Study on geostationary orbit transfer using lunar gravity assist and multiple paths

Abstract

It is possible to launch a satellite to Geostationary Equatorial orbit (GEO) from Naro Space Center in South Korea even though that is located in the mid-latitudes of the northern hemisphere. However, when launched from this site, the equatorial inclination after separation will be 80° due to the constraints of launch vehicle and launch site. Since the inclination of GEO satellite has been mainly 0°, a very large plane change maneuver is required to lower the initial inclination to 0°. To solve this problem, the trajectory was designed to greatly reduce the ∆V and put the satellite into the targeted GEO using lunar gravity assist (LGA). There are multiple paths using LGA that depends on the combination of the Earth departure (ascending and descending), free-return trajectory (Circumlunar and Cislunar), and node of Moon’s orbit (ascending node and descending node). In this dissertation, a total of eight (8) multiple paths are defined using these combinations, and a numerical search technique using Newton-Raphson method is applied to launch a GEO satellite from the NRSC, flyby the Mon and enter the GEO. As a result of simulation for each path, it was found that the ∆V was generally larger in case of circumlunar free-return than in case of cislunar free-return. A “∆V compensation” concept was defined to perform direct comparisons with previous studies as the initial orbit size and inclination used in previous studies are different. Through the direct comparison, it was confirmed that the path of ADCSL requires the least ∆V. In addition, it was also confirmed that the ∆V required for this case is about 1,762 m/s, which is about 260 m/s larger than the 1,500 m/s required for going from GTO to GEO without plane change maneuver. These results indicate that trajectory presented in this dissertation can be used in practice if a GEO satellite secures more fuel or if the launch vehicle puts the satellite into an orbit higher than the GTO. In addition, in order to investigate how large the effect on seasonal variations is, a scenario of monthly basis for one year was constructed and simulations were performed. As a result of seasonal variation, it was found that the case of DACSL and ADCSL required low ∆V, and it was confirmed that summer and winter seasons was required a larger ∆V than spring and fall seasons.