Self-Adaptive Energy-Aware Power Management for CubeSat Electrical Power Systems Under Dynamic Orbital Conditions

The operational longevity of CubeSat missions is frequently compromised by the tight coupling between limited energy harvesting, battery degradation, and thermal instability. To address these conflicting constraints, this study proposes a robust Energy Management System (EMS) that synergizes Model Predictive Control (MPC) with a bio-inspired Sparrow Search Algorithm (SSA). Unlike conventional static schedulers, the proposed SSA-MPC framework dynamically optimizes control parameters to navigate the non-convex landscape of orbital power flows, explicitly integrating non-linear battery aging models and radiative thermal dynamics into the decision process. Comparative simulations under realistic LEO conditions demonstrate that the SSA-MPC strategy significantly outperforms baseline architectures. Specifically, the proposed method increases critical load availability from 90.5% to 99.9% while virtually eliminating load shedding events. Furthermore, the strategy enhances system sustainability by reducing residual energy waste by 17%, lowering peak internal temperatures by 2°C, and improving end-of-mission battery State-of-Health. Validated through extensive Monte Carlo robustness analyses, this framework offers a resilient, computationally efficient solution for autonomous energy regulation in next-generation nanosatellites, ensuring reliable performance even under stochastic environmental uncertainties.