Thruster layout in spacecraft with tight control requirements – optimization models and analysis

The present thesis work focuses on the search for an appropriate thruster configuration, suitable for the Next Generation Gravity Mission (NGGM). The goal of the NGGM mission is to measure the temporal variation of the gravitational field in the Earth’s system. Following the success of the previous GOCE and GRACE missions, the NGGM one is intended as the Earth Observation mission of the future. The main technological innovation introduced, concerns the exploiting of the laser interferometry technique, between satellite pairs, to collect data with an unprecedented spatial and temporal resolution. As a monitoring mission, the NGGM formation needs to fly at relatively low altitudes (of around 350 km), where the presence of the atmospheric drag is not negligible, for a long period of time (of around 7 years, according to the current scientific requirements). In addition, the laser interferometer and accelerometer payloads impose strong requirements in terms of attitude and orbit control, with the aim of compensating the drag perturbations and guaranteeing precise pointing and angular rate control. The control law, thus obtained, in terms of force and torque request, represents the reference operational scenario. The main objective of the entire study is the identification of an optimized thruster layout, in the context of minimizing the total propellant consumption w.r.t. to the entire mission. The problem thus formulated belongs to the class of large-scale optimization problems, which are very challenging to solve tout court by any global optimization approach. Due to the nonlinear and non-convex nature of the equations involved, the identification of sub-problems easy to handle represents a good strategy: following this logic, two different models have been determined. The first one provides the thruster positions and their approximate orientations; the second performs a refinement of the orientations previously found. Considering the strong relationship between the two models, the present thesis work focuses on the description of both, with a particular care to the first phase of the second model. This represents the phase of verification of the feasibility and/or minimization of the infeasibility. In particular, the treatment will describe the development of three parts: the first part concerns the adjustment and the extension of the relevant mathematical model; the second part concerns the implementation of the algorithm of resolution associated to the model; the third part concerns the experimental data analysis. The analyses conducted are based on a dedicated local optimization approach. Finally, the basis for the next development phase have been laid: the next step is to verify if the thruster layout found by the present analysis also meets the ultimate objective of minimizing the fuel consumption; if this does not happen, new orientations need to calculate.