• Like
CSAR Legacy
Upcoming SlideShare
Loading in...5
×
Uploaded on

 

More in: Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Be the first to comment
    Be the first to like this
No Downloads

Views

Total Views
989
On Slideshare
0
From Embeds
0
Number of Embeds
0

Actions

Shares
Downloads
1
Comments
0
Likes
0

Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
    No notes for slide

Transcript

  • 1. CSAR Legacy Michael T. Heath February 21, 2007 ©2007 Board of Trustees of the University of Illinois ©
  • 2. Center for Simulation of Advanced Rockets
    • Detailed, whole-system simulation of solid propellant rockets under normal and abnormal operating conditions
    • Accurate models of physical components
    • Subscale simulations of materials and accident scenarios
    • Software framework to facilitate component integration
    • Computational infrastructure to support large-scale simulations
    • Research collaborations with government laboratories and rocket industry
    Overarching Goal: Simulation of solid propellant rockets from first principles STS Discovery
  • 3. Solid Rocket Booster Igniter Case Core flow Propellant Nozzle Joint & slot Ammonium perchlorate (oxidizer) Polymeric binder and aluminum particles (fuel) 200  m Micrograph of solid propellant
  • 4. Challenges in Rocket Simulation
    • Full 3-D modeling essential to capture physics
    • Strong, nonlinear coupling among components
    • Complex, dynamically changing geometry
    • Extremely diverse spatial and temporal scales
    • Complex material properties and physical processes
    • Enormous computational capacity required for high-resolution simulation of full burn
    • Scalability to 1000s of processors essential
  • 5. CSAR Legacy
    • Rocstar code
    • Major simulations
    • Fundamental research
    • Impact
      • Campus
      • State and local
      • National and international
    • Human resources
    • Lessons learned
    Aft Joint Slot Forward Joint Slot Titan IV Turbulence
  • 6. Rocstar Code Suite
    • Nation’s only 3-D integrated simulation capability for solid rocket motors
    • How did we accomplish this?
      • Adequate financial resources
      • Long time horizon
      • Critical mass of talent across disciplines
      • Availability of massively parallel platforms
    • Now in position to solve problems of national interest, and already doing so
      • Five-segment RSRM for lunar program
    Fully-coupled RSRM — fluid temperature and propellant stress profiles
  • 7. Rocstar Code Suite
    • Scalable to thousands of processors
    • Applicable to many other problems
      • Fluid-structure interaction
      • Reactive, multiphase flows
    • Examples
      • Pyrovalve
      • Helicopter blades
      • Volcanoes
      • Tall buildings
    NASA Booster Separation Motor
  • 8. Major Simulations “Science with the Code”
    • Full RSRM
    • Titan propellant slumping
    • Turbulence around flexible inhibitor
    • Aluminum impingement on nozzle
    Al particles impacting nozzle Flow field near slumping propellant
  • 9. Verification, Validation, and Uncertainty Quantification
    • Verification
      • Convergence studies
      • Comparison with analytical solutions
      • Method of manufactured solutions
      • Comparison between codes
        • e.g., do Rocflo and Rocflu give similar results for same problem?
      • Code coverage
      • Coupled codes — verify component codes first
    • Validation
      • RSRM (ATK/Thiokol & NASA)
      • Titan IV (Lockheed Martin)
      • 15 and 70 lb BATES (AFRL)
      • Lab scale rocket (NAWC)
      • Attitude Control Motor (Aerojet)
      • Shock tube with thin steel panel
    • New sampling techniques for UQ
  • 10. Navy “StarAft” Motor
    • Full burnout
    • Tactical motor
  • 11. Titan IV Motor
    • Heavy-lift booster
    • Accident scenario
    • Detailed vorticity studies
  • 12. NASA RSRM
    • Late stage in full burn (~100 s)
    • Flexible inhibitor
    • 3-D vorticity study
  • 13. NASA RSRM
    • Burnout simulation
      • Red is burning propellant; blue is at insulated surface
    • Viewing
      • Propellant/fluid interface
      • Inhibited surfaces
    • Unique capabilities
      • High rate “time zooming”
      • Significant burnout — never seen in industry
      • Case constraints
      • Propellant walkback
      • Inhibitor regression
      • Dynamically changing topology
    Initially inhibited surface
  • 14. Fundamental Research in Fluids and Combustion
    • Multiphase flow
      • Equilibrium Eulerian method for fine particles
      • Lagrangian superparticles
    • Turbulence modeling
      • Optimal LES
    • Time zooming
    • Propellant morphology
      • Parallel packing code
    • Propellant modeling
      • 3-D simulations with Rocfire
  • 15. Fundamental Research in Structures and Materials
    • Constitutive and damage modeling
      • Heterogeneous propellants and HE
      • Metallic components
    • Crack propagation
      • Burning, pressure driven
    • Multiscale materials modeling
    • Molecular-level modeling of material interfaces
    • Space-time discontinuous Galerkin methods
  • 16. Fundamental Research in Computer Science & Applied Math
    • Parallel programming environments
      • Charm++, AMPI, ParFUM
    • Parallel I/O
      • Panda
    • Parallel performance monitoring/modeling
      • Rocprof
    • Meshing
      • Mesh smoothing, repair, and remeshing
      • Mesh quality metrics
      • Hybrid geometric/topological partitioner
    • Visualization
      • Rocketeer
  • 17. Fundamental Research in Integration Methodology
    • Object-oriented integration framework
    • Flexible parallel orchestration
    • Stable component coupling and time-stepping
    • Accurate and conservative data transfer based on common refinement
    • Stable and efficient explicit surface propagation
  • 18. Computational Science and Engineering Education Program 13 departments 130 faculty associates 130 grad students 10 graduate fellows Research Program Computational Science & Engineering Option Center for Simulation of Advanced Rockets Center for Process Simulation and Design DOE funded $4-5 million per year 10 year program 20 faculty 35 graduate students 5 undergrads 20 professional staff NSF funded $6 million over 6 years 12 faculty 13 students & postdocs Midwest Structural Sciences Center AFRL funded $5 million over 5 years 10 faculty 15 students & postdocs ©2007 Board of Trustees of the University of Illinois
  • 19. Campus Impact
    • CSAR put CSE “on the map”
    • CSAR followed by two additional centers hosted by CSE
      • Center for Process Simulation and Design
      • Midwest Structural Sciences Center
    • CSE seen as “center of centerness” on campus, advising other groups seeking or establishing centers
    • CSE taken as model in developing other new interdisciplinary programs (e.g., bioinformatics)
    • CSE’s Turing cluster
      • Established value of large-scale computing facilities for local users
      • Financial commitment from campus administration
  • 20. Turing G5 Xserve Cluster
    • 1536 compute processors
    • Dual 2GHz G5 processors per node
      • 4GB RAM per node
      • Myrinet interconnection network
    • 6 Xserve head nodes
      • 49TB Xserve RAID
    • 512 and 256 node partitions
      • 23 racks (15 and 8 racks, respectively)
    • Mac OS 10.4 Server
  • 21. State and Local Impact
    • Economic development through industrial collaborations
      • Boeing, Caterpillar
    • Visibility of simulation promoted through various venues
    • Recurring appearances in media (TV, newspapers)
    RSRM inhibitor vorticity Flexible inhibitor
  • 22. National and International Impact
    • Publications (~900 journal articles and proceedings)
    • Conference sessions
    • Industrial collaborations
      • ATK Thiokol
      • Aerojet
      • Boeing
      • Caterpillar
    • Other government agency funded research
      • USAF
      • NASA
    • Technology transfer to and from NNSA labs
    • Participation in Integrated Product Team for USAF solid propulsion program
  • 23. Pedagogical Impact 1st edition, 1997 2nd edition, 2002
  • 24. Human Resources
    • Supported 160 graduate students, about 40 of whom have gone on to DOE labs and many others to positions in government or industry
    • Recruited and developed 50 research staff members, many of whom have gone on to positions at DOE labs, industry, and academia
    • Contributed to faculty career development (promotions, awards, etc.), counter to conventional wisdom about collaboration
  • 25. CSAR Students and Staff in DOE/NNSA Labs
      • Boyana Norris, ANL
      • Michael Parks, SNL
      • Jason Petti, SNL
      • Ali Pinar, LBNL
      • Jason Quenneville, LANL
      • Christopher Siefert, SNL
      • Donald Siegel, SNL
      • Christopher Tomkins, LANL
      • Michael Tonks, LANL
      • Jason Weber, BBL
      • Bradley Wescott, LANL
      • Amanda White, LANL
      • Steven Wojtkiewicz, SNL
      • Jin Yao, LLNL
      • Jack Yoh, LLNL
    • Former CSAR Staff Now at DOE Labs (5)
      • Michael Ham, ORNL (LANL)
      • Dennis Parsons, LLNL
      • James Quirk, LANL
      • Mark Short, LANL
      • Jeff Vetter, LLNL (ORNL)
    • Former CSAR Students Now at DOE Labs (34)
      • Tyler Alumbaugh, LLNL
      • Balakumar Balasubramanium, LANL
      • Michael Bange, LANL
      • Benjamin Chorpening, SNL
      • Nathan Crane, SNL
      • Zhiqun Deng, PNNL
      • Erik Draeger, LLNL
      • Michelle Duesterhaus, SNL
      • Jeff Grossman, LBNL (LLNL )
      • Arne Gullerud, SNL
      • Thomas Hafenrichter, SNL
      • Rebecca Hartman-Baker, ORNL
      • Jason Hales, SNL
      • Jonghyun Lee, ANL
      • Vanessa Lopez, LBNL
      • Xiaosong Ma, ORNL
      • Greg Mackey, SNL
      • Burkhard Militzer, LLNL
      • Jeffrey Murphy, SNL-L
    Gray now in other employment.
  • 26. What Worked Well?
    • Single, over-arching application
      • Maintain focus, avoid mission creep
    • Pre-existing, interdisciplinary framework independent of individual academic departments
    • Tight integration of CS with application disciplines
    • Professional staff in addition to faculty and students
    • Wealth of thesis-sized research issues for graduate students
    • Pipeline of students from Illinois to lab employment
  • 27. What Could Have Worked Better?
    • Collaborations with NNSA labs were numerous, but could have been tighter and more comprehensive
    • Suggestions for improvement
      • Common application focus between labs and university
      • Deeper commitment to joint activities, such as long-term visits and joint hiring (e.g., postdocs)
      • Two-way software exchanges
  • 28. ©2007 Board of Trustees of the University of Illinois http://www.csar.uiuc.edu ©
  • 29. Prof. Michael T. Heath, Director Center for Simulation of Advanced Rockets University of Illinois at Urbana-Champaign 2270 Digital Computer Laboratory 1304 West Springfield Avenue Urbana, IL 61801 USA [email_address] http://www.csar.uiuc.edu telephone: 217-333-6268 fax: 217-333-1910