During on-orbit operation, thermal control coatings on spacecraft surfaces are exposed to space radiation, causing a gradual increase in solar absorptance and absorbed heat flux. To address this issue, this study investigates the radiation transfer mechanism, on-orbit sensor monitoring, and data-driven prediction of coating degradation. A Monte Carlo method is employed to analyze the radiative transfer characteristics of coatings at both mesoscopic and macroscopic scales. An on-orbit monitoring device was developed to acquire temperature telemetry data from eight types of coatings, including S781, over 31 months. Using these data, the solar absorptance degradation curve of S781 was obtained through inversion, showing good agreement with literature data. A sliding window-based dynamic parameter identification method is proposed to enable continuous tracking and adaptive updating of degradation rate coefficients. Validation using real on-orbit data demonstrates that the method can identify early-stage data-driven fitting variations and achieve stable fitting after adaptive refitting. This study demonstrates the feasibility of integrated sensing, measurement, and intelligent data analytics for on-orbit coating degradation monitoring.

thermal control coatings,radiation characterization,Monte Carlo method,prediction model,degradation analysis