Integrated Computational Materials Science & Engineering

Size: px

Start display at page:

Download "Integrated Computational Materials Science & Engineering"

Cody Davidson
6 years ago
Views:

1 Integrated Computational Materials Science & Engineering 3/3/2012 Gregory Ho, Ph.D. Advanced Technology Laboratories Cherry Hill, NJ 1

Thesis Statement Future innovation in defense-related

co-selection of material properties, product design, and

Material Choices: Brick Straw Wood Composite Gingerbread

Manufacturing Process: Whittle down sticks Bundle 10 to a

2 Thesis Statement Future innovation in defense-related products is critically dependent upon simultaneous co-selection of material properties, product design, and manufacturing process. Material Choices: Brick Straw Wood Composite Gingerbread Interdependent! Manufacturing Process: Whittle down sticks Bundle 10 to a tow Precision placement Glue? Product: Roof, walls, door Hurricane Safe Well-ventilated (airflow/hour) Max Cost < $1000 Max Weight Min Volume 2

3 Thesis Statement This type of co-design is only possible with the maturation of advanced computational methods: 1. Multiscale modeling 2. Material informatics 3. Computational manufacturing The ultimate goal of ICMSE (Integrated Computational Materials Science & Engineering) is to demonstrate we can do simultaneous co-design! 3

4 Outline Introduction: Why do we need ICMSE? Case Studies Conclusion and Future Work: What problems need to be solved? 4

5 $ Billion Federal Budget Outlook (Outlays $ Billions) $4,000 $3,500 $3,000 Defense 050 (Including OCO) Non-Defense Interest on Debt Mandatory Programs $2,500 $2,000 Annual Deficit $1,500 $1,000 $500 $ Fiscal Year Source: OMB Interest on Debt Exceeds Defense Spending Starting in FY2013 5

6 DOD Budget by Title ($ Billions) $700 $600 Overseas Contingency Operations $500 $400 $300 Supplementals Operations & Maintenance Other $200 Military Personnel $76.1B $100 RDT&E Investment Procurement $112.9B $ Source: OMB and DOD FY11 Budget Request Documents Projected Eng. and Mfg. Affordability are King in next Decade(s) Funding covers much wider threat spectrum; flexibility needed 6

7 Solution: Advanced Materials e.g., carbon nanostructures Need multiscale modeling with robust model linkages Fast, simultaneous co-design in simulation to reduce redesign Virtual Microscope Simultaneously, maximize performance and efficiency (cost, weight, manufacturability) device / structure micro scale meso scale Atomic scale Quantum Scale Improved products at for lower cost 7

8 Outline Introduction: What is ICMSE, why do we need it? Case Studies Conclusion and Future Work: What problems need to be solved? 8

Scalable Nanocomposites http://www.compositesworld.

9 Scalable Nanocomposites 9

10 Scalable Nanocomposites 10

11 Multiscale Modeling in the Loop Material models from nano to micro Macroscale models of Product Topological optimzation Material Properties Product Design Manufacturing Methods Process Models 11

Dynamics (LAMMPS), Concurrent Finite Element (ABAQUS, MDS) Multiscale Modeling can be an Art,

12 Multiscale Modeling Hierarchy Solid Mechanics Analytical model Solid Mechanics (ABAQUS) (B-spline analysis tool, BSAM) First Principles (VASP) Molecular Dynamics (LAMMPS) Molecular Dynamics (LAMMPS), Concurrent Finite Element (ABAQUS, MDS) Multiscale Modeling can be an Art, Few Standard Tools. Sometimes problems can be too complex to model! 1 Picture credits: Brian Wardle, et al. 12

Informatics Analysis Tools Multiple Linear Regression (MLR) K Means Clustering (PCA) Partial Least Squares (PLS) Support Vector Machines (SVM) Random Forest (RF) Star Plots Y-Scramble Validation

13 Informatics Analysis Tools Multiple Linear Regression (MLR) K Means Clustering (PCA) Partial Least Squares (PLS) Support Vector Machines (SVM) Random Forest (RF) Star Plots Y-Scramble Validation Predictive Analysis Sensitivity Analysis Multivariable Comparative Analysis Multivariable Predictive Analysis Inverse-QSAR A combination of linear and nonlinear modeling with data visualization addresses model interpretability and predictivity 13

14 The Importance of Informatics 14

15 3D Printing Revolution 15

16 Computational Manufacturing Need to understand: What are the properties of the material you create? What beneficial design changes does this allow you to make? PRINTED SAMARAI: (3D printed; <15 mins assembly) 16

17 Pouring an Airplane? 17

18 Outline Introduction: What is ICMSE, why do we need it? Case Studies Conclusion and Future Work: What problems need to be solved? 18

19 Conclusion Future innovation in defense-related products is critically dependent upon simultaneous co-selection of material properties, product design, and manufacturing process. This type of co-design is only possible with the maturation of advanced computational methods: Multiscale modeling Material informatics Computational manufacturing 19

20 Future Critical Needs for ICMSE Automated, dynamically reconfigurable factories for the production of any product using any material with tailored properties in small quantities Foundational failure physics method development Bonded joints / Large fastened structures Moisture intrusion / impact in sandwich structures Rigorous quantification of model error Scale linkages in multiscale modeling: Mapping from first principles to semi-empirical Standardization of model interoperability (process models and material models) Multiscale meta-workflows and data shared across the community of interest 20

Android Analog for Interoperable Structural Models and Design Tools Structures Wing Armor Gunner Seat Pi Joint Missile Body Compression After Impact CNTs Graphene Foams Functional Grading Unitized

21 Android Analog for Interoperable Structural Models and Design Tools Structures Wing Armor Gunner Seat Pi Joint Missile Body Compression After Impact CNTs Graphene Foams Functional Grading Unitized Structure Design Framework Adaptive Stiffness Tailored Stiffening Load Case Environment Bonded Joint ASTM Tests Model Libraries Adhesives Resin Systems Thermoplastic Fibers Sandwich Structures Fiber Treatments Modeling Tools Model Linkages Multiscale Linkages Interface Models ABAQUS LAMMPS VASP NASTRAN OpenFOAM Materials Informatics MOLDFLOW Analytical 21

Enabling Materials Innovation for Future Economic Growth

Enabling Materials Innovation for Future Economic Growth Building an Integrated MGI Accelerator Network Georgia Tech Hotel and Conference Center 5 June 2014 Rick Barto, Ph.D. Sr. Manager, Applied Sciences