Due to the influence of the external environment, the satellite's surface is often irregularly folded, and these folds have some influences on the optical properties. So folds need to be taken into account in the modeling of the optical characteristics of space objects. However, a large number of surface cells in fold surface will lead to a dramatic increase in computational complexity. This paper considers pleats as a kind of ‘material’. A method based on macroscopic optical scattering cross section measurement is proposed to obtain the bidirectional reflectance distribution function (BRDF) data of the fold material. Furthermore, the error back propagation neural network is used to establish BRDF model of fold material, which replaces the complex modeling process of folds and greatly simplifies the calculation. The problem of poor real-time performance is solved under the acceptable accuracy. By combining the experiment with the simulation to compare the BRDF model designed in this paper with the traditional BRDF model, it is verified that the error of the model designed in this paper is much smaller than that of the traditional model.