Table Of Content

Report Documentation Page Form Approved OMB No. 0704-0188 Public reporting burden for the 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 the 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 Washington Headquarters Services, Directorate for Information Operations and Reports, 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 a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. 1. REPORT DATE 3. DATES COVERED 2002 2. REPORT TYPE 00-00-2002 to 00-00-2002 4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER NAVO MSRC Navigator. Spring 2002 5b. GRANT NUMBER 5c. PROGRAM ELEMENT NUMBER 6. AUTHOR(S) 5d. PROJECT NUMBER 5e. TASK NUMBER 5f. WORK UNIT NUMBER 7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION Naval Oceanographic Office (NAVO),Major Shared Resource Center REPORT NUMBER (MSRC),1002 Balch Boulevard,Stennis Space Center,MS,39522 9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S) 11. SPONSOR/MONITOR’S REPORT NUMBER(S) 12. DISTRIBUTION/AVAILABILITY STATEMENT Approved for public release; distribution unlimited 13. SUPPLEMENTARY NOTES 14. ABSTRACT 15. SUBJECT TERMS 16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF ABSTRACT OF PAGES RESPONSIBLE PERSON a. REPORT b. ABSTRACT c. THIS PAGE Same as 28 unclassified unclassified unclassified Report (SAR) Standard Form 298 (Rev. 8-98) Prescribed by ANSI Std Z39-18 The Director’s Corner Steve Adamec, NAVO MSRC Director Sweeping Series of Enhancements Engulf NAVO MSRC The next few months will be notable ones here at ically important for us to redouble our efforts in the NAVO MSRC. We are completing a sweeping assessing and implementing common user envi- series of center enhancements, designated as ronments, practices, and tools within and across Technology Insertion for FY02 (TI-02), across the SRCs. Your continuing individual and collec- several major technology areas within the MSRC. tive user feedback makes it clear that you consid- The centerpieces of TI-02 are the acquisition of a er this to be one of your highest priorities for the new IBM POWER4 HPC system and completion SRCs. In response, the SRCs have intensified and of the NAVO MSRC Remote Storage Facility made significant progress toward strategic cross- (RSF). When complete, the TI-02 enhancements cutting collaborative efforts in technical areas will provide almost 8.5 teraflops of aggregate such as mass storage and archival, metacomput- peak computing capability with commensurately ing, HPCMP-wide shared information environ- balanced storage and networking capabilities. ments, and security. Here at the NAVO MSRC, The RSF will permit us to offer what is perhaps we are redoubling our efforts to supplement one of the most resilient and best performing those activities by strengthening the links to the HPC storage environments in the world today. new HPCMP Programming Environment and This enormous computational capability, coupled Training (PET) program that's more tightly with a sustained 11-year NAVOCEANO focus on focused than ever on user environment, tools, supporting the largest and most demanding DoD and productivity. computational applications, will continue to My staff and I look forward to seeing you in June enable unparalleled advances in the DoD science at the 2002 HPCMP Users' Conference in Austin, and technology areas served by the HPC Texas. As always, please take every opportunity Modernization Program (HPCMP). to let us know how we can better serve you — With all of this diverse computational capability your feedback is critically important to us and to that's been fielded across 20+ shared resource the HPCMP. centers (SRCs) by the HPCMP, it has become crit- About the Cover: This image was generated by the POP model running on HABU, an IBM RS/6000 SP located at the NAVO MSRC. The simulation represents a state-of-the-art eddy-resolving effort that will eventually be used as the ocean component of a coupled global air/ocean/ice prediction system for Navy needs as well as for short-term climate studies. 2 SPRING 2002 NAVO MSRC NAVIGATOR The Naval Oceanographic Office (NAVO) Contents Major Shared Resource Center (MSRC): Delivering Science to the Warfighter The NAVO MSRC provides Department of The Director’s Corner Defense (DoD) scientists and engineers with high performance computing (HPC) resources, including leading edge computational systems, large- scale data storage and archiving, scientific visuali- 2 Sweeping Series of Enhancements Engulf zation resources and training, and expertise in spe- NAVO MSRC cific computational technology areas (CTAs). These CTAs include Computational Fluid Dynamics (CFD), Climate/Weather/Ocean Feature Articles Modeling and Simulation (CWO), Environmental Quality Modeling and Simulation (EQM), Computational Electromagnetics and Acoustics (CEA), and Signal/Image Processing (SIP). 4 Active Swirl and Fuel Modulation to Control Combustion Instability in Gas Turbine Engines NAVO MSRC Code N7 7 NAVO MSRC Visualization Center Supports 1002 Balch Boulevard the Retrieval of the Ehime Maru Stennis Space Center, MS 39522 1-800-993-7677 or [email protected] 11 Knowledge Management: A Case for Intelligent Data Archives Scientific Visualization 13 Improvements in the Use of Color Wheel NAVO MSRC Navigator Visualization of Global Modeling Data www.navo.hpc.mil/navigator 14 Towards a High-Resolution Global Coupled NAVO MSRC Navigator is a bi-annual technical publication designed to inform users of the news, Navy Prediction System: The Ocean Component events, people, accomplishments, and activities of the Center. For a free subscription or to make address changes, contact NAVO MSRC at the Programming Environment and Training address above. EDITOR: 18 VSIPL/ERI: An Enhanced Reference Implementation Gioia Furness Petro, [email protected] of the Vector Signal and Image Processing Standard DESIGNERS: 20 NAVO MSRC PET Update Cynthia Millaudon, [email protected] Kerry Townson, [email protected] Lynn Yott, [email protected] The Porthole Any opinions, conclusions, or recommendations in this publication are those of the author(s) and do 22 A Look Inside NAVO MSRC not necessarily reflect those of the Navy or NAVO MSRC. All brand names and product names are trademarks or registered trademarks of their Navigator Tools and Tips respective holders. These names are for information purposes only and do not imply endorsement by the Navy or NAVO MSRC. 24 File Transfer Techniques: Transferring Between HABU and Other Systems Approved for Public Release Upcoming Events Distribution Unlimited 27 Conference Listings NAVO MSRC NAVIGATOR SPRING 2002 3 Active Swirl and Fuel Modulation to Control Combustion Instability in Gas Turbine Engines Chris Stone and Suresh Menon, Georgia Institute of Technology Combustion-generated pollutants such control techniques as nitrous oxide (N O), carbon monox- (e.g., using struc- 2 ide (CO), unburned hydrocarbons tural changes) (UHC), and soot can be reduced signif- cannot deal with icantly if lean burning combustion sys- all possible tems can be used in gas turbine changes, and engines. However, as the fuel-air mix- therefore, in order ture becomes lean, small perturbations to suppress these in the flow can change the unsteady oscillations, active heat release pattern, and if the heat control strategies release is in-phase with the acoustic are needed. This fluctuations in the combustor, the pres- article describes the sure (acoustic) fluctuations can grow study of two active con- into high-amplitude, low-frequency trol methods applied to pre- oscillations. These oscillations can mixed combustion in a gas tur- cause flame extinction (often called bine engine: modulation of the Lean Blowout, or LBO), reduce engine swirl imposed on the incoming fuel- Figure 1(b) life span, and/or cause catastrophic air mixture (which is at a fixed equiva- structural damage under extreme cir- lence ratio) and modulation of the fuel upstream of the combustor. Fuel is then cumstances. If these oscillations can be content in the premixed mixture for a injected into this swirling airflow in the controlled and even leaner mixtures fixed swirl condition. These studies premixer so that a swirling, premixed can be burned stably, then not only can have been conducted using a state-of- fuel-air mixture enters the combustor. emissions be reduced but the over fuel the-art Large-Eddy Simulations (LES) Swirling inflow is used in dump com- consumption can also be decreased sig- model developed at the Georgia bustors to exploit a fluid dynamic phe- nificantly. Since these oscillations occur Institute of Technology. nomenon called Vortex Breakdown in a dynamic manner, passive The first active control technique stud- (VB) that makes the flame zone com- ied here employs modulation of the pact and stabilizes the flame. As the inlet swirl velocity of the incoming swirling flow expands into the combus- premixed mixture. In a typical tor, an adverse pressure gradient is gas turbine engine, swirl is formed in the flow, and this slows down introduced into the air the axial motion of the mixture. For a from the compressor given geometry, when the swirl intensi- by swirl vanes (typical- ty exceeds a critical value, a recircula- ly small airfoils at high tion bubble is formed (called VB) along angles of attack) in the the centerline of the combustor. This premixer located bubble acts as an effective bluff body in Figures 1(a)/(b).Swirl modulation effect on flame-vortex dynamics:(a) high-swirl condition,(b) low-swirl condition. At the high-swirl state,the vortex rings (purple) rapidly break down,and the flame surface (gray) is compact.On the other hand,at low swirl,the vortex rings are more coherent,and they drag the flame with them,making the Figure 1(a) flame longer.No VB occurs when the swirl is low,and therefore,the flame in the low-swirl case oscillates and is unstable,whereas at high swirl,VB stabilizes the flame. 4 SPRING 2002 NAVO MSRC NAVIGATOR the flow, and along the centerline, the shared-memory parallel platforms (SGI an increase in swirl stabilizes the flame flow from the inlet actually stops (stag- Origin 2000/3800). In fact, due to its and reduces the fluctuation in both the nates) just upstream of the bubble. This scalability, LESLIE3D is now a bench- inflow mass flow rate and in the pres- phenomenon helps to stabilize the mark code for all Department of sure fluctuations. flame just upstream of the stagnation Defense (DoD) High Performance The governing physics of combustion point. However, in current gas turbine Computing (HPC) sites and is being instabilities is quite complicated owing engines, since the swirl vanes are fixed, used to evaluate new systems’ per- to the fact that not one, but several formances. The current study the swirl introduced into the flow is stat- physical processes are coupled and in employed 1.3 million grid cells and ic and is typically optimized for only contention. The dominant process for required approximately 1000 CPU one operating condition. This study combustion instabilities is the interac- hours (on HABU) for one flow-through explores how swirl magnitude can be tion of acoustic waves (pressure fluctua- time. Typically, 10-15 flow-through actively changed to stabilize the com- tions) with the unsteady heat source (in times are simulated to obtain sufficient bustion as the fuel air mixture is this case, the flame surface). When the data for statistical analysis. changed from rich to lean. pressure (p') and heat-release fluctua- The first investigation looked at chang- tions (q') are "in-phase" (the product of The second technique studied here is ing the level of swirl imposed at the the two being positive), energy is the dynamic adjustment of the fuel inlet. In an open-loop study conducted added to the instabilities. As the fuel content in the fuel-air mixture. For a earlier, the pressure fluctuation was concentration is decreased, two impor- fixed mass flow rate and a fixed inlet shown to reduce when swirl intensity tant controlling parameters, the com- swirl intensity, we dynamically change the fuel-air mixture ratio by changing the equivalence ratio of the incoming mixture. The goal of this study is to determine how changing the fuel content (which, in turn, will affect the amount of heat released and peak temperature reached) will affect the combustion dynamics in the combustor. The LES methodology is used here to simulate the combustor response to both control techniques. In LES, the large scales of motion are fully resolved in both space and time, and only the small scales (smaller than the grid resolution) are modeled. Premixed combustion is modeled using a thin-flame model in which a scalar field that represents the propagating flame front is Figure 2.Mass flux measured at the inlet during swirl modulation.At low swirl, evolved, along with the LES momen- the fluctuating mass flux is nearly 100% but drops sharply as the swirl vane tum and energy transport. To model angle is increased (i.e.,when the swirl velocity is increased).The pressure fluctuation in the combustor is also decreased as the swirl intensity is increased. the fuel-air mixture change (in the second control technique), an additional (conserved) scalar equation is mod- was increased. To dynamically repro- bustion temperature (flame tempera- eled, which tracks the change in local duce this effect, a simulation was start- ture) and the burning velocity (flame equivalence ratio. The local equiva- ed with a low-swirl S (0.56), and then speed), are subsequently reduced. The lence ratio is then used to calculate the swirl was steadily increased so that S nearer one is to the LBO limit, the burning rate and combustion tempera- reached a final value of 1.12. Analysis more pronounced is this effect (See ture. shows that after a small delay, the pres- Figure 3a). At a lower burning velocity, sure oscillation amplitude dropped by unsteady flames are more susceptible The LES algorithm (LESLIE3D) is par- 6.6 decibels (dB). Figures 1 and 2, to large-scale flow perturbations since allel (using Message Passing Interface) respectively, show the effect of swirl on the recovery time (i.e., the time need- and performs very well on a variety of vortex-flame dynamics and on the ed to return to the mean or stable distributed (Cray T3E (SEYMOUR), mass flow-rate fluctuations. Results RS/6000 SP3 (HABU), etc.) and Article Continues... show that for a fixed fuel-air mixture, NAVO MSRC NAVIGATOR SPRING 2002 5 Figure 3(a) Figure 3(a)/(b).Instantaneous view of fuel modulation control.Shown are the flame surface (red isosur- face),vortex rings (gray rings),product temperature (color contours), and fluctuating pressure signal (shown below the figure).In (a),the fuel mixture entering the combustor has become lean,and this results in a decrease in temperature (blue core region),and the pressure oscillation increases.This time corre- sponds to the far right of the p'(t) signal.In (b),the inflow fuel mixture equivalence ratio is increased (rich mixture),and as a result,the flame temperature increases,resulting in a hotter core region (red temperature contours).At this phase of the control cycle,the pressure fluctuations 5 begin to decrease (first quarter of the p'(t) signal). 0 -5 location) is lengthened. In the pres- flow in a confined domain: acoustic not only to achieve stable combus- ent simulation, the magnitude of waves, vortex motion, and unsteady tion in a regime that is not possible flame fluctuations is reduced at high- heat release due to combustion. otherwise, but also to understand er burning velocities (See Figure 3b). Active control of this instability, as how the dynamics have been Another contributing factor to the demonstrated here, can then be used changed due to control actuation. increased stability is the modification of the relevant time scales. As the Figure 3(b) combustion temperature is altered, the acoustic velocity (speed of sound) is also modified, resulting in a shift in the phase between the pressure and heat-release fluctuations. This effect can be significant in either increasing or decreasing the instabilities, depending on the phase between the p' and q'. In some combustion devices, instabilities are more intense for higher equivalence ratios due to this effect, while, as simulated here, the reverse may also occur. Simulations such as these can be used to understand the complex 5 nature of combustion instability that 0 is a result of nonlinear interactions between the three modes of wave -5 motion in a compressible reacting Acknowledgements This work was supported in part by General Electric Power Systems and Army Research Office (ARO). Computational support was provided by the DoD High Performance Computing Modernization Program Office (HPCMO) at HPC Major Shared Resource Center (MSRC) sites (Naval Oceanographic Office (NAVO), U.S. Army Engineer Research and Development Center (ERDC), and Army Research Laboratory (ARL)), under an ARO Challenge Project. 6 SPRING 2002 NAVO MSRC NAVIGATOR NAVO MSRC Visualization Center Supports the Retrieval of the Ehime Maru Pete Gruzinskas, NAVOMSRC Visualization Center On 9 February 2001, a collision speed and direction, temperature, er resolution (500-m) nest, that between a U.S. Navy submarine and salinity fields. Initially, a 2-km bounded the operation area just and a Japanese fishing trawler sent grid that covered the entire south of Oahu. The model was run the fishing vessel to the bottom of Hawaiian Islands was generated, but on the newest and most powerful the Pacific Ocean about 10 miles this model provided input to a high- supercomputer at the NAVO MSRC, south of Diamond Head, the RS/6000 SP3 (HABU). Oahu, Hawaii. The decision Part of the NAVO MSRC was made to retrieve the ves- Visualization Center mission is sel, which lay in approxi- the development of analytical mately 1800 feet of water. A software environments for the Crisis Action Team was Department of Defense (DoD) formed, and the NAVO research community. Some of MSRC Visualization Center these tools, designed to ana- was asked to provide visuali- lyze ocean model output, zation support for the Ehime were modified to analyze the Maru Retrieval Operation. Navy's operational model out- There are many support put. These software tools pro- aspects in an operation like vided significant diagnostic this, but the first challenge capability, which assisted in was to visualize the high-res- the validation of the model olution model output gener- output in a very complex ated by the Shallow-Water environment. Analysis and Forecast System While virtual environments (SWAFS) a three-dimension- may not provide all of the al (3D) ocean circulation answers to data analysts, the model that produces time- Screen capture of SWAFS probe compared to series images with correct the ADCP current measurements. Article Continues... SWAFS surface currents in the operational area. Vertical profile of SWAFS over the Ehime Maru. NAVO MSRC NAVIGATOR SPRING 2002 7 fact of the matter is the real world ocean model analysis is portability was critical to the provision of an environment is in 3D. Features with- — the ability to run on a variety of analysis environment and data to in the environment have 3D struc- hardware architectures, including forward-deployed personnel on- ture that can be difficult, if not laptop computers. scene at the recovery site. Other impossible, to realize in two-dimen- support products were developed to Portability is accomplished by savvy sional (2D) space. The same tech- render the recovery area in 3D from application of graphics techniques, nology that was built to help high-resolution bathymetry (water as well as smart Input/Output and research and development modelers depth) data to delineate critical memory management. This ability scrutinize their model output areas. Conceptual animations was applied to an opera- were built to demonstrate Shallow-Water Recovery Site Deep-Water Recovery Site tional scenario where the the mechanics of an 21º17.52’N 21º04.85’N speed and direction of extremely difficult recovery 157º56.40’W 157º49.46’W the ocean currents were operation. critical factors. The operation was success- An additional feature ful, and all mission objec- added to the application, tives were accomplished. unique to the Ehime This recovery operation Maru retrieval operation, demonstrates the excellent was the display of an synergy that has developed Acoustic Doppler Current between the operational Profiler (ADCP) buoy. Navy and the DoD The display of the ADCP research infrastructure built data provided analysts a by the High Performance direct comparison Computing Modernization between model output Office (HPCMO), and is a and in-situ current meas- Final Relocation Site stellar example of the urements in near-real 21º1.00’N NAVO MSRC realizing the time. Another significant 158º7.86’W HPCMO maxim of feature of the virtual "Delivering Science to The 3D shaded relief of OPAREA using high-resolution environments built for bathymetry collected by the USNS SUMNER.Imagery Warfighter." shows where the ship was resting,the shallow-water retrieval area,and the final relocation site. 02 days Particle 02 days Particle 22 hours Pick On 22 hours Pick On 59 minutes 59 minutes 0 45 Speed (cm/s) 0 45 Speed (cm/s) Screen capture showing ocean currents at depth Another snapshot of currents at depth.Note the leg- over an exaggerated (3:1) model of the Ehime Maru. end has been "tuned" to accentuate faster currents. 8 SPRING 2002 NAVO MSRC NAVIGATOR Still images from conceptual animation showing retrieval of the Ehime Maruby the vessel Rockwater. Still image from conceptual animation showing the lift pro- cedure.A conceptual animation was generated depicting how the towing harness would be attached to the Ehime Maru,which rested in approximately 1800 feet of water. Still image showing Ehime Maruretrieval over real bathymetry data.The animation was designed to concep- tualize/visualize what the recovery would look like in the context of the real data collected by the USNS SUMNER. NAVO MSRC NAVIGATOR SPRING 2002 9