Transcript Station-Keeping Performance of a Large High-Altitude

Near-Space Station-Keeping Performance
of a Large Notional Airship
Dave Schmidt
Professor Emeritus
Dept. Mechanical & Aerospace Engineering
University of Colorado at Colorado Springs
[email protected]
719 262-3580
www.uccs.edu/~sansrl
as presented at the SAE ACGSC
Williamsburg, VA
October 12, 2006
Partially Sponsored by the Army Space & Missile Defense Battle Lab
Outline of the Presentation
Near Space
What is “Near Space”?
Why Are We Interested in It?
Is it Feasible?
- Key Problem - Long-Duration Station Keeping
Wind Environment
Aerodynamics - Drag
Guidance & Control in Turbulence
Solar - Powered Propulsion
Station - Keeping Power Analysis
Summary & Conclusions
Typical “Mission” Communication Or Observation
Example SpaceBased Solution to
Wide-Area
Surveillance
What if We Could
Place a Sensor at a
Lower Altitude
and Keep It There?
Persistent Surveillance Key
Persistent (24/7) Communication and/or Surveillance Platform
Desirable Near-Space Platform Characteristics
• Persistent 24/7, multi-month, all-weather capability
• You can bring it down and fix it…
• Low-cost platform, rapid reconstitution of capabilities
• Improved performance of most space sensors
• “Local” control
• …..
• This is the Promise of Near Space
Space
Where Is Near-Space?
GEO
37,160 km
MEO
21,000 km
100 – 1000 km
Near-Space
LEO
100 km (327,000 ft)
Air
20 km (60,000 ft)
20 km
Surface
SATCOM
GPS
Iridium
US Near-Space Efforts
(Selected)
NASA
USAF
(98,863 feet – World Record)
Helios
NM State
Univ
Advanced Aerobody
Payload: 220 - 500 lbs
Altitude: 50-98K feet
Duration: > 1 week
AFRL
Payload: <1000 lbs
Altitude: 60 – 100K feet
Duration: 3 months
Free Balloons
>2,500 flights since 1951
Payload: up to 7000 lbs
Altitude: Up to 140,000 feet
Army/MDA
High Altitude Airship
Payload: 4K – 12K lbs
Altitude: 65K to 85K
Duration: 1 month
(near-term)
3 - 5 yrs (objective)
“Ascender”
Maneuvering Vehicle
Payload: 100-1000 lbs
Altitude: 80 – 120k feet
Duration: 4-10 days
UCCS
Weather Research
Payload: 6 lbs
Altitude: 100K feet
International Efforts to Develop
Near-Space Platforms
United States
Canada
United Kingdom
Germany
Israel
South Korea
Japan
Malaysia
Outline of the Presentation
Near Space
What is “Near Space”?
Why Are We Interested in It?
Is it Feasible?
Key Problem - Long-Duration Station Keeping
Select a Typical Vehicle for Analysis
Model Wind Environment
Model the Aerodynamics
Model Solar - Powered Propulsion
Perform Station - Keeping Power Analysis
Assess Feasibility - Identify Possibilities
Summary & Conclusions
Key Issue - Station Keeping
Objective: Remain on Station in Presence of Winds
Issue: On-Board Power Available vs Power Required
Power Available  Power Required
1

2
(I Solar ηSolarCells ηStorage  PPayload ) ηMotor ηProp   ρVWind
 CD AVWind + PManeuver
Auxiliary
2

Requires Assessments of VWind CD PManeuver ISolar η's
Notional Airship Investigated
Vehicle Characteristics:
Volume = 6.1x106 ft3
Length = 450 ft
Width = 100 ft
WOEW = 30,000 lbs
Solar-electric powered
Electric motors/propellers
Payload Power = 3 kW
Performance Req’m’ts:
Operating Altitude - 65,000 ft
Endurance ~ 1 Year
Maintain Position in Winds
Wind Analysis
White Sands (EPZ) and Akron (PIT)
(32.5 deg North, 106.5 West), (41 deg North, 81.5 deg West)
Statistical Wind Modeling
STATISTICS:
Possible “Sweet
Spots”
Year 2004 Wind Variation:
White Sands (EPZ) and Akron (PIT)
Monthly vs Annual Wind Statistics
Annual Stats
(2004)
Monthly Stats
(2004)
White
Sands
(July)
Akron
Speed
(kts)
Speed
(kts)
White
Sands
Akron
Speed
(kts)
Speed
(kts)
50
15.7 kts
13.4 kts
50
27.4 kts
34.4 kts
95
31.4 kts
35 kts
95
35.5 kts
68.5 kts
99
38.6 kts
47.5 kts
99
38.4 kts
85.6 kts
P Winds
P Winds
(Dec)
Key Points: Monthly Variations Can Be Significant, and
Probability distribution is key, not just means
Altitude, K ft
Dryden Turbulence Model
•
Frequency Spectra


2 
1
+
L
ω/U




u
0
Φu (ω) = σ u (2Lu /πU 0 ) 
2
g
1
 1 + 3  L ω/U 2 
2
v
0

Φ v (ω) = σ v (L v /πU 0 ) 
2 2
 1 +  L v ω/U 0   

 
g
RMS Turbulence Intensity, fps
Aerodynamic Drag Estimation
Drag
Power Required
• UCCS Low-Speed Tunnel
• Lift-Drag Force Balance
• Body geometry L/D = 3.9
• ReL = 1.3 x 105 (Low)
• Turbulence Grid Needed
• Mmax = 0.045 (50 fps, 30 kts)
Airship Model in Tunnel Test Section
Drag = (1/2 V2) CDAref
CD - Empirically & Experimentally Determined
Airship Dynamics Modeling
Non-dimensional Aero Coefficients and Mass Properties
From Nagabhushan & Tan (1995)
Mass Properties Scaled
Apparent Masses Scaled
Static Stability Restored
Roll DOF Ignored
Sluggish, Heavily Damped Response
Degree of
Freedom
u (fps)
Transfer Functions
From Thrust, t (lbs)
0.000925 (s+0.261) (s+0.527)
(s+0.261) (s+0.0156) (s+0.527)
v (fps)
0
r (rad/s)
0
Transfer Functions
From Rudder, r (rad)
0
-0.0958 (s+13.9) (s+0.0156)
(s+0.261) (s+0.0156) (s+0.527)
-0.0390 (s+0.0824) (s+0.0156)
(s+0.261) (s+0.0156) (s+0.527)
Station Keeping In Turbulence
Steady Wind, VWind
Plus u and v Gusts
ug vg
Objective:
Drive Position Error To Zero
In Presence Of Wind Gusts
X
Desired Position
(Looking From Above)
Y Error
X Error
Y
Motion Decomposition:
Vehicle Relative to Steady Air Mass
Air Mass Relative to Inertial
Gust Superimposed on Steady Air Mass
Guidance & Control Architecture
GPS Based
Wind Gusts (turbulence)
ControlsInertial
Inertial
Thrust &
Position
Velocity
Fin Defl.
Desired Position
-
Guidance
Algorithm
-
Attitude
Controller
Vehicle
Dynamics
Kinematics
Xperturbation = (U 0 + u) - VWind = u
Yperturbation = U 0 ψ + v
Control laws synthesized via classical loop shaping
Allows for assessment of maneuver-power required
Closed-Loop Time Responses
Position Error X, ft
Time Responses
Position Error Y, ft
Heading deg
Surge velocity u, fps
Lateral velocity v, fps
Yaw rate r, deg/sec
Thrust t, lbs
Rudder defl. r, deg
Time, sec
Stochastic Control-Power Assessment
Linear Closed-Loop Dynamic System (BW)
+
Turbulence Model
Lyapunov Equation
Response Covariance Matrix Y
Maneuver-Thrust Required
Severe Turbulence
Longitudinal Performance
100000
RMS X error, ft
10000
Increased Control
Bandwidth
1000
100
10
1
1
10
100
1000
RMS Thrust, lbs
Note: To Keep rms Position Error < 1000 ft Requires 20 lbs Thrust rms
Increasing Total rms Station Keeping Thrust Required ~6%
Pulling it All Together
# Wind Statistics and Probability Distributions
# Aerodynamic Drag Models
# Maneuver Power Required
Solar Geometric Effects
Vehicle geometry - large airship
Geo latitude
Diurnal and seasonal effects
Energy-Subsystem Efficiencies
Power Available
Versus
Power Required
Average Solar Insolation Available
Purple = Akron (kW)
Blue = White Sands (kW)
Includes Effects of
Day Length
Sun Angles
Average solar insolation by month
Akron 41 deg
White Sands 32 deg
600
500
W/m^2
400
300
200
100
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Diurnal & Vehicle Geometry Effects
1.01
N-S
0.80.8
Fraction
0.60.6
Of
Useful
Area
0.40.4
E-W
0.41 Ave NS
0.32 Ave EW
0.20.2
00 0
0
5
10
15
10
Hour of Day
20
20
Airship Power System Schematic
Solar
Incident
Energy
Payload
Auxiliary
Systems
Solar
Cells
Energy
Storage
Power Mgmt. Options:
Instant Usage
- - Store Then Use
Motors
Propellers
= Energy Conversion
Power Required vs Available
(Mean (P=0.5) by Month
power required and power available
Akron required
Akron available
White Sands required
White Sands available
109kW
70
60
Ave.
Power
Deficit
Excess Ave.
Power in Feb.
50
kW
40
30
20
10
0
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Power Required vs. Power Available
Monte Carlo Simulations
Variation from random velocity
600
Battery Power, kW*hrs
battery type:
all to battery
Cd:
0.12
capacity:
600 kWh
month:
Feb
motor efficiency:
80%
solar cell efficiency: 6%
battery efficiency: 60%
500
400
Ave.
Excess
Power
>0
300
200
100
0
0
no poweravailable
1
2
3
days
Time, days
4
5
6
Summary of Key Results
• Near-Space Offers Promise In Communication and Surveillance Missions
- Cost-Effectiveness, Reliability, Flexibility, Persistence
• Key Feasibility Issues Include Power Limitations and Wind Vulnerability
This Airship Predicted to Have Insufficient Average Power at Akron in December.
Average Excess Power Available - Necessary, But Not Sufficient Condition
to Assure Viability – More Detailed Analysis Required.
Also, Probable Insufficient Power at Akron in Other Months, Due to
“Gambler’s Ruin” Phenomenon.
Large Seasonal Variation in Winds - Wind Statistics & Distribution
Power Required vs. Power Available - Out of Phase
Power Balance Very Sensitive to Wind Speed (~V3), Drag Coefficient,
and Power-system Efficiencies (linear multiplicative dependence)
Conclusions Based on Results
This Vehicle is Very Power Limited
The Vehicle Could Not Remain Aloft for a Year in Winds Like Akron’s
Slight Change in Power-system Efficiencies - Large Change in Results
-- Payoff from Additional Technological Advancements
In Particular, Novel Long-Term Power Storage Technology Would
Have High Payoff
Other Vehicle Configurations Need Similar Assessment,
To Determine Sensitivity of Results to Vehicle Configuration
Back Up Slides
Potential Fields of View
At Sub-Orbital Altitudes
80,000 feet
60,000 feet
700 Miles
600 Miles
Near Space Potentially Offers An Alternative
Cost-effective,
persistent
wide area surveillance,
communications, etc.
This is Why
There is Interest
Rudder Control-Power Required
Lateral Performance
100000
RMS Y error, ft
10000
Increased Control
Bandwidth
1000
100
10
1
1
10
100
RMS Rudder Deflection, deg
Note: To Keep rms Position Error < 1000 ft Requires 9 deg Rudder rms