Coating structures with dynamically adjustable infrared emissivity are crucial in spacecraft components to cope with the transient thermal environments of space. For a long time, thermochromic phase change materials have been widely used in applications requiring emissivity adjustment, and optimizing the range of adjustable infrared emissivity has always been at the forefront of research. However, reducing the absorption of solar radiation has significant implications for the practical application and thermal stability of spacecraft components in space environments. In this paper, we propose a multilayer film structure based on the phase change material VO₂ combined with the materials ZnSe and ITO to achieve low solar radiation absorption and adjustable infrared emissivity for intelligent thermal radiators in space. Through finite element simulation analysis of the structure, we achieve a solar radiation absorption rate of 0.3 and an adjustable infrared emissivity of 0.49. According to Stefan–Boltzmann’s law, the structure exhibits strong radiative heat dissipation at high temperatures and weak energy dissipation at low temperatures to maintain the thermal stability of the device and ensure efficient operation. The intelligent thermal radiator operates based on the principles of Fabry–Perot resonance. Therefore, the multilayer structure based on the phase change material VO₂ demonstrates excellent performance in both solar radiation absorption and adjustable infrared emissivity, showcasing its tremendous potential in the field of intelligent thermal control in aerospace.

Video super-resolution methods based on convolutional kernels or optical flow often face challenges such as limited utilization of multi-frame detail information or strong reliance on accurate optical flow estimation. To address these issues, this paper proposes a novel super-resolution reconstruction network named Dual Motion Compensation and Multi-scale Structure–Texture Prior (DCST-Net). Dual motion compensation performs direct and progressive motion mapping in parallel, effectively mitigating estimation bias in motion modeling. A multi-scale structure–texture prior is introduced to enhance high-frequency details through feature fusion, alleviating over-smoothing caused by warping and fusion processes. The proposed DCST-Net method is validated on datasets containing both large and small targets, demonstrating its effectiveness and robustness.