An Optimization Approach for Electrical Power System Supervision and Sizing in Low Earth Orbit Satellites

Optimal management of the electrical power system (EPS) in low Earth orbit satellites requires the design of dedicated algorithms. In this work, a power flow supervisor is proposed by defining decision strategies based on a dynamic EPS model that distinguishes between different operating modes. This approach ensures the primary constraint of load service is met while optimizing battery charging/discharging, and efficiently tracking the maximum power point of photovoltaic panels. The transitions among different modes are governed by exogenous inputs, load power demands, and battery state of charge, providing flexibility and adaptability to varying conditions. In each mode the supervisor determines the duty cycle of the DC-DC buck converter used as the interface between the photovoltaic panels and the battery on the unregulated bus and the current to be absorbed by a bleeder circuit connected in parallel to the battery. The overall design of the EPS supervisor was obtained as the solution of a two-level optimization procedure. Indeed, fine-tuning of the parameters are obtained as the solution of a mathematical problem and then an optimization approach was used to determine the optimal sizing of the battery and photovoltaic panels. The simulation results demonstrated the effectiveness of the optimization strategy.