According to the latest satellite observations, clouds cover about 67% of the earth's surface and affect the radiation budget of the earth-atmosphere system by scattering and absorbing short-wave radiation from incident sunlight and longwave radiation from the ground. The characteristics of cloud particles at different altitudes are complex and changeable in space and time, the microphysical properties of real clouds are still unknown. As a kind of discrete random medium, cloud inevitably causes background interference to satellite-to-earth link laser communication, satellite remote sensing, etc. Therefore, the research on light scattering and radiation characteristics of clouds is not only helpful for predicting climate change and understanding the radiation budget of the global atmospheric system, but also has important significance for other fields in atmospheric physics. This paper starts with the modeling of the spatial distribution of the cloud, studies the spatial distribution patterns and internal microstructures of clouds, uses the multi-scale superposition algorithm in fractal theory to establish a three-dimensional spatial distribution model of clouds, and controls the parameters of the different types of clouds in the algorithm. The cloud layer can be regarded as an irregular structure composed of water droplets or ice crystal particles of various sizes in a three-dimensional space. The modeling of the spatial distribution of clouds includes two main parts: the modeling of clouds in spatial morphology and the modeling of the clouds in microscopic physical structure. For modeling of the spatial distribution of clouds, a multi-scale overlay algorithm is used. This algorithm has many controllable parameters and flexible control. It can generate various types of clouds according to needs, and can simulate dynamics by increasing the fractal dimension cloud structure. The multiscale superposition algorithm is used to establish the spatial distribution models of different types of clouds.