Satellite Battery Lifetime Extension via Scheduling of Onboard Computing and Energy Harvesting

With advances in onboard processing, satellites are becoming more capable of solving complex tasks. This added capability increases computational load and energy consumption. We propose strategies to extend the battery lifetime of satellites by tailoring energy consumption and energy harvesting to two key constraints: computational workload and sunlight-eclipse pattern. Satellites operate in periodic sunlight and eclipse cycles: during sunlight, they harvest solar energy to charge their batteries; during eclipse, they rely solely on stored energy. Repeated charging and discharging accelerates battery degradation, progressively reducing capacity over time. Once degradation surpasses a critical threshold, satellite operation becomes unreliable. To model and mitigate battery aging, we adopt a degradation model based on therainflow counting algorithm. The model accounts for the satellite’s sunlight and eclipse pattern to predict battery degradation over charge and discharge cycles more accurately. We formulate an optimization problem to minimize battery degradation by scheduling computational tasks across sunlight and eclipse phases, and by determining the appropriate amount of energy to harvest for battery charging. The proposed scheme reduces battery degradation by a factor of three, thereby extending the operational life of the satellite.