MDO Investigation of

1

• MDO Investigation of
• Advanced Design Concepts
• Applied to the
• Blended Wing-Body Configuration
• Six Month Progress Review
• NASA Langley Research Center, Grant NAG 1-02024
• July 24, 2002
• Hampton, VA

2
Our Team

• Students
• Vance Dippold, III
• Andy Ko
• Serhat Hosder
• Faculty
• R. Arieli (visiting faculty)
• B. Grossman
• R.T. Haftka (University of Florida)
• W.H. Mason
• J.A. Schetz

Meeting weekly via telecon with web
presentations and archiving
3
Overview and Tasks
The Job Use MDO to integrate advanced
propulsion concepts that address the problems of
emissions and noise Use an advanced concept as
the baseline the BWB

• Task 1 Models for Distributed Propulsion
• Task 2 MDO with Distributed Propulsion and Noise
• Task 3 Noise

4
Key considerations

• Developed an MDO baseline similar to the 1994/96
  BWB
• 1996 report, CCD-2, is publicly available and
  detailed
• Dino Roman, Boeing BWB Team, looking at and
  commenting on our work
• Fuel volume/tank weight issue of Hydrogen fuel
• Structural weight of pressurized cabin
• Noise in conceptual design

5
Presentation Outline

• Introduction - Bill Mason
• Analysis and Modeling of Distributed Propulsion
  and Noise for MDO - Vance Dippold
• Blended Wing Body MDO Methodology with
  Distributed Propulsion and Noise - Andy Ko
• Green Engineering Hydrogen Fuel - Andy Ko
• Future Work- Joe Schetz

6

• Analysis and Modeling ofDistributed Propulsion
  and Noisefor MDO
• Vance Dippold, III
• and
• J.A. Schetz

7
Advantages of Distributed Propulsion

• Improve Propulsive Efficiency by filling in the
  wake behind the entire vehicle (or most of it)
• Reduce Noise
• Elimination of flaps and control surfaces and
  maybe tails with thrust vectoring
• Tailor flowfield
• Small engines produce high frequency range noise
• Can be absorbed by materials
• Dissipate faster
• Improve Safety
• Engine redundancy
• Improve Affordability with small, interchangeable
  engines
• Disadvantages
• Total Engine Weight penalty
• Higher Fuel Consumption

8
Suggested Arrangements of Distributed Propulsion
9
Consider a BWB Case(Boeing 1996 Design)
Small Engine Size Data (dia lt 2.5 ft)(from
Aviation Week 2002 Source Book)
10
Consider an Example withMultiple Engine Sizes

• Can tailor the flowfield depending upon lift and
  drag distributions
• But, multiple engine sizes could diminish
  affordability advantage

11
Possible Engine Arrangements

• One engine size class
• 50 equal-size engines shown
• Engine spacing 2.4 dia

• Two engine size classes
• 12 larger engines, 48 smaller engines shown
• Engine spacing 2.2 dia

12
Simplified Airframe Noise Rating for MDO(Follow
G.M. Lilley Approach)

• Turbulent flow on upper surface near TE is
  critical for clean aircraft
• Noise Intensity
• (Lilley, 2001)
• Note key role of characteristic turbulent
  velocity u0TKE1/2
• Rank aircraft designs by max TKE on upper surface
  at TE

13
Analysis Requirements

• Viscous (Turbulent)
• Need TKE for noise
• Need mixing for propulsive efficiency
• Model engine thrust
• Transonic
• 3D
• Challenging CFD Problem!

14
First Consider 2D Sliceof BWB Flowfield

• Upstream Region
• Use MSES with BL interaction to obtain p(x) and
  inviscid part of inflow to embedded region
• Use Virginia Tech BL codes on web with TKE model
  to obtain BL part of inflow to embedded region
• Embedded Region containing Modeled Engine
• Use ANSYS Flotran
• Turbulent Navier-Stokes (K-epsilon)
• Includes force field models (for engine)
• Finite Element Method

15
Model Effect of Propulsor on Upstream Region

• Major effect of distributed propulsion on
  Upstream Region is to change circulation and lift
• Model this gross effect as a flap at TE
• Test approach by comparing a flap and a bump near
  TE to increase CL
• For the same 10 increase in CL, p(x) and BL are
  the same in the Upstream Region
• Our Approach
• Compute p(x) and BL from airfoil alone
• Add engine in N-S computation
• Compute new lift, and model with flap
• Compute new p(x), BL with flap
• N-S with engine and new BCs
• Iterate

Bump
Flap
16
Planned 3D Approach

• Upstream Region
• FELISA, FLO22, or GASP for 3D inviscid flow
• Strip method for BL
• Embedded Region
• ANSYS Flotran

17
ANSYS FlotranIdealized Engine Test Case
18
Getting StartedA 2D Analysis using 1996 BWB

• Select a wing section
• 62 of span
• Chord 26.2 ft
• Sweep 32 deg
• CL2D 0.69
• Cruise conditions
• Minf 0.85
• 2D Airfoil Section
• SC(2)-0710 Airfoil
• Chord 22.2 ft
• CL 0.78
• M2D 0.72
• Re 35 Million

19
MSES Velocity Vector Field
20
VT JAVA Boundary Layer Applets
21
Engine Cowl Design

• Engine cowl design based on streamlines over
  airfoil alone
• Must correctly design nacelle, for low-speed and
  high-speed flow conditions

22
Cowl Design ConsiderationsSimilar to
Conventional Nacelle Design

• Low-speed thrust augmentation
• Thrust reverser
• High-alpha conditions
• Noise suppression

23
Roadmap

• Conduct analyses for modest number of cases
• Simplified ducts
• Distributed actuator volumes for engines
• Selected spanwise locations
• Collect key output data
• Max TKE on TE(s)
• Propulsive Efficiency
• Lift and lift distribution
• Drag and drag distribution
• Moments
• Construct response surfaces for MDO purposes

24
Blended Wing BodyMDO methodology with
distributed propulsion and noise

• By Andy Ko
• July 24, 2002

25
Overview

• Goal MDO program for distributed propulsion and
  noise
• Previous MAD Center experience
• Using ModelCenter
• Program Flowchart
• MDO problem statement
• Methods
• Validation

Source Revolutionary Aerospace Systems Concepts
Quiet Green Transport Study
Source Revolutionary Aerospace Systems Concepts
Quiet Green Transport Study
26
Previous Experience

• BWB design methodology builds on Virginia Techs
  HSCT and SBW research
• BWB code similar to SBW code, but architecture is
  ModelCenter based
• Variable fidelity modeling
• Response surface methods
• Use analysis methods already developed
• Programming practices
• Flexibility
• Different objective functions
• Different mission parameters
• Modular approach
• Version control

27
ModelCenter

• What is Model Center?
• It is an integration software written by Phoenix
  Integration
• It is a visual environment
• Allows the user to perform design studies
  including parametric studies, optimization and
  response surfaces.
• Optimization capability using DOT by Vanderplaats
  RD
• Response surface capability using Response
  Surface Toolkit by Adoptech
• Cross-platform environment
• Can be used on PC or UNIX machines
• Individual programs running on different machines
  and different platforms can be integrated.
• Reduced integration and debug time
• Support from Phoenix Integration readily available

28
Program Flowchart
Legend
Component completed
Component not completed or Integrated
ModelCenter Functions
29
MDO problem statement

• Objective function
• Different objective functions can be used
• Takeoff Gross Weight
• Noise
• Cost (if module available)
• Combination of objective functions
• Objective functions can also be specified as
  constraints
• Design variables
• A total of 21 design variables available for
  preliminary code
• Chord (5)
• t/c (5)
• Sweep (4)
• Spanwise station position (3)
• Wing span
• design variables to be added
• Number of engines and distribution
• Parameters
• Various parameters to allow for different
  configurations include
• Number of passengers

• Fuel weight
• Average cruise altitude
• Thrust

30
MDO problem statement

• Constraints in place
• Fuel volume
• Balanced Field Length
• Landing distance
• Second segment climb gradient
• Missed approach climb gradient
• Approach velocity
• Cabin planform area
• Cabin height
• Top of climb
• Constraints to be added
• Balance
• Stability Control
• Noise (can also be specified as an objective
  function)
• Fuel volume for hydrogen fuel

Source http//www.geocities.com/witewings/bwb/gal
lery.html
31
Methods

• Wing weight
• Wing weight equation obtained from Beltramo et.
  al (NASA CR-151970)
• Empirical equation is from current aircraft data
• Future work will involve a better structural
  model for the cabin and wing
• Other weights
• Calculations based on 1994 BWB report by Liebeck
  et. al (NASA CR-4624)
• Propulsion (preliminary modeling)
• Engine thrust and weights obtained from
  Isikverens Ph.D. dissertation
• SFC model obtained from a GE90-like engine deck
• Future work to include distributed propulsion

32
Methods

• Aerodynamics
• Induced drag
• Calculated using a discrete vortex method which
  uses a Trefftz plane analysis to determine the
  load distribution
• Friction drag
• Configuration is divided into a number of
  spanwise strips with the transition location
  estimated based on the Reynolds number and sweep.
• Wave drag
• Approximated with the Korn equation
• Off-line CFD calculations will be done later and
  implemented using response surface methods

33
Methods

• Cruise performance
• Weight fractions for startup, takeoff and climb
• Climb calculation will be improved
• Breguet range equation for cruise segment
• No allowance for descent and landing segments
• Rate of climb at initial cruise altitude (Top Of
  Climb) calculated

34
Initial Validation Mission

• 800 Passengers
• Based on 1994 report mission

35
Initial validation results
Approximated values
36
Initial Validation Results
1994 BWB design
Current VT BWB design
37
Status

• Essential framework in place
• More work needs to be done
• Additional validation
• Important constraints need to be added
• Balance
• Stability Control
• Noise (as developed)
• Higher fidelity methods need to be used
• Wave drag
• Structures wing cabin weight
• Fuel volume for hydrogen fuel
• MDO of the distributed propulsion concept will be
  included as developed

38
Green EngineeringHydrogen fuel

• By Andy Ko
• July 24, 2002

39
Overview

• Fuel tank design and data
• Large fuel volume required Major challenge
• Most hydrogen fuel aircraft place fuel tanks in
  the fuselage
• Wings tanks not efficient for pressurization and
  heat transfer
• Hydrogen engine data
• Liquid hydrogen (LH2) fuel tank placement
  possibilities in the BWB
• Implications of using hydrogen
• Implications to distributed propulsion

40
Fuel tanks

• Almost all LH2 aircraft designs place the fuel
  tanks in the fuselage
• However, we do not have a fuselage!
• Only aircraft design found with wing LH2 tanks
  designed is for a high altitude, high endurance
  aircraft
• Integral and non-integral wing fuel tanks designed

Source Brewer, G.D., Hydrogen Aircraft
Technology, CRC Press, 1990
Source Brewer, G.D., Hydrogen Aircraft
Technology, CRC Press, 1990
Nonintegral, pillow tank design
Integral, conformal tank design
41
Hydrogen engine

• Obtained engine data designed by Brewer, D.G.
• Provides dimensions, weights and engine deck
• Scaling equations are also provided
• Comparison of GE90 engine and hydrogen engine

42
BWB LH2 tank placement

• LH2 fuel tanks always in the outboard wing
  sections
• 3 choices for inboard section
• Option 1 LH2 tank replaces upper passenger deck
  
• Option 2 Large LH2 tank in the center section
  beside passenger decks
• Option 3 Smaller LH2 tank in the center section.
  Remaining volume used for passengers
• External tanks?

LH2 fuel tank on upper deck Passenger cabin on
lower deck
LH2 fuel tank
LH2 fuel tank
Passenger cabin
Passenger cabin
LH2 fuel tanks
LH2 fuel tanks
LH2 fuel tanks
Option 1
Option 2
Option 3
Schematic diagram only, fuel tanks are not sized.
43
Implications of using hydrogen

• According to Carline A.K. (1975)
• LH2 fueled aircraft requires 3.7 times the volume
  of JP fueled airplanes
• Takes into account lower TOGW for LH2 aircraft
• Based on our current BWB design
• Volume of JP fuel used 8000 ft3
• Volume available in the outboard wings 17200
  ft3 (2.15)
• Approximate volume needed if using LH2 29600
  ft3 (3.7)
• Implications
• We need about 72 more fuel volume for LH2
• Possible solution
• Increasing t/c at the outboard sections
• Additional tanks in center section
• Other issues
• This result is based on a 800 passenger BWB
  aircraft
• Smaller BWB aircraft will have smaller relative
  available volume
• Possible implications to distributed propulsion
• Fuel systems and engine placement

44
Future Plans

• Task 1 Models for Distributed Propulsion
• Parametric CFD studies
• Generate Response Surfaces
• Task 2 MDO with Distributed Propulsion And Noise
• MDO Framework and Baseline
• CFD and Response Surfaces for Noise based on TKE
  Level at TE
• Without Distributed Propulsion
• With Distributed Propulsion
• MDO with Distributed Propulsion, Hydrogen Fuel
  and Noise
• Parametric MDO Studies
• Refined MDO Models
• Task 3 Noise
• Implementation of TKE Level at TE Method for MDO
• Continuing Studies of Applicability of Lilley
  Approach
• Consider Other Possible Approaches

45
Backup slides
46
Consider the Engine Weight
47
Engine Weight an Example withMultiple Engine
Sizes
48
Airframe Noise Rating for MDO(Follow G.M. Lilley
Approach)

• Noise Intensity on approach with flaps
• (Lilley, 2001)
• Note u0TKE1/2 is still critical

49
MSES Results(No Engines)

• Upstream Region flowfield
• Initial design engine duct

Pressure Distribution on SC(2)-0710
Airfoil Minf0.721, alpha0.0, Re38.7 million
Cp Comparison of MSES-Viscous andGASP N-S
Analysis
50
Typical Boundary Layer Analysis Results
51
Ejector Noise Reduction System

• Studied by W. Presz, G. Reynolds, C. Hunter for
  Gulfstream GII
• Alternating Lobe Mixer Ejector Concept reduces
  noise
• Simple (no moving parts)
• Offers low-speed thrust augmentation
• Pays for itself (with respect to drag)
• Ejector shroud design is critical to performance

52
List of Sources

• Aircraft with Distributed Propulsion NASA
  Aeronautics Blueprint, 2002
• Conventional Aircraft with Distributed Propulsion
  (many engines electric or jet) NASA Aeronautics
  Blueprint, 2002
• Typical Stagnation Streamlines on Nacelle(GE
  Engine Handbook)

53
Geometry definition

• 5 spanwise stations specified
• Chord
• t/c
• Position of stations
• ¼ chord sweep between stations
• Passenger cabin is defined inboard of span
  station 3
• Only 60 of the chord at that section is used for
  passengers
• Remaining afterbody is used for cargo and baggage
• Fuel tanks are located in the wings outboard of
  span station 3 up to 95 of wing span.

Passenger cabin
Fuel tanks
Spanwise stations
54
Fuselage design

• Wing is designed separate from passenger cabin
• Wing weight is based on empirical wing weight
  equations from Beltrano et. al (NASA CR-4624)
• Cabin design
• Pressure membranes and cabin pressure barriers
  are modeled separate from the wing skin panels
• Non-integral design
• We do not have weight equations for integral
  cabin/wing design

55
ModelCenter screenshots