(A) study on optimal deployment and maintenance strategy of satellite mega-constellation

Abstract

The new era of space-based broadband service has emerged, backed by a satellite mega-constellation in the low Earth orbit. One example is OneWeb which plans to offer high-speed, worldwide internet access using a 6,372-satellite constellation. Space X is also rapidly building up the Starlink constellation, composed of 12,000 satellites. In addition, with the miniaturization of electronics, low manufacturing cost, adequate performance, and reliability, new constellations composed of hundreds to thousands of small satellites to capture the demand of global service have been proposed and deployed. However, the cost involved in these large constellations is enormous, necessitating an integrated strategy for efficient deployment and maintenance. This doctoral thesis addresses issues related to deploying and maintaining satellite mega-constellations. The design of constellation deployment mission and maintenance strategies considered in this thesis can be interpreted as a multi-echelon dual-channel supply chain model. The deployment problem is projected onto the established supply chain model, then is formalized after substituting it with a dynamic network flow problem. The satellite constellation maintenance problem is formalized by replacing it with an inventory management problem that utilizes the established dual-channel supply chain model for spare options that replace failed satellites with new satellites. The optimal deployment problem is represented by mixed-integer linear programming, and the two-phase framework is proposed to establish optimal strategies. The proposed framework consists of an evaluation phase and an optimal planning phase. In order to verify the proposed framework, a OneWeb constellation was selected as a subject for study. The effectiveness of the two-phase framework was analyzed and proved in each phase. The optimal maintenance strategy design problem is expressed as a non-linear integer programming that utilizes inventory management's (s, Q) policy. The channels that constitute a dual-channel supply chain are divided into a normal and an emergency mode. The effectiveness of the proposed model is established through numerical examples and sensitivity analysis on the maintenance of large satellite constellations. The contributions of this thesis are threefold. First, the optimal deployment and spare replenishment problems are defined and mathematically formulated. Second, a novel solution procedure and the framework to deploy and replenish the satellite mega-constellation are defined and developed. Finally, realistic numerical examples are conducted to demonstrate the effectiveness of the proposed problem and developed framework. The results can provide benchmarks for designing and evaluating the constellation deployment and replenishment architecture in the preliminary phase of the mission operation.