Solar Array Power Prediction for Near-Space Vehicles Based on a Hybrid Method

Abstract

Compared with other energy sources in near-space, solar energy is more renewable, stable and accessible, which makes solar cells the main source of energy for long-endurance near-space vehicles. However, due to the limited solar cells laying area and complex flight conditions, near-space vehicles are facing tense energy supply situation during flight. Therefore, fast and accurately online power prediction of the solar array, which has important significance for power management, efficient use and reliable operation of the energy system, is the key point to solving the situation at present. In this paper, a hybrid online power prediction method combined particle filter and power calculation model is proposed. Firstly, a comprehensive solar cell array power calculation model for the near-space vehicle is established. Then, the particle filter algorithm is used to estimate optimally the important parameters in the power calculation model with the flight data, to realize the dynamic, rapid and accurate prediction of the power generation of the vehicle during flight. Finally, the data obtained from flight tests of the stratospheric aerostat and high-altitude scientific balloon are verified and compared with the other four models. The results demonstrate that the generation power prediction method proposed in this paper has higher prediction accuracy, which will be a good guiding significance for the actual flight of different types of near-space vehicles.