Satellite collocation control strategy in COMS

Transcription

1 SpaceOps Conferences May 2016, Daejeon, Korea SpaceOps 2016 Conference / Satellite collocation control strategy in COMS Yoola Hwang *1 Electronics and Telecommunications Research Institute, 218 Gajeong-ro, Yuseong-gu, Daejeon 34129, Republic of Korea, Byoung-Sun Lee 2, 3 Electronics and Telecommunications Research Institute, 218 Gajeong-ro, Yuseong-gu, Daejeon 34129, Republic of Korea University of Science and Technology, 217 Gajeong-ro, Yuseong-gu, Daejeon 34113, Republic of Korea, and Unseob Lee 4 Satrec Initiative, 441 expo-ro, Yuseong-gu, Daejeon 34051, Republic of Korea Orbital slots and frequency bands for geostationary satellites are not enough to operate them in Korea. At , Communication, Ocean, Meteorological Satellite (COMS) is placed and two more GEO-KOMPSAT2 (GK2) A and B satellites in the same window will be launched in 2018 and Thus the strategy of collocation is required to operate satellites safely. The most popular strategy is to control eccentricity and inclination vectors to operate within narrow control-box. However, the uncertainty of orbit determination (OD) using single station tracking data is so high in the operation of three satellites at the from the ground station of the Also another issue is since three satellites have only one side solar panel they have to perform wheel-off loading (WoL) maneuver for momentum dumping twice per day. Collocation strategy in COMS should overcome these OD uncertainties and take into account daily WoL maneuver time. According to the operation scenarios of COMS, GK2A, and GK2B, both longitude separation method and the novel combined eccentricity and inclination vector control strategy will be applied and validated by calculating the relative position differences for each satellite. I. Introduction he geostationary satellite is normally operated to maintain within limited box area. The geostationary satellites are getting to collocate in the same longitude due to the growing number of geostationary satellites and limited frequency band. In case of two geostationary satellites collocation, in the past, longitude separation was typically used. However, recently eccentricity and inclination vector control is usually used because of the narrow control band operation. Communications, Ocean, Meteorological satellite (COMS) located at East (E.)128.2 will be collocated with other two satellites, Geo-Korea-multi-purpose satellites A and B (GK-2A and B). Tentatively COMS will be separated into longitude with other two satellites: COMS will be at the nominal E. longitude and GK-2A and B will be collocated at E. Collocation control strategies of geostationary satellites have been researched by many engineers. Lee et al. [1] studied collocation of three geostationary satellites at 116 E. and they found best solution by eccentricity and T 1 Principal Researcher, Aerospace Systems Research Section, ylhwang@etri.re.kr, Director, Aerospace Systems Research Section, lbs@etri.re.kr 3 Professor, Aerospace System Engineering, lbs@ust.ac.kr 4 Senior Research Staff, Ground System Division, uslee@satreci.com Copyright 2016 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.

2 inclination vector control. Each satellite separated by more than 8 km distances for six months. Also they analyzed two geostationary satellites and inclined geosynchronous satellite [2]. Park et al. [3] showed how to manage multiple satellites using the parallel condition of eccentricity and inclination vector and eccentricity control bias limit conditions. In this research we consider two test-cases: one is to place COMS at E. and the other satellites are collocated at E. Actually they seem to collocate the longitude of E within 0.1 control limit box. The other case is to collocate at E. within 0.05-degree control limit box using eccentricity and inclination vector control. For each test cases, we calculated the distances between each satellite and estimated the required velocity increments for station-keeping planning. A. Covariance Analysis II. Orbit Determination Uncertainty COMS has special issue in determining OD. Distance between satellite and ground station is too close to adjust satellite s states and azimuth bias at the same time. Therefore, the azimuth bias should be corrected into azimuth measurements before OD. The OD error is an important factor in planning to collocate satellites. COMS OD has been estimated using ranging data and 13m antenna tracking data of single station. The analyzed orbit error is roughly 2~3 km RSS by measurement biases and noises [4]. If two ground stations, Daejeon (KARI) and Dongara stations, which are used as the error factor for East-West (EW) dead band allocation can be reduced into 1~2 km Root-Sum-Squares (RSS) from 2~3 km single station OD error. OD covariance analysis was achieved using GEODA [5] program, distributed to the public. Table 1 summarizes the noises and biases of tracking and ranging data for each ground station. Measurements noises and biases are given by COMS s as the GEODA input. Figure 1 shows OD error using single station s ranging and tracking data. The OD errors of along-track and crosstrack directions show larger than radial direction error. Figure 2 is the results of OD using ranging data of two ground stations. Along-track OD error occupies most of position error. 3-dimension OD uncertainty using two ranging data shows almost 0.8~1.0 km RSS by the GEODA covariance analysis. Table 1 Ground stations information Daejeon Longitude: Latitude: Altitude: 100.0m Tracking data 0.01-degree noises degree biases 1 h interval Ranging data 10m noises 20m biases 1 h interval Dongara Longitude: Latitude: Altitude: 251.1m Ranging data 10m noises 20m biases 1 h interval Figure 1. OD uncertainty using tracking and ranging data of single ground station

3 Figure 2. OD uncertainty using ranging data of two ground stations III. Station Keeping Maneuver Strategy The geostationary satellite is perturbed by Earth gravity, solar radiation pressure and luni-solar gravity. When the satellite is located at E., it drifts to the 75 E., because of Earth s tesseral harmonics [1]. Thus, in order to prevent geostationary satellite from drifting to westward, East-West Station-Keeping (EWSK) is performed periodically. Luni-solar gravity is caused by inclination changes and right ascension ascending node [6]. North- South Station-Keeping (NSSK) maneuver keeps geostationary satellite out of plane change. A. Allocation of dead band To allocate possible station-keeping box, Lee et al. [7] analyzed error factors due to the drift and eccentricity control, orbit determination (OD), maneuver realization, and luni-solar perturbation. In longitude of E., as previously mentioned, OD uncertainty using single ground station showed roughly 3km RSS [4]. Using two ground stations, the OD uncertainty can be reduced to 1~1.5 km RSS [4, 5]. EW error due to the drift is and eccentricity error becomes And the luni-solar perturbation can be allocated to Maneuver realization error can be allowed to the amount of 10 % from velocity increments, less than 0.05 m/s, for 7-day period. Thus, according to the E. EW dead-band allocation calculation (Table2), drift and eccentricity control for more than three satellites should be less than Table 2 EWSK allocation at for a 7-days [7] Effects Guard band for OD and maneuver error Guard band for luni-solar perturbations Allocation for Drift Allocation for eccentricity Allocated Values 0.02 deg deg deg deg.

4 B. Collocation Strategy To place three satellite around E. two test cases are planned. One is to use control-box within 0.1 like longitude separation strategy. COMS will be placed at E. and the other satellites will be located at E. Other satellites are operated using the combined eccentricity and inclination vector control. For both scenarios, three satellites station-keeping maneuvers are weekly performed as given by Table 3. And Wheel off-loading (WoL) maneuver is performed twice per day. The maneuver time and velocity increments follows real COMS operation concept. SK maneuvers for collocation are achieved only once per day for each satellite. Table 3 Maneuver Schedules Sun Mon Tue Wed Thu Fri Sat COMS NS EW GK2A NS EW GK2B EW NS *Everyday WoL maneuver is performed twice. EWSK maneuver uses one-burn method and NSSK applies Minimum-Fuel-Target (MFT) strategy in this simulation. In NSSK method, at first NSSK maneuver control initial values using MFT Initialized method to target the biased position. And then the inclination vector is controlled by MFT strategy. One-burn controls westward drift and eccentricity vector by sun-pointing perigee. If one-burn is difficult to control eccentricity vector then two-burn method is adapted. Table 4 shows orbital elements used to simulate collocation strategy. Orbit elements of COMS at E. and area to mass ratio are used commonly to GK2A and GK2B. In the beginning of simulation, large maneuvers are performed to place at the targeting position. Table 4 Three satellites Parameters for Colocation Three satellites Epoch :10: Semi-major (km) Eccentricity Inclination (degrees) Argument of Perigee (degrees) Right Ascension Ascending Node (degrees) Mean Anomaly (degrees) Area to mass ratio (m 2 /kg) A. Case-1: Longitude Separation and eccentricity and inclination control Case1 is to separate satellites in different longitudes. COMS is normally operated at E. Other satellites are collocated by controlling eccentricity and inclination vector. GK2A and GK2B can be separated by x-axis reference of eccentricity and inclination vector. Figures 3 and 4 show the collocation operation scenario for case 1 in detail. Figure 3 implies the longitude separation and figure 4 shows the eccentric and inclination vector control circle. As provided in Table 5, y-components of eccentricity and inclination vector are biased to separate two satellites to be able to collocate.

5 Figure 3. Longitude separation for COMS and GK2A and B Figure 4. Eccentricity and inclination vector controls for GK2A and B Table 5 Three satellites information for case1 COMS GK2A GK2B Longitude Control Target Eccentricity X Control Target Eccentricity Y Control Target Inclination X Control Target Inclination Y Figures 5 to 8 show the results of collocation for case 1. Figure 5 shows mean eccentricity evolution for one year. The eccentricity vector follows sun direction by counter-clockwise direction for both GK2A and GK2B. Figure 6 displays inclination vector evolution. The initial NSSK maneuver was performed by MFT initialized method adding inclination vector bias to send to the target position. To find proper initial inclination vector to be located, NSSK maneuver fired abnormally to the biased direction at the first time. The eccentricity vector of GK2B evolves downward to revolve in the reference of the biased target value as shown in figure 5. In figure 6, the first abnormal inclination vector traces are excluded in GK2B. The simulation of collocation begins from April, 2016 and ends in the middle of March in 2017.

6 Figure 5. Eccentricity vector history for one year (GK2A and GK2B) Figure 6. Inclination vector history for one year (GK2A and GK2B) Figure 7 shows the longitude variations according to the time. They show the same pattern for longitude variations and the patterns have a tendency to be shifted due to the maneuver control by different dates. COMS drifts around E. GK2A and GK2B collocate at E. Figure 8 shows the longitude versus semi-major axis changes. Table 6 shows the total velocity increments used for one year. EWSK maneuvers are planned to use 1.9 m/s and NSSK maneuver roughly uses 40 m/s. Here maneuver realization errors are excluded.

7 Figure 7. Longitude evolutions for COMS and GK2A and B (case1) Figure 8. Longitude versus semi-major axis variations for COMS and GK2A and B (case1) Table 6 Required total velocity increments for case1 Satellite Delta-V EWSK (m/s) Delta-V NSSK (m/s) COMS GK2A GK2B B. Case-2: Collocation by Eccentricity and Inclination Vector Control Case 2 controls eccentricity and inclination vector to collocate three satellite at E. The simulation of collocation for the COMS, and GK2A and B use orbital elements given in Table 4. They are evolved according to the eccentricity and inclination vector control strategy. Every week NSSK and EWSK maneuvers are performed as provided in Table 3. As they use same orbital elements, the first maneuver is performed exceptionally to find targeting biased position. Figure 9 shows how to manage the COMS, and GK2A and B by eccentricity and inclination vector. Table 7 provides eccentricity vector and inclination vector bias values used in this simulation. The eccentricity limit circle depends on the area to mass ratio of each satellite.

8 Figure 9. Eccentricity and inclination vector controls for three satellites Table 7 Target bias information of three satellites collocation for case2 COMS GK2A GK2B Longitude Control Target Eccentricity X Control Target Eccentricity Y Control Target Inclination X Control Target Inclination Y Figures 10 to 13 show the results of collocation for case 2. Figure 10 shows the eccentricity vector history for one year for three satellites. Since we use same area to mass ratio, same magnitudes for the eccentricity circle limit value were used. Thus the sizes of the evolved eccentricity vectors are almost same magnitude. This phenomenon also happens to the inclination vector control. The eccentricity vectors trace sun revolution direction. Figure 11 shows the inclination vector history. Figure 12 indicates the longitude variations for three satellites. Figure 13 shows the drift of semi-major axis according to the longitude variations. Figures 12 and 13 show that three satellites keep ±0.05 control-box well. Table 8 shows the total magnitude of the planned velocity increments for one year. The velocity increments of EWSK and NSSK maneuvers seem to show the similar magnitudes used in real COMS operation. Figure 10. Eccentricity vector history of three satellites for one year

9 Figure 11. Inclination vector history of three satellites for one year Figure 12. Longitude evolutions for COMS and GK2A and B (case2) Figure 13. Longitude versus semi-major axis variations for COMS and GK2A and B (case2) Table 8 Planned total velocity increments for case2 Satellite Delta-V EWSK (m/s) Delta-V NSSK (m/s) COMS GK2A GK2B

10 IV. Satellite Relative Distances In collocation simulation, we propagated satellite position including velocity increments according to the following dynamics models: the gravity model, luni-solar perturbation and solar radiation pressure. For each case, COMS-GK2A, COMS-GK2B, and GK2A-GK2B distances are calculated into true-of-date Cartesian coordinate system. Figure 14 shows the relative distances between three satellites. Since the COMS is actually located at different longitude from GK2A and GK2B the relative distances with respect to the COMS are larger than the relative positions of GK2A and GK2B. The collocated position of GK2A and GK2B keeps more than 5km RSS as expressed in grey color. Figure 15 shows the each relative distance for three satellites which are collocated at They keep more than 5 km RSS relative position differences within ±0.05 control-box. Figure 14. Orbit differences between COMS-GK2A, COMS-GK2B, and GK2A-GK2B for case1 Figure 15. Orbit differences between COMS-GK2A, COMS-GK2B, and GK2A-GK2B for case2

11 V. Conclusions To allocate three satellites around E. OD analysis and EW dead bad allocation were studied. OD using single station including maneuver error is occupied to 0.02 and the other perturbation errors including drift error are allowed to have Case1 is to operate that COMS is controlled in the left side from E within 0.05 control-box and GK2A and GK2B are controlled by eccentricity and inclination vector on the right side from E. In case2 eccentricity and inclination vectors are controlled for three satellites to be operated within ±0.05 control-box. Also three satellites keep them to have minimum distances more than 5 km RSS. In spite that we consider WoL maneuver, the collocation strategy for three satellites using bilateral triangle for eccentricity and inclination vector controls satisfactorily within ±0.05 mission limit box. While the scenario of case 1 allocates ± 0.1 control-box, the scenario of case 2 makes it possible to maintain three-satellite within ±0.05 control-box at E. References [1] Lee, B.-S., Lee, J.-S., and Choi, K.-H, Analysis of station-keeping maneuver strategy for collocation of three geostationary satellites, Control Engineering Practice, Vol. 7, pp , [2] Lee, B.-S., and Choi, K.-H, Collocation of two GEO satellites and one inclined GSO satellite, Aerospace Science and Technology, Vol. 4, pp , [3] Park, B.-K, Tahk, M.-J., Bang, H.-C., and Choi, S.-B., Station Collocation Design Algorithm for Multiple Geostationary Satellites Operation, Journal of spacecraft and rockets, Vol.40, No. 6, Nov.-Dec., [4] Hwang, Y., Lee, B.-S., Validation of geostationary satellite orbit determination using single-station antenna tracking data, Journal of spacecraft and rockets, Vol.50, No. 6, pp , [5] Montenbruck, O. and Gill, E., Satellite Orbits Models Methods Applications, 1st ed., Springer-Verlag, NewYork, [6] Pocha,J.J., An Introduction to mission design for geostationary satellites, Dordrech: Reidel, [7] Lee, B.-S., Hwang, Y., Kim, H-Y., and Park, S., East-West Station-Keeping maneuver strategy for COMS satellite using iterative process, Advances in space research, Vol.47, pp , Acknowledgement This work was supported by the Space Core Technology Development Program of NRF. [NRF- 2015M1A3A3A , Development of core S/W standard platform for GEO satellite ground control system]