System Architecting for GEO Communication Satellite Considering On-Orbit Refueling

TL;DR

This work addresses GEO satellite design under commercial on-orbit refueling (OOR) by jointly optimizing design lifetime $T_{life}$ and initial propellant $m_{p,des}$ within a multi-objective framework $\left[E_V, E_V/\sigma_V\right]$, balancing profitability and risk. A surrogate-model-assisted framework atop a detailed satellite lifecycle simulation evaluates economic outcomes under uncertainty, with revised mass/cost models that decouple lifetime from propellant mass. Through case studies of chemical and electric propulsion, the study shows OOR can drive architectural shifts for chemical propulsion when OOR capacity is high and cost is low, but yields limited impact for electric propulsion due to small propellant mass fraction. The results provide actionable thresholds for service capacity and pricing and establish a scalable methodology to assess OOR-enabled architectures across space systems. The approach offers a practical path for operators and manufacturers to quantify value under servicing-enabled life-extension while incorporating risk via a mean-variance-inspired objective.

Abstract

This paper introduces the problem of selecting a satellite system architecture considering commercial on-orbit refueling (OOR). This problem answers two questions: "What design lifetime should the satellite have?" and "How much propellant should be carried at launch?" We formulate this as a mathematical optimization problem by adopting design lifetime and initial propellant mass as design variables and considering two objective functions to balance the returns and risks. To solve this problem, we develop a surrogate model-based framework grounded in a satellite lifecycle simulation. The framework captures various uncertainties and operational flexibility, and integrates a modified satellite sizing and cost model by adjusting traditional models with OOR. Based on the developed framework, we conduct a case study of GEO communication satellites to examine current target service performance and explore the potential of a new system architecture that diverges from traditional design trends.

Figures (10)

• Figure 1: Flow diagram of the satellite lifecycle simulation.
• Figure 2: Decision timings and alternatives.
• Figure 3: Proposed satellite system architecting framework.
• Figure 4: Contour plot and heatmap of $\hat{J}_1$ (left) and $\hat{J}_2$ (right) for the baseline scenario.
• Figure 5: Pareto front of baseline scenario in design space (left) and normalized objective space (right).