In response to the challenging issue of distinguishing the types of internal defects of wind turbine blades in service, this paper proposes the theory and method of dynamic thermal imaging detection based on natural daylight excitation. Finite element simulation and experimental analyses are conducted to investigate the dynamic thermal imaging patterns of different internal defects of wind turbine blades under natural daylight excitation. Firstly, a finite element heat transfer simulation model of wind turbine blade slices is established, and the variation rule of thermal characteristics of two types of typical internal defects, namely, debonding and water, is revealed by numerical calculation under the daylight-excited heat conduction physical field. Secondly, blade slice is homogeneously processed, and a thermal imaging detection platform is constructed using unmanned aerial vehicles for daylight-induced thermal imaging. Finally, daylight-induced thermal imaging experiments are conducted during different times of day under natural daylight conditions. The simulation and experimental results indicate that the two types of typical defects of debonding and water inside the wind turbine blade will lead to different trends in the surface temperature field under daylight excitation, and debonding defects will lead to the dynamic evolution of hightemperature anomalies to low-temperature anomalies in the corresponding areas on the surface of the wind turbine blade, while water is the opposite, which will provide a new methodology for the intelligent operation and maintenance of in-service wind turbine blades.