Spectral emissivity calculation based on extreme learning machine

Abstract

Spectral emissivity has important research significance and application value in the fields of national defense and medical science. The network training is carried out on the test samples through the ELM modeling method in this paper, and the results show that the blackbody is band infrared radiation brightness model in the range of 3 ~ 5μm and 8 ∼ 12μm, used to calculate the different wavelengths of blackbody spectral emissivity. The comparison of the standard blackbody shows that the maximum relative error between the spectral emissivity and the actual spectral emissivity of the ELM network model is less than 1%, and the method is applied to the calculation of the target spectral emissivity of the aircraft skin.