An Empirical ModelSRP IGSWS2018-PS06 for the Orbit Normal … · 2019-01-02 · SRP accel....
Transcript of An Empirical ModelSRP IGSWS2018-PS06 for the Orbit Normal … · 2019-01-02 · SRP accel....
Definition of -TBMECOM
- version with just 2 parameters coveringSRP SRP due to SP:
- can be ignored for short-arcs =>β (T20, T30)ECOM TM-
- SRP due to sat.-body may be absorbed by stochastic orbitparameters ( ) or by an a priori box modelECOM-TBMP
Definition -TBof ECOM
- SRPacceleration during ON described in TERM system as ...
- ... by 9 parameters:SRP
- β-dependency can be ignored for short-arcs => ECOM-T
- stochastic orbit parameters (i.e., pulses or accelerations) canbe added optionally => ECOM TBP-
Introduction
The Center for Orbit Determination in Europe ( ) is contributing toCODE
the Multi- Extension ( ) since its start in 2012 with anIGS GNSS MGEX
orbit and clock solution. `s ( ) product includes theCODE MGEX COM
satellite systems , , Galileo, BeiDou2 ( 2), andGPS GLONASS BDSQZSS. GEO satellites are not included, so far.
The ECOM2 solar radiation pressure model, used for all satellites( )SRP
in the COM solution, has been designed for yaw-steering ( ) attitude.YS
Consequently, the orbit quality was degraded for BDS2 and QZS-1spacecrafts ( ) at times when operating under orbit-normal ( )SC ON
attitude, i.e., during or close to eclipse seasons.
Here we introduce a family of empirical models taylored for the -SRP ONmode, which can either be used on top of an a priori model or as aSRPstand-alone model.
International ServiceGNSSWorkshop 201829 Oct. - 02 Nov. 18 Wuhan China20 , ,
L .. Prange , R. Dach , G Beutler , A. Villiger ,1 1 1 1
S. Schaer , D. Arnold , A. Jäggi2 1 1
1Astronomical Institute, University of Bern, Switzerland
2Federal Office of Topography swisstopo, Wabern, Switzerland
An Empirical ModelSRPfor the Orbit Normal Mode
Poster compiled by ,L. Prange October 2018Astronomical Institute, University of Bern, Switzerland
@aiub.unibe.chlars.prange
IGSWS2018-PS06
Contact address
Lars PrangeAstronomical Institute, University of BernSidlerstrasse 53012 Bern (Switzerland)[email protected]
Posters and other publications from theSatellite Geodesy Group:AIUB
http://www.bernese.unibe.ch/publist
Attitude Modes
Most SCs are oriented with attitude. SomeGNSS YSBDS QZS ON2 satellites and -1 switch to attitude forsmall elevation angles of the Sun w.r.t. the orbital plane.βAll spacecrafts maintain the attitudeGEO ONpermanently.
�u�u
�
Z
Z
Z
Y
Y
Y
Y
X
X
Spacecraft-fixed reference frame( aw- teering attitude):Y S
X
X
SPN = D
Fig. 1: Satellite-fixed ref. frames under - and -mode.YS ON
ECOM SRP Model Frames
During -mode the co-rotating decompositionYS ECOM
is practical ( due to solar panels absorbed by oneSRPcoeff.; Y=0). For -mode non-rotating frames are betterONsuited (e.g., frames fixed to local orbit or orbital plane). Weuse the Terminator System Decomposition ( ).TERM
�u�u
�
Z
Z
Z
X
X
Spacecraft-fixed reference frame( rbit- ormal attitude):O N
X
X
Y = W
SPN
D
Y = W
Y = W
Y = W
�u�u
�
= D
T3 = D
T3 = D
T3 = D
T2
Terminator
line
T1
T1
T1
T2
T2
T2
TERM T( erminator) frame:
T1
Fig. 2: model frame definition.SRP
SRP YS ONin - and -modes
Features of the attitude mode are:YS- solar panel ( ) plane is always parallel to the terminator plane andSP
normal to the vector Sun => satellite ( )D
- C`sY-sides are never illuminated (Y=0, apart fromY-bias)S- ody (+Z, -Z, +X panels) causes an signal, which is periodicSC b SRP
w.r.t. the argument of latitude difference betw. SC and Sun ( )Δu
- rapid noon/midnight turns for small elevation angles ( ) of the Sun w.r.t.βthe orbital plane
Features of the attitude mode are:ON- no noon and midnight turns- SP-normal ( ) deviates from the vector D by -angleSPN the β- power generation is reduced compared to the -modeYS- due to SPcan be absorbed by 2 parameters, which are constantSRP SRP
w.r.t. u, but not w.r.t.Δ β- all panels of body can be illuminated - causing periodic signalsSP SRP
w.r.t. u andΔ β
SRP ONSimulation for
Simulations based on released meta data of -1 have beenQZSperformed separately for SP and SC body. Simulated SRPacceleration has been projected into the terminator referenceframe (TERM, Fig. 2). Coefficients of significant size havebeen selected for ( u) and ( u, ) -ECOM-T ECOM-TBΔ Δ βempirical SRP models for the ON-mode.
-100 -50 0 50 100Beta [deg]
-20
-15
-10
-5
0
5
10
15
20
SRP accel. [nm/s**2]
SRP coeff: TERM 0010 T20 for QSP
-100 -50 0 50 100Beta [deg]
-120
-100
-80
-60
-40
-20
0
SRP accel. [nm/s**2]
SRP coeff: TERM 0010 T30 for QSP
Fig. 3: Simulated due to in system.SRP SP TERM
0 50 100 150 200 250 300 350 400
du [deg]
-60
-55
-50
-45
-40
-35
-30
-25
-20
SR
P T
3 [n
m/s
**2]
Simulated SRP acceleration, QBX
0 50 100 150 200 250 300 350 400
du [deg]
-15
-10
-5
0
5
10
15
SR
P T
2 [n
m/s
**2]
Simulated SRP acceleration, QBX
0 50 100 150 200 250 300 350 400
du [deg]
-6
-4
-2
0
2
4
6
SR
P T
1 [n
m/s
**2]
Simulated SRP acceleration, QBX
Fig. 4: Simulated due to box in system. ColorsSRP SC TERM
represent following -angles: , ,β green, dotted: -80° cyan: -30° red:0°, , .blue: +30° green, solid:+80°
Validation -1of QZS
- Simulated and estimated SRPcoefficients have comparablecharacteristics (apart from singularities) and show littlevariations from day to day:
- clear improvement of QZS-1 orbits and clocks:
- agreement with external QZF orbit solution improves:
Validation 2of BDS
- smallest deviation of SRPcoefficients from assumed valuesobserved for MEOs, shorter orbit arcs, and ECOM-TBM(P)
- for IGSOs the periodic SRPcoefficients are not well enoughdetermined to represent the orbit over longer periods, whichaffects the remaining parameters:
- consequently, the orbits are degraded as well in case the fullECOM-TB is used for long-arc POD of BDS2 IGSOs:
Reference to productsCOM
Prange, L.; Arnold, D.; Dach, R.; Schaer, S.; Sidorov, D.;Stebler, P.; Villiger, A.; Jäggi, A. (2018). CODE productseries for the projectIGS MGEX . Published byAstronomical Institute, University of Bern.URL CODE MGEX: http://www.aiub.unibe.ch/download/ _ ;DOI: 10.7892/boris.75882.2.
Validation-Summary and Conclusions
- is a suitable stand-alone SRPmodelQZS-1: ECOM-TB
- : addition of pulses (=> ) or of an aBDS2 MEO ECOM-TBP
priori box model is recommended- : periodic SRPcoefficients cannot be estimatedBDS2 IGSO
for long-arc orbits with current network and orbit model (=>use or ECOM-TBM + an a priori box model)ECOM-TBMP
Possible reasons for reduced performance of ECOM-TB forBDS2:
- sampling problem for periodic SRPcoefficients due topoor observation coverage over EastAsia (affecting IGSOsmore than globally observed MEOs; see Fig. 10)
- open questions regarding BDS orbit model (meta data!)
-160
-155
-150
-145
-140
-15 -10 -5 0 5 10 15
SRP
acce
lera
tion
[nm
/s**
2]
Beta [deg]
T30C1b
Estimated I1Estimated I2Simulated
-30
-25
-20
-15
-10
-5
0
-15 -10 -5 0 5 10 15
SRP acceleration [nm/s**2]
Beta [deg]
T20S3bEstimated I1Estimated I2Simulated
-150
-100
-50
0
50
100
150
0 50 100 150 200 250 300
SLR residuals [cm]
DOY 2015
YS ECOM2 ON ECOM-TB
0
0.5
1
1.5
2
2.5
3
3.5
4
0 50 100 150 200 250 300
RMS of linear clock fit [ns]
DOY 2015
YS ECOM2 ON ECOM-TB
Fig. 5: vs. coefficients of ECOM-TB refer-Simulated estimatedring to different ON-periods and in 2015.I1 I2
Fig. 6: Orbit and CLK-valid. of 2015 or .ECOM2 ECOM-TB
-180
-160
-140
-120
-100
-80
-60
-40
-20
0
-4 -3 -2 -1 0 1 2 3 4
SRP
acce
lera
tion
[nm
/s**
2]
Beta [deg]
T30C1b IGSO 1-day
ECOM-TB ECOM-TBP ECOM-TBM ECOM-TBMP
-180
-160
-140
-120
-100
-80
-60
-40
-20
0
-4 -3 -2 -1 0 1 2 3 4
SRP
acce
lera
tion
[nm
/s**
2]
Beta [deg]
T30C1b IGSO 3-day
ECOM-TB ECOM-TBP ECOM-TBM ECOM-TBMP
Fig. 8: SRPcoefficientT30C1b ofdifferentECOM-TBversions forBDS2 MEOs(top) andIGSOs(bottom) with1- (left) vs. 3-day arcs
-500
-400
-300
-200
-100
0
100
200
300
-10 -5 0 5 10
SLR
resi
dual
s [c
m]
Beta [deg]
BDS2 IGSO 1-day
ECOM2 ECOM-TB ECOM-TBP ECOM-TBMP
-500
-400
-300
-200
-100
0
100
200
300
-10 -5 0 5 10
SLR
resi
dual
s [c
m]
Beta [deg]
BDS2 IGSO 3-day
ECOM2 ECOM-TB ECOM-TBP ECOM-TBMP
Fig. 9: SLR-validation ofBDS2 MEOs(top) andIGSOs(bottom) with1- (left) vs. 3-day arcs(right).
Fig. 10: Formal radial orbiterror of QZS-1, BDS2 MEOs,and BDS2 IGSOs in early2015 (YS and ECOM2applied to all).
Vali.-Method SLR, IQR [cm] CLK-fit, median [ns] OMC, median [cm] ORB-fit, med. [cm]
Sat.-System BI BM Q1 BI BM Q1 BI BM Q1 BI BM Q1
ECOM2 20.5 21.0 62.0 1.72 1.61 1.43 55.9 29.2 42.4 23.0 10.1 14.1
ECOM-TB - - 15.2 - - 0.35 - - 14.2 - - 5.6
ECOM-TBP - 12.2 - - 0.69 - - 9.8 - - 6.8 -
ECOM-TBMP 12.2 - - 0.72 - - 27.1 - - 44.0 - -
Fig. 7: QZS-1 orbit comparison betw. COM and QZF ACs(screenshot taken from )http://mgex.igs.org/analysis
ECOM2 used for COM orbitECOM-TBused forCOM orbit
Tab. 1: Results of orbit validation with SLR residuals (SLR), RMS of linear sat. clock fit (CLK), orbit misclosures (OMC), and 3-day long-arc fit ( )ORB-fit BI for BDS2 IGSO ( ), BDS2 MEO (BM), and QZS-1 (Q1) satellites with different SRP models during
ON-periods in 2014 and 2015.-180
-160
-140
-120
-100
-80
-60
-40
-20
0
-4 -3 -2 -1 0 1 2 3 4
SRP
acce
lera
tion
[nm
/s**
2]
Beta [deg]
T30C1b MEO 1-day
ECOM-TB ECOM-TBP ECOM-TBM ECOM-TBMP
-180
-160
-140
-120
-100
-80
-60
-40
-20
0
-4 -3 -2 -1 0 1 2 3 4
SRP
acce
lera
tion
[nm
/s**
2]
Beta [deg]
T30C1b MEO 3-day
ECOM-TB ECOM-TBP ECOM-TBM ECOM-TBMP
-100
-50
0
50
100
-10 -5 0 5 10
SLR
resi
dual
s [c
m]
Beta [deg]
PRN C11 (MEO) 1-day
ECOM2 ECOM-TB ECOM-TBP ECOM-TBMP
-100
-50
0
50
100
-10 -5 0 5 10
SLR
resi
dual
s [c
m]
Beta [deg]
PRN C11 (MEO) 3-day
ECOM2 ECOM-TB ECOM-TBP ECOM-TBMP
�u�u
�
E3 = D
E2
E1
E2E2
E2
E1
E1
E1
ECOM CE O M( mpirical rbit odel) frame:ODE
E3 = D
E3 = D
E3 = D
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
39 40 41 42 43 44 45
Form
al e
rror
rad
ial
[cm]
Day of year 2015
BDS2 IGSO BDS2 MEOQZS-1