Skip to main content

Full text of "NASA Technical Reports Server (NTRS) 19960047100: Optimization of Air-Breathing Engine Concept"

See other formats


NASA Technical Memorandum 107303 


Optimization of Air-Breathing Engine Concept 


Surya N. Patnaik 
Ohio Aerospace Institute 
Cleveland, Ohio 

Thomas M. Lavelle and Dale A. Hopkins 
Lewis Research Center 
Cleveland, Ohio 


Prepared for the 

Computational Aerosciences Workshop 
sponsored by NASA Headquarters 
Ames Research Center, August 13-15, 1996 



National Aeronautics and 
Space Administration 


Optimization of Air-Breathing Propulsion Engine Concept 

SuryaN. Patnaik 

Ohio Aerospace Institute, Cleveland, Ohio 44142 
spatnaik@lerc.nasa.gov, (216) 962-3135 

Thomas M. Lavelle and Dale A. Hopkins 
NASA Lewis Research Center, 21000 Brookpark Road, Cleveland, Ohio 44135 
tlavelle@lerc.nasa.gov, (216) 977-7042; dhopkins@lerc.nasa.gov, (216) 433-3260 

Summary 

The design optimization of air-breathing propulsion engine concepts has been accomplished by 
soft-coupling the NASA Engine Performance Program (NEPP) analyzer with the NASA Lewis 
multidisciplinary optimization tool COMETBOARDS. Engine problems, with their associated 
design variables and constraints, were cast as nonlinear optimization problems with thrust as the 
merit function. Because of the large number of mission points in the flight envelope, the diversity 
of constraint types, and the overall distortion of the design space; the most reliable optimization 
algorithm available in COMETBOARDS, when used by itself, could not produce satisfactory, 
feasible, optimum solutions. However, COMETBOARDS’ unique features — which include a 
cascade strategy, variable and constraint formulations, and scaling devised especially for difficult 
multidisciplinary applications — successfully optimized the performance of subsonic and supersonic 
engine concepts. Even when started from different design points, the combined COMETBOARDS 
and NEPP results converged to the same global optimum solution. This reliable and robust design 
tool eliminates manual intervention in the design of air-breathing propulsion engines and eases the 
cycle analysis procedures. It is also much easier to use than other codes, which is an added benefit. 
This paper describes COMETBOARDS and its cascade strategy and illustrates the capabilities of 
the combined design tool through the optimization of a high-bypass-turbofan wave-rotor-topped 
subsonic engine and a mixed-flow-turbofan supersonic engine. 

Introduction 

The NASA Engine Performance Program (NEPP) can be used for the analysis and preliminary 
design of subsonic and supersonic air-breathing propulsion engine concepts. NEPP can evaluate the 
performance of an engine over its flight envelope for various mission points, which are defined by 
different Mach number, altitude, and power-setting combinations. It also can optimize engine 
parameters at specified mission points. However, NEPP can experience difficulties with optimiza- 
tion, producing infeasible suboptimal solutions that require manual redesign. In an effort to 
eliminate the optimization deficiency of the NEPP code and improve its reliability, we combined 
NEPP with COMETBOARDS (Comparative Evaluation Test Bed of Optimization and Analysis 
Routine for the Design of Structures). This combined tool has successfully optimized a number of 
subsonic and supersonic engines. Some of COMETBOARDS’ key features and unique strengths 
that assisted in optimizing the engines include a cascade optimization strategy, constraint and 
design formulations, and a global scaling strategy. This paper presents a brief introduction to the 
COMETBOARDS design tool and the NEPP analyzer. The design optimization capability of the 
combined tool is illustrated by considering a subsonic wave rotor topped engine and a mixed-flow- 
turbofan supersonic engine as examples. 



COMETBOARDS Test Bed 

The multidisciplinary design optimization test bed, COMETBOARDS, which is used in the design 
of air-breathing propulsion engines, has the modular organization depicted in figure 1 . Some key 
features of the test bed are multidisciplinary optimization (with separate objective, constraints, and 
variables for each discipline), substructure optimization in sequential and parallel computational 
platforms, and state-of-the-art optimization algorithms. An analysis approximation by means of 
linear regression analysis and neural networks is being added. The COMETBOARDS system first 
formulates the design as a nonlinear mathematical programming problem, and then it solves the 
resulting problem. The problem can be formulated (variables, constraints, objective, etc.) by the 
analysis tools available in the “Analyzers” module reading specified data in the “Data files” 
module. A number of analysis tools (RPK/NASTRAN (ref. 1) for structural analysis, NEPP (ref. 2) 
for air-breathing engine performance analysis, FLOPS (ref. 3) for aircraft flight optimization 
analysis, etc.) are available in COMETBOARDS, and provision exists for the soft-coupling and 
quick integration of new analysis tools. The NEPP and FLOPS analyses are interfaced to 
COMETBOARDS through system calls. The COMETBOARDS solution technique exploits 
several of the unique strengths that are available in its “Optimizers” module, such as a cascade 
optimization strategy, the formulation of design variables and constraints, and a global scaling 
strategy. COMETBOARDS, which is written in the FORTRAN 77 language, is currently available 
for Cray and Convex computers and Iris and Sun workstations. Successful COMETBOARDS 
solutions for a number of diverse industrial problems (such as components of the space station, the 
rear divergent flap of a downstream mixing nozzle for a High Speed Civil Transport (HSCT) 
engine, system optimization for subsonic and supersonic aircraft, thrust optimization for 
multimission HSCT mixed-flow-turbofan engines, and optimization of a wave-rotor concept in 
propulsion engines) illustrate its versatility and robustness. 

Cascade Optimization Strategy 

COMETBOARDS can solve difficult optimization problems by using the cascade strategy depicted 
in figure 2. This strategy uses more than one optimizer to solve a complex problem when individual 
optimizers face difficulties. With COMETBOARDS, users have considerable flexibility in develop- 
ing cascade strategies: selections can be made from a number of optimizers, their convergence 
criteria, analysis approximations, and the amount of random perturbations between optimizers. 
Consider, for example, a four-optimizer cascade (one optimizer followed by three other optimizers) 
that was used to successfully solve a subsonic aircraft problem. For such a cascade, individual con- 
vergence criteria can be specified for each optimizer. For example, a coarse stop criterion may be 
sufficient for the first optimizer, whereas a fine stop criterion may be necessary for the last 
optimizer. Likewise, an approximate analysis may suffice for the first optimizer, although an 
accurate analysis can be reserved for the final optimizer. The amount of pseudorandom perturbation 
for design variables may be specified between the optimizers at the discretion of users. A more 
indepth description of COMETBOARDS can be found in references 4 to 6. 

NASA Engine Performance Program (NEPP) 

The NEPP engine simulation computer code performs zero-dimensional, steady-state, thermo- 
dynamic analysis of turbine engine cycles. By using a flexible method of input, a set of standard 
components are connected at execution time to simulate almost any turbine engine configuration 
that the user may contemplate. Off-design performance is calculated through the use of component 


2 



performance maps. The compressor and turbine performance maps are scaled by the code to match 
the design point pressure ratio, corrected weight flow, and efficiency of the engine being modeled. 
The default thermodynamic routine used in the code is preset for a mixture of air and JP4 fuel. A 
chemical equilibrium model is incorporated as an option to adequately predict thermodynamic 
properties when chemical dissociation occurs as well as when virtually any fuel is used. To 
determine the performance of an engine over a flight envelope, the user will define many different 
operating conditions representing different Mach number, altitude, and power setting combinations. 
Each one of these points represents a separate analysis problem. Often when a cycle is being 
studied there are several values that can be varied to give best engine performance. For example, an 
engine design may have a variable geometry fan that allows fan rotor blade angles to be set to give 
the best fuel consumption subject to certain performance constraints such as fen surge margin. 
Thus, when creating a simulation of this cycle, an optimization scheme is needed to determine the 
"best" fan rotor blade angles for a given engine operating condition. NEPP currently uses Powell's 
conjugate direction method for optimization, but experience in using this algorithm with NEPP has 
shown it to be lacking. Often the results are not the optimum values and require further fine-tuning 
by the engineer. One common problem is that the optimizer fails to push the design hard up against 
a constraint, even when doing so would improve the results. Combining COMETBOARDS and 
NEPP is an attempt to compensate for NEPP’s deficiency. 

Design of a Wave-Rotor-Topped Engine 

Conceptually, a wave rotor replaces a burner in conventional air-breathing engines. The wave-rotor 
topping can lead to higher specific power in the engine, or to more thrust for less fuel consumption. 
Design op timizat ion was carried out for a high-bypass-ratio-turbofen wave-rotor-enhanced 
subsonic engine with four ports (the burner inlet, burner exhaust, compressor inlet, and turbine 
exhaust ports). Figure 3 depicts the 47 mission points. NEPP generated the engine performance 
analysis and the constraint and objective formulations, whereas COMETBOARDS optimized the 
design. To examine the benefits that accrued from the wave-rotor enhancement, we designed the 
engine under the assumption that most of the baseline variables and constraints were passive and 
that the im portant parameters directly associated with the wave rotor were active. The active 
variables considered were the rotational speed of the rotor and the heat added to it. Important active 
constraints included limits on the maximum speeds of all compressors, a 1 5 -percent surge margin 
for all compressors, and a maximum wave-rotor exit temperature. The engine thrust was selected as 
the merit function. The wave-rotor-engine design became a sequence of 47 optimization 
subproblems (one for each mission point). Only by using the cascade strategy could the problem be 
solved successfully for the entire flight envelope. Figure 4 shows the convergence of the two- 
optimizer cascade strategy for the mission point defined by Mach = 0.1 and altitude = 5000 ft. The 
first optimizer produced an infeasible design at 67 061 -lb thrust in about five design iterations. The 
second optimizer, starting from the first solution with a small perturbation, produced a feasible 
optimum design with a thrust of 66 901 lb. For these 47 mission points, figure 5 shows the 
optimum solutions obtained with the combined tool and normalized with respect to the NEPP 
results. This figure depicts the benefits of optimizing wave-rotor design with the combined 
COMETBOARDS-NEPP design tool. Figure 5 shows that the combined tool produced a design 
with a higher thrust over all 47 mission points than did NEPP, with maximum increases around 
mission points 12, 26, and 32. Both NEPP and COMETBOARDS— NEPP produced identical 
optimum thrust values for a few mission points; however, the maximum difference in thrust 
exceeded 5 percent for several mission points. These differences could be significant if the design 


3 



points with increased thrust were used to size the engine. The combined COMETBOARDS-NEPP 
tool successfully solved the subsonic wave-rotor-engine design optimization problem. 

Mixed-Flow-Turbofan Supersonic Engine for High Speed Civil Transport System 

Optimization of a 122-mission-point mixed-flow-turbofan supersonic engine also was attempted 
with the COMETBOARDS-NEPP combined tool. This optimization required the solution of a 
sequence of 122 optimization subproblems (again, one for each mission point). For each sub- 
problem, the thrust of the engine was considered as die merit function. The important active design 
variables considered were engine bypass ratio, fan operating point determined by fan speed, and 
surge mar gin. The important constraints considered were maximum speed on all compressors, 
acceptable surge margin for all compressors, compressor discharge temperatures, and maximum 
mixer corrected flow. Because of the sequence of a large number of optimization subproblems, the 
diverse constraint types, and the overall ill conditioning of the design space, the most reliable 
individual optimization algorithm available in COMETBOARDS could provide feasible results for 
only a portion of the 122-mission-point flight envelope. A four-optimizer cascade strategy could 
successfully solve the engine design problem for the entire 1 22-mission-point flight envelope. 
Furthermore, calculations for the cascade strategy converged to die same global solution when 
begun from different design points. The cascade solution was normalized with respect to the NEPP 
results, which were obtained by using an individual optimizer. This normalized solution, which is 
shown in figure 6, was found to be superior for most of the 122 mission points, except for a few 
cases for which both the COMETBOARDS and NEPP optimum results agreed. For flight around 
mission point 70, optimum thrust was about 10 percent higher for COMETBOARDS than for 
NEPP. COMETBOARDS successfully solved die 122-mission-point, mixed-flow-turbofan engine 
design problem. 


Summary 

The COMETBOARDS design tool, when augmented with the NEPP analyzer for air-breathing 
propulsion engines, successfully solved a number of subsonic and supersonic engine design 
problems. COMETBOARDS’ advanced features and unique strengths made engine design 
problems easier to solve. Its cascade optimization strategy was especially helpful in generating 
feasible optimum solutions when an individual optimizer encountered difficulty. Calculations for 
the cascade strategy converged to the same optimum design even when they started from different 
initial design points. For most mission points, the combined tool increased the value of the 
optimum thrust by a few percentage points. Such improvements can become critical especially 
when engines are sized for such mission points. The research-level software COMETBOARDS, 
with some enhancements and modifications, can be used by the aircraft industry. 

References 

1. RPK/NASTRAN. COSMIC. University of Georgia, Athens, GA, 1994. 

2. Klann, J.N.; and Snyder C.A.: NEPP Programmers Manual. NASA TM-106575, 1994. 

3. McCullers, L.A.: FLOPS: Aircraft Configuration Optimization. NASA CP-2327, 1984. 

4. Guptill, J.D., et al.: COMETBOARDS Users Manual, NASA TM-4537, 1996. 

5. Patnaik, S.N., et al.: Comparative Evaluation of Different Optimization Algorithms for Structural 

Design Applications, Int. J. Num. Meth. Eng., vol. 39, 1996, pp. 1761-1774. 

6. Patnaik, S.N.; Gendy A.S.; and Hopkins D.A.: Design Optimization of Large Structural Systems 

With Substructuring in a Parallel Computational Environment. Comp. Systems Eng., vol. 5, 
no. 4-6, 1 994, pp. 425-440. 


4 



Optimizers 


Approximations 
Neural nets 
v Regression > 


Data files 


Engine 

optimization 


Analyzers 




Number of design iterations 


Fig. 2. — Cascade solution for a subsonic aircraft. 


5 







Normalized optimum thrust Altitude, 


4x1 0 4 



0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 
Mach number 

Fig. 3. — Forty-seven mission points for high- 
bypass-ratio-turbofan wave-rotor-topped 
engine. 


COMETBOARDS solution 

NEPP solution 


1.06 
1.04 
1.02 
1.00 
0.98 

0 10 20 30 40 4750 

Mission points 

Fig. 5. — Value-added benefit in design of a 47-mission- 
point, high-bypass-turbofan subsonic engine using a 
wave rotor. 



6 


Normalized optimum thrust 


7.75x104 



0 2 4 6 8 10 12 


Number of design iterations 

Fig. 4. — Cascade solution for a wave-rotor-topped 
subsonic engine. 



Figure 6. — Value-added benefits in the design of 
a mixed-flow turbofan engine. 



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 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, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503. 


1 . AGENCY USE ONLY ( Leave blank) 


2. REPORT DATE 

August 1996 


3. REPORT TYPE AND DATES COVERED 

Technical Memorandum 


4. TITLE AND SUBTITLE 


Optimization of Air-Breathing Engine Concept 


6. AUTHOR(S) 


Surya N. Patnaik, Thomas M. Lavelle, and Dale A. Hopkins 


5. FUNDING NUMBERS 


WU-505-63-53 


7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 

National Aeronautics and Space Administration 
Lewis Research Center 
Cleveland, Ohio 44135-3191 


8. PERFORMING ORGANIZATION 
REPORT NUMBER 


E- 10227-3 


9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 

National Aeronautics and Space Administration 
Washington, D.C. 20546-0001 


10. SPONSORING/MONITORING 
AGENCY REPORT NUMBER 


NASA TM- 107303 


11. SUPPLEMENTARY NOTES 


Prepared for the Computational Aerosciences Workshop sponsored by NASA Headquarters, Ames Research Center, 
August 13-15, 1996. Surya N. Patnaik, Ohio Aerospace Institute, 22800 Cedar Point Road, Cleveland, Ohio 44142; 
Thomas M. Lavelle and Dale A. Hopkins, NASA Lewis Research Center. Responsible person, Thomas M. Lavelle, 
organization code 2430, (216) 977-7042. 


12a. DISTRIBUTION/A VAILABILITY STATEMENT 

Unclassified - Unlimited 
Subject Category 39 

This publication is available from the NASA Center for AeroSpace Information, (301) 621-0390.| 


12b. DISTRIBUTION CODE 


13. ABSTRACT (Maximum 200 words) 


The design optimization of air-breathing propulsion engine concepts has been accomplished by soft-coupling the 
NASA Engine Performance Program (NEPP) analyzer with the NASA Lewis multidisciplinary optimization tool 
COMETBOARDS. Engine problems, with their associated design variables and constraints, were cast as nonlinear 
optimization problems with thrust as the merit function. Because of the large number of mission points in the flight 
envelope, the diversity of constraint types, and the overall distortion of the design space; the most reliable optimization 
algorithm available in COMETBOARDS, when used by itself, could not produce satisfactory, feasible, optimum 
solutions. However, COMETBOARDS’ unique features — which include a cascade strategy, variable and constraint 
formulations, and scaling devised especially for difficult multidisciplinary applications — successfully optimized the 
performance of subsonic and supersonic engine concepts. Even when started from different design points, the combined 
COMETBOARDS and NEPP results converged to the same global optimum solution. This reliable and robust design tool 
eliminates manual intervention in the design of air-breathing propulsion engines and eases the cycle analysis procedures. 
It is also much easier to use than other codes, which is an added benefit. This paper describes COMETBOARDS and its 
cascade strategy and illustrates the capabilities of the combined design tool through the optimization of a high-bypass- 
turbofan wave-rotor-topped subsonic engine and a mixed-flow-turbofan supersonic engine. 


14. SUBJECT TERMS 

Design optimization; Air-breathing engine; Wave rotor; Multiflow turbofan engine 

15. NUMBER OF PAGES 

08 

16. PRICE CODE 

A02 

17. SECURITY CLASSIFICATION 

18. SECURDY CLASSIFICATION 

19. SECURITY CLASSIFICATION 

20. LIMITATION OF ABSTRACT 

OF REPORT 

OF THIS PAGE 

OF ABSTRACT 


Unclassified 

Unclassified 

Unclassified 



NSN 7540-01-280-5500 


Standard Form 298 (Rev. 2-89) 
Prescribed by ANSI Std. Z39-18 
298-102