Applications of Linear Programming Techniques to Satellite Power Management and Scheduling

Abstract

The power system is one of the most important subsystems for a successful space mission. Any failure in this subsystem leads to a direct loss of a satellite. This creates a need for a power schedule to be effectively and efficiently produced, especially if requirements are constantly changing. This paper presents the application of linear programming techniques to solving the power schedule problem, with the more specific usage of mixed-integer linear programming (MILP). The illustration of the approach is applied to a Swedish student satellite, which consists of the necessary subsystems and eight separate experiments. Two programs are developed, one studying the satellite lifetime in terms of orbital cycles and the other studying the individual orbit cycle. Simulating the lifetime of the satellite over 5000 orbit cycles, the battery level did not decline below 76.35%. Using a computer with an Intel i4 processor, this simulation took 3.2 hrs, with individual orbits taking 2.3 s each. Further work includes developing the program to be completed on-aboard the satellite, adapting to new scenarios, and incorporating a model for the decline of battery performance over time.