optimization drives Design) 4. Direct connection to development program (management) Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization of Assemblies – getting Strategic! Calculating weight saving..... 20% Calculating weight saving ..... 45% FE-Based optimization technology topology, topography, sizing, shape, staking, free-shape, multi-disciplinary... 35 24/03/2015 Optimization of Assemblies • Multi-model optimization (MMO) aims to simultaneously optimize multiple parts or configurations with common design variables • Simplify the iterative design process associated with optimizing multiple parts, especially when conflicting requirements exist • Models may be entirely different or share identical parts • Models may be assigned individual objectives and constraints, while global responses may also be defined • Performance and scalability are taken into consideration 86 Optimization of Assemblies • Similar models with different meshes (e.g. coarse and fine) • Similar models with subcase-dependent configurations or characteristics (e.g. damping) • Different models sharing identical designable parts • Different models connected at designable locations • Different models with combined objectives or constraints W1 W2 • Any combination of the above min (W1+W2) 87 36 24/03/2015 Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization of Assemblies OptiStruct: Multi-Model Optimization Optimization Problem Single Model Optimization Multi-Model Optimization Copyright © 2013 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization of Assemblies • Modal Performance versus Weight Study • Study creates visibility on relations between performance and weight • Example is showing a design limit not known to customer Modal Performance v BIW Weight Require Major Architectural Changes 140% 135% Risk Torsion Mode (Hz) 130% 125% 120% 115% Target 110% 105% Optimum Modal Performance versus Weight 100% 95% 93% 94% 95% 96% 97% 98% 99% 100% 101% 102% BIW Weight (Kg) 37 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Optimization - Fundamentals Robust and Repeatable Analysis Predict the Trend Fast and Accurate Analysis Performance and Scalability Robust Optimization Algorithms and Processes Convergence and Efficiency Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Altair Solver Technology Structural Analysis Crash, Safety, Impact & Blast Thermal Analysis Fluid Dynamics Systems Simulation Manufacturing Simulation ElectroMagnetics Multiphysics Analysis and Optimization 38 24/03/2015 Copyright © 2012 Altair Engineering, Inc. Proprietary and Confidential. All rights reserved. Altair HyperWorks: Performance Enhancing Software 39