Maximizing Propellant Consumption To Prepare An Impaired Satellite For Retrieval
Open Access
- Author:
- Shenoy, Ashish
- Graduate Program:
- Aerospace Engineering
- Degree:
- Master of Science
- Document Type:
- Master Thesis
- Date of Defense:
- December 12, 2019
- Committee Members:
- Robert Graham Melton, Thesis Advisor/Co-Advisor
Amy Ruth Pritchett, Program Head/Chair
Puneet Singla, Committee Member - Keywords:
- PSO
Particle Swarm Optimization
Satellite
Retrieval
Impaired Satellite
Propellant Consumption
Orbital Elements
Gauss Variational Equations
GVE - Abstract:
- A man-made Earth satellite is a complex device consisting of a payload and a variety of subsystems that is placed into an orbit around Earth. Due to environmental factors and technological errors, the satellite’s functions may become impaired which reduces or halts its ability to complete or continue its mission. An impaired satellite can either be left in orbit or could be retrieved by a retrieval mission if the satellite’s stakeholder deems the satellite valuable enough to warrant repair. Any unused propellant remaining in the satellite must be used up to ensure a safe retrieval. It cannot simply be vented because it could create hazards for other satellites, such as adsorbing onto optical-sensor surfaces, but also because there is generally no means provided for venting unused propellants. The only other method to use up remaining propellant on a satellite is to turn on the thrusters for a specified burn period; however, doing so will change the orbit of the satellite which will require extra effort and resources to determine a suitable rendezvous point for a retrieval mission. This necessitates a solution that involves firing the satellite thrusters to completely use up all propellant without significantly changing the satellite’s orbital characteristics. This thesis used Particle Swarm Optimization (PSO) techniques to determine such a solution. PSO is an efficient meta-heuristic algorithm that uses swarm interactions and behavior to determine the best solution that fits the constraints of the problem it tries to solve. PSO was applied to the Gauss Variational equations to obtain the optimal timevariable thrust-angle functions that maximize the propellant use and minimize the change in the orbital elements of a satellite. 2-D and 3-D PSO formulations of the Gauss Variational Equations were tested on several sample orbits. Each orbit was tested in two scenarios, one where the burn time was user-fixed and one where the burn time was PSO-determined. The PSO algorithm was able to determine optimal thrust-angle functions for all cases presented. The effectiveness of solutions gathered was demonstrated by the fact that each solution resulted in a final set of orbital elements equal to the initial set of orbital elements for each orbit to at least 10-12 accuracy while also consuming all of the propellant. The success of applying PSO methodology to minimize the change in orbital elements under specific conditions supports the notion that PSO can be applied to other types of specific or general orbital mechanics problems.