1. report date 2. report type briefing charts
TRANSCRIPT
REPORT DOCUMENTATION PAGE Form Approved
OMB No. 0704-0188 Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE (DD-MM-YYYY) 24 September 2015
2. REPORT TYPEBriefing Charts
3. DATES COVERED (From - To) 01 Sept 2015 – 24 Sept 2015
4. TITLE AND SUBTITLE Challenges and Opportunities in Propulsion Simulations
5a. CONTRACT NUMBER
5b. GRANT NUMBER
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) Venkateswaran Sankaran
5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER Q12J
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION REPORT NO.
Air Force Research Laboratory (AFMC) AFRL/RQR 5 Pollux Drive Edwards AFB, CA 93524-7048
9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) Air Force Research Laboratory (AFMC) AFRL/RQR 11. SPONSOR/MONITOR’S REPORT
5 Pollux Drive NUMBER(S)
Edwards AFB, CA 93524-7048 AFRL-RQ-ED-VG-2015-370 12. DISTRIBUTION / AVAILABILITY STATEMENT Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES For presentation at invited seminar at University of Michigan, Ann Arbor, MI, PA Case Number: # 15590; Clearance Date: 9/24/2015 14. ABSTRACT Briefing Charts/Viewgraphs
15. SUBJECT TERMS N/A
16. SECURITY CLASSIFICATION OF:
17. LIMITATION OF ABSTRACT
18. NUMBER OF PAGES
19a. NAME OF RESPONSIBLE PERSON
V. Sankaran
a. REPORT Unclassified
b. ABSTRACT Unclassified
c. THIS PAGE Unclassified
SAR 40 19b. TELEPHONE NO
(include area code)
N/A Standard Form
298 (Rev. 8-98) Prescribed by ANSI Std. 239.18
Venke Sankaran AFRL/RQ
Challenges & Opportunities in Propulsion Simulations
University of Michigan Ann Arbor, 24 Sept 2015
Distribution A – Approved for public release; Distribution Unlimited
Distribution A – Approved for public release; Distribution Unlimited
AFRL Mission
Leading the discovery, development, and
integration of affordable warfighting technologies for our air, space, and
cyberspace force.
Distribution A – Approved for public release; Distribution Unlimited
Space Vehicles • Space&Electronics&• Space&Environmental&Impacts&&&Mi4ga4on&
• Space&OE/IR&• Space&Experiments&• Pla;orms&&&Opera4ons&Technologies&
&
Materials and Manufacturing • Func4onal&Materials&&&Applica4ons&
• Manufacturing&&&Industrial&Technology&
• Structural&Materials&&&Applica4ons&
• Support&for&Opera4ons&&
Sensors • Advanced&Devices&&&Components&
• Layered&Sensing&Exploita4on&
• Mul4GInt&Sensing&(RF/EO)&
• Spectrum&Warfare&&
• Fuze&Technology&• Muni4ons&AGN&C&• Muni4ons&System&Effects&Science&
• Ordinance&Sciences&• Terminal&Seeker&Sciences&&
Munitions
Aerospace Systems • Air&Vehicles&• Control,&Power&&&&Thermal&Management&
• High&Speed&Systems&• Space&&&Missile&Propulsion&
• Turbine&Engines&
&
Directed Energy • Directed&Energy&&&EO&for&Space&Superiority&
• High&Power&Electromagne4cs&
• Laser&Systems&• Weapons&Modeling&and&Simula4on&
Information • Autonomy,&C2,&&&&Decision&Support&
• Connec4vity&&&Dissemina4on&
• Cyber&Science&&&Technology&
• Processing&&&Exploita4on&
Human Performance • BioGeffects&• Decision&Making&• Human&Centered&ISR&• Training&&
AF&Office&of&Scien4fic&Research&
• Aerospace,&Chemical&&&Material&Sciences&
• Educa4on&&&Outreach&• Mathema4cs,&Informa4on,&&&life&sciences&
• Physics&&&Electronics&&
AFRL Technical Competencies
‹#›
Aerospace Systems Directorate
MISSION/VISION: Leading discovery and development of world class integrated Aerospace Systems S&T for national security
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Edwards AFB, CA
Wright-Patterson AFB, OH
Established 1917
Established 1947
• Liquid Rocket Engines • Solid Rocket Motors • Spacecraft Propulsion
Technologies • Gas Turbine Engines • Vehicle Flight Systems • Airframe Aerodynamics
and Structures • Hypersonic Propulsion
Technologies
~580 programs (OH ~470 / CA ~110)
~1814 on-site people
(OH ~633 gov’t, ~771 contractors) (CA ~203 gov’t, ~207 contractors)
~$540M / year (core S&T)
(FY14 PB includes FY14-FY18) (OH ~$460M / CA ~$80M)
~$3.9B in facilities
(OH ~$1.9B / CA ~$2B)
Aerospace Systems Directorate
As of 31 July 2013
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
!
MOJAVE!
BORON!HWY 58!
LANCASTER!
AVENUE E!
HIG
HWAY
14!
LANC
ASTE
R BL
VD.!
140t
h ST
REET
EAS
T!
RESERVATION BOUNDARY!
0! 5! 1!0!
SCALE IN MILES!
HWY 395!
ROSAMOND BLVD.!
MERCURY BLVD.!
ROCK
ET S
ITE
ROAD
!
EDWARDS AIR FORCE BASE! Air Force!Research!
Laboratory!Site!
ROGERS!DRY LAKE!
ROSAMOND!DRY LAKE!
AFTC!
Edwards AFB
D.C.
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
• F-1 engine testing for the Saturn V Rocket that put Men on the Moon
History of the Rock
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Available Facilities
Bench-level Labs High Thrust Facilities • 19 Liquid Engine stands, up to
8,000,000 lbs thrust • 13 Solid Rocket Motor pads, up to
10,000,000 lbs thrust"Altitude Facilities • From micro-newtons to 50,000
lbs thrust
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Space and Missile R&D Building Block Process
6.1 6.2 6.3
Distribution A: Approved for public release; unlimited distribution
‹#›
45-80% of directed energy weapon
weight and volume
70-90% of launch weight 40-60% of system cost
40-70% of cruise missile weight; the critical factor in survivability, lethality, & reach
60-80% of tactical missile weight the critical factor in range & time-to-target
40-60% of aircraft TOGW 20-40% of system life
cycle cost
Propulsion & Power are Important!
50-70% of satellite weight 25-40% of
system cost the life-limiting factor
Air Force fuel costs were $6B in FY07 alone
Distribution A: Approved for Public Release; Distribution Unlimited
Distribution A – Approved for public release; Distribution Unlimited
Vision
! Maintain hands-on rocket M&S tool expertise ! Develop rocket physics and numerics expertise ! Promote modular computational infrastructures ! Lead in new and emerging research areas
Establish leadership in rocket M&S
Distribution A – Approved for public release; Distribution Unlimited
Themes
! Relevance to customers and programs ! Strong experimental interactions ! Model evaluation & development focus ! Partnership with community
Lead adoption of model-driven development
Distribution A – Approved for public release; Distribution Unlimited
Levels of Analysis
! Level 0 – Empirical relations ! Level 1 – 0D or 1D analysis ! Level 2 – Multi-dimensional analysis ! Level 3 – RANS coupled to multi-physics ! Level 4 – LES/DES/DNS simulations
Combustion CFD example
Distribution A – Approved for public release; Distribution Unlimited
Types of Codes
! Commercial – Fluent, STAR-CCM ! Small Business – CRAFT, CFD++ ! University/In-house – LESLIE, GEMS ! Open-Source – OpenFOAM ! Govt Codes – NCC, Coliseum, CREATE
A combination of code solutions is necessary!
Distribution A – Approved for public release; Distribution Unlimited
Multi-Level Hierarchy
Engineering Models (e.g. Galerkin expansion solns)
Analytical Solution (e.g. Linear Euler)
Medium-Fidelity (e.g. Non-linear Euler, URANS)
High Fidelity: (e.g. LES or Hybrid RANS/LES)
Incr
ease
d Fi
delit
y
Increased Cost
Experiments
Response Functions
Near Term Spinoff
Response Functions feed back to improve lower cost models
Combustion stability example
Utilize high-fidelity solutions to develop next-gen design tools
‹#›
Anderson (Purdue) • AFOSR • NASA CUIP • ALREST • AFRL Frederick (UAH) • NASA CUIP • AFRL • ALREST Karagozian (UCLA) • AFOSR Leyva, Talley (AFRL) • AFOSR • ALREST Cavitt (Orbitec) • AFRL • ALREST Santoro (PA State) • AFOSR (core) • NASA CUIP • ALREST Yu (Maryland) • NASA CUIP Zinn (GA Tech) • AFOSR Nestleroad Engin’ng • MDA
Data-Centric Model Development
Full Scale (existing and HCB)
Catalyst
Bed
Oxidizer Manifold
Fuel Injector Manifold
Highly Instrumented Chamber Section
Oxidizer Post
Chamber Extension
Discretely Variable Chamber Length
Converging Nozzle
= High Frequency Pressure Measurement Location Model Data
GA
Tec
h
Acoustics
AFR
L - c
oax
Driven jets
Longitudinal CI
Pur
due
Standing CI
Pur
due
– m
ulti
elem
UA
H –
sin
gle
elem
Experiments Spinning CI
HCB to be heavily instrumented to provide CI data
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Payoffs
Damaged F-1 engine injector faceplate due to combustion instability
• The Past: Test Driven Development F-1 > 3000 tests (59 R&D engines) J-2 > 1000 tests (43 R&D engines) SSME >900 tests (27 R&D engines) RL-10 >700 tests
• The Future: Model Driven Development 20-50 tests?? ICBMs: $25B in Life Cycle Cost savings
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Integrated Motor Life Management
In-House: • Validation of A&S modeling capability• AFNWC funded supported for ANDESimprovement (Automated NDE DataEvaluation System)
The WOWs • Potential to provide >20% reduction in LCC• Provide accurate, near-real-time motor healthcondition (diagnostics)
• Provide individualized service life estimates(prognostics)
• Transition opportunity ~ 2018
Goals: Reduce predictive uncertainty of future state of a motor on an individual basis by 20%/50% (near/far term goals)
Analysis Command &
control
Initial state and inspection data
Data processing and
storage
Sensors to include: temperature, humidity, case damage, propellant slump, acceleration, and TVA displacement and load
First integration of motor specific sensor data to
advanced aging models to provide a
individualized service life
estimate
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
MCAT (Motor Component Assessment Technology)
What are we doing? Developing new solid rocket motor (SRM) components and M&S tools that decrease inert weight by 20%.
Customer why? High-speed penetrator weapons will enable attack of deeply-buried targets.
Tech Reason? New M&S tools may show possibility of higher efficiencies from SRM designs.
Transition? 3 of 6 FY12 task orders support an AFRL FCC. 1 of 6 FY12 task orders supports AFNWC
In-House: Experiments to validate new models
The WOWs • The AFNWC propellant task is part of a plan that may save $2.1B in future acquisition costs • We are only gov’t lab doing solid rocket motor R&D for launch & strategic needs
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Electric Propulsion
Plasma propulsion increases Isp by 10x, reducing s/c propellant 10x, lighter and/or more capable s/c
Developing new technologies that enable less expensive, more maneuverable and agile s/c
Reducing launch mass substantially reduces launch cost, increases payload fraction, and enables missions otherwise not possible
In-House: • Test facilities o 8 vacuum
chambers o Thruster design o Diagnostics o Validation
• Advanced numerics
The WOWs: • AEHF requested assistance with thruster
performance verification • Developed propulsion module for FalconSat-5
tech demo, including spacecraft interaction diagnostics
• Cubesat EP propulsion module selected by 2 constellations
• National M&S effort for EP coordination
AEHF SV2 Sensor
Package
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Coliseum
• Engineering tool to study EP plumes and their effect on spacecraft • Realistic Geometries • Flexible Materials databases
OB
JEC
TVES
A
PPR
OA
CH
• Develop C-based framework code (Coliseum) and plasma submodules (Draco, Aquila, Ray)
• Couple with HPHall – hybrid fluid/PIC code
Realtime coupling between HPHall and Coliseum allows us to track evolution of time dependent
features all the way from the anode to many thruster lengths downstream
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Next-Gen Framework
• Need new computational framework to leverage modern computer science, algorithms and hardware acceleration and provide much-improved capabilities to user base
OB
JEC
TVES
A
PPR
OA
CH
• Build modular C++ object-oriented framework with architecture to leverage Nvidia GPU accelerators • Release common
computational infrastructure as Distro A for collaboration
• Add physics modules as either Distro C or A to accomplish ITAR mission
• Version 1 (est. beta release end of 1QFY16) • Coliseum replacement capability
- Electrostatic pushes - Triangulated spacecraft geometries - Electrostatic plasma solvers (Boltzmann & Poisson) - Volumetric collisions - Hooks to communicate with HPHall - Macroparticle surface/boundary interactions
• Version 2 (est. beta release end of 4QFY16) • HPHall replacement capability
• Version 3+ • Higher fidelity device models (HET and FRC)
LOOKING AHEAD
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Basic Plasma Propulsion Research IM
PAC
TS
• Close coupling of 6.1/6.2 programs enables cutting edge academic and lab research to transition into engineering codes
• Provides exceptionally qualified workforce
• Develop hybridized fluid / kinetic solvers to efficiently study multiple scales present in many plasma processes
• Develop more computationally efficient, higher-fidelity Collisional-Radidative (C-R) and radiation transport models to improve simulations and mirror experiments
• Hybridize Vlasov and multifluid models • Apply advanced fluid simulation methods to FRC to
develop true design capability
Two-stream plama instability (Vlasov)
APPROACH
Distribution A: Approved for Public Release; Distribution Unlimited
Distribution A – Approved for public release; Distribution Unlimited
Hybrid CPU-GPU Framework
24
‹#›
Feature Titan Summit
Application Performance Baseline 5-10x Titan
Number of Nodes 18,688 ~3,400
Node performance 1.4 TF > 40 TF
Memory per Node 38GB (GDDR5+DDR3) >512 GB (HBM + DDR4)
NVRAM per Node 0 800 GB
Node Interconnect PCIe 2 NVLink (5-12x PCIe 3)
System Interconnect (node injection bandwidth)
Gemini (6.4 GB/s) Dual Rail EDR-IB (23 GB/s)
Interconnect Topology 3D Torus Non-blocking Fat Tree
Processors AMD Opteron™ NVIDIA Kepler™
IBM POWER9 NVIDIA Volta™
File System 32 PB, 1 TB/s, Lustre® 120 PB, 1 TB/s, GPFS™
Peak power consumption 9 MW 10 MW
Titan vs. Summit
Source: R. Sankaran, ORNL
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Hydrocarbon Boost
In-House: • Subscale facility to mitigate
combustion stability risk
The WOWs: • Design, build, test ORSC LOx/Kerosene Liquid Rocket Engine Tech Demonstrator
• 250K-lbf with high Throttle Capability
• 100 Life Cycle with 50 cycle overhaul
Advanced LRE Tech Base
– Required to replace Russian RD-180 on EELV
– Establishes Ox-rich staged combustion (ORSC) tech base for U.S.
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Liquid Rocket Combustion Instability
Source: Purdue University
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Transverse Mode Instabilities
Source: Purdue University
Distribution A: Approved for Public Release; Distribution Unlimited
Distribution A – Approved for public release; Distribution Unlimited
Next-Gen R&D
29
High-Fidelity • Modular framework • Efficient grid types • High-Order Accuracy • Adaptive Mesh • Adaptive Physics • Advanced models • Emerging architectures
Multi-Fidelity • Use high-fidelity to train
low fidelity • LES simulations
– Limited number off-line calculations with DOE
• Reduced Order Model – Obtain response
functions from LES • Design Tool
– Non-linear Euler with response functions
Distribution A – Approved for public release; Distribution Unlimited
Mesh Types for Reacting-LES
30
Unstructured Mesh • Rarely automated • Very inefficient • Usually limited to
second-order accuracy • Difficult to adapt • Good at capturing
complex geometries • Very good for boundary-
layer resolution
Cartesian Mesh • Automatic generation • Highly efficient • High-order accuracy
– Usually fifth- or seventh-order accurate
• Amenable to adaption • Poor geometry definition • Proper boundary-layer
resolution is inefficient
Solution: Combine unstructured near-body mesh with Cartesian off-body mesh
Distribution A – Approved for public release; Distribution Unlimited
Dual-Mesh Paradigm
Extend'CREATE,AV’s'dual,mesh'paradigm'to'internal'rocket'flowfields''•'Cartesian'codes'are'10x'faster'than'unstructured''•'FiDh,order'accuracy'means'10x'fewer'grid'points''•'AdapFve'mesh'puts'grid'refinement'where'needed''•'AdapFve'physics'can'be'tailored'to'combusFon'
31 Source: Wissink et al., CREATE
Distribution A – Approved for public release; Distribution Unlimited
Turbulent Combustion
32
‹#›
Advanced Numerics
MUSCL scheme!
LESLIE3D!Central scheme!
Ref: 2014 – Cocks et al. “Towards Predictive Reacting Flow LES”
Need to determine OPTIMAL discretization schemes for Reacting LES
Algorithm comparisons: - Identical SGS - Differences in numerical schemes’ dissipation
American Institute of Aeronautics and Astronautics
2
In a previous work3, the current authors conducted a careful to study to demonstrate that even when using the same grid, time-step, boundary conditions and physical models, it is possible to get completely different results if different CFD codes (with different numerical schemes) are used. A key result from that work is reproduced here in Figure 1 to illustrate the differences in the size and number of the turbulence length scales and the differences in the nature of the flame shape altogether. As further evidence of unrepeatability, the same simulation is repeated using a single CFD code with the same boundary conditions but two different numerical schemes (a central differencing scheme with no artificial dissipation and a MUSCL scheme). Here the numerical model includes the grid size, numerical scheme, artificial viscosity, if any, and boundary conditions. Physical models include the closure models for subgrid turbulence and subgrid turbulence-chemistry interaction, and characteristics of the LES filter. Figure 2 highlights the underlying problem that a given combination of numerical model and physical models can yield a drastically different and unpredictable solution.
Figure 1: Instantaneous temperature contours from simulations using different codes3 for the Volvo test case13-15. Codes top to bottom: CharLES, LESLIE3D, OpenFOAM and Fluent.
Figure 2: Instantaneous temperature contours from simulations using the LESLIE3D code that utilize a MUSCL scheme (top) and a central scheme (bottom).
Dow
nloa
ded
by A
ir Fo
rce
Rese
arch
Lab
orat
ory
(DET
-7 A
FRL/
RQO
I) - E
dwar
ds A
FB o
n N
ovem
ber 1
4, 2
014
| http
://ar
c.ai
aa.o
rg |
DO
I: 10
.251
4/6.
2014
-082
6
American Institute of Aeronautics and Astronautics
2
In a previous work3, the current authors conducted a careful to study to demonstrate that even when using the same grid, time-step, boundary conditions and physical models, it is possible to get completely different results if different CFD codes (with different numerical schemes) are used. A key result from that work is reproduced here in Figure 1 to illustrate the differences in the size and number of the turbulence length scales and the differences in the nature of the flame shape altogether. As further evidence of unrepeatability, the same simulation is repeated using a single CFD code with the same boundary conditions but two different numerical schemes (a central differencing scheme with no artificial dissipation and a MUSCL scheme). Here the numerical model includes the grid size, numerical scheme, artificial viscosity, if any, and boundary conditions. Physical models include the closure models for subgrid turbulence and subgrid turbulence-chemistry interaction, and characteristics of the LES filter. Figure 2 highlights the underlying problem that a given combination of numerical model and physical models can yield a drastically different and unpredictable solution.
Figure 1: Instantaneous temperature contours from simulations using different codes3 for the Volvo test case13-15. Codes top to bottom: CharLES, LESLIE3D, OpenFOAM and Fluent.
Figure 2: Instantaneous temperature contours from simulations using the LESLIE3D code that utilize a MUSCL scheme (top) and a central scheme (bottom).
Dow
nloa
ded
by A
ir Fo
rce
Rese
arch
Lab
orat
ory
(DET
-7 A
FRL/
RQO
I) - E
dwar
ds A
FB o
n N
ovem
ber 1
4, 2
014
| http
://ar
c.ai
aa.o
rg |
DO
I: 10
.251
4/6.
2014
-082
6 Premixed flame:
33 Distribution A: Approved for Public Release; Distribution Unlimited
Distribution A – Approved for public release; Distribution Unlimited
Adaptive Physics
• Combustion calculations are extremely expensive – Detailed combustion kinetics
• Entails large numbers of species and reaction steps – Turbulent combustion closures
• Linear Eddy Model (LEM) involves sub-grid solutions • The “Silver Lining“
– Detailed chemistry and closures only needed locally – Most of the flowfield has unburnt or burned propellants
• Adaptive physics approach needs to be derived – Apply detailed models only in specific blocks – Block-based solver structure is ideally suited to
adaptive physics implementation
34
Distribution A – Approved for public release; Distribution Unlimited
Rocket'Code' SoYware&Integra4on&Tes4ng&Valida4on&Applica4ons&
SPACE Program
35
CREATE,AV''Framework'
Meshing&Domain&Decomposi4on&Parallel&Processing&GUI&
CFD''Solvers'
Cartesian&Solver&
Strand&Solver&
CombusFon''Physics'
Equa4on&of&state&Turbulence&Combus4on&Turbulent&combus4on&
GPU'Mul4&CPU/GPU&&Accelera4on&
Version 1: Mixing & combustion Version 2: Combustion stability Version 3: Thermal management Version 4: Ignition
‹#›
Modular Vision
CFD Engine
Turbulence Models
Multiphase Models
Combustion Kinetics
Equation of State
Turbulent Combustion
Partners: • AFRL (East & West), HPCMO-CREATE, AEDC, Eglin • NASA – MSFC, GRC • DOE – Sandia/CRF, Oakridge • Academia – Georgia Tech, Purdue, UCLA • Industry – Aerojet-Rocketdyne, SBIR & STTR
• Redlich-Kwong • Peng-Robinson • REFPROPS
• RANS • DDES • LES/Smagorinsky
• Primary Atomization • Kelvin-Helmholtz • Taylor-Analogy Breakup
• Global mechanisms • Reduced mechanisms • Flamelet libraries
• Linear Eddy Model • Flamelets • FMDF
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Road Ahead • Develop in-house code/modeling expertise
– In-house use of codes and models developed externally
• Focus on core rocket physics expertise – State-of-art physics sub-models and numerics for:
• High pressure EOS (Equation of State), LES sub-grid models, Combustion Kinetics, Multiphase, Turbulent combustion, Structures
• Modular physics with well-defined interfaces – Build infrastructure with core-CFD algorithms
– Build partnerships for sub-model development
• Variable Level of Fidelity – Vision for model hierarchy from high fidelity physics-based models to
lower-fidelity engineering models
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Collaboration
• In-house Activities – Modeling and Simulations Forum
– Coordination of M&S and diagnostics research
• RQ Interactions – CFD VTC’s held periodically
– Working groups in common interest areas – eg., turbulent combustion
• AFOSR Coordination – Rocket propulsion, turbulent combustion, flow control, plasmas, materials,
propellants, computational math
• External – Collaboration with HPCMO/CREATE, NASA, Sandia, universities, industry,
small businesses
Distribution A: Approved for Public Release; Distribution Unlimited
‹#›
Opportunities
• Multi-scale modeling of turbulent combustion – Compressible turbulence and reaction kinetics
• Emphasis on emerging computing architectures – Focus on GPUs
• Multi-fidelity hierarchy for design tool development – Reduced order model development
• Optimization framework for design & analysis – Including error estimation & uncertainty quantification
• Data analysis and processing – Advancing test science
Distribution A: Approved for Public Release; Distribution Unlimited