Infrared temperature measurement system for hot-formed aluminum alloys based on non-Gaussian rough surface and dual-scale corrected Monte Carlo

During the manufacturing and processing of aluminum alloys, temperature is an important indicator that determines product performance. However, the peak–valley skewed microstructure formed during the ring rolling process gives the workpiece surface a non-Gaussian character, resulting in significant deviations in emissivity calculation and background radiation correction in existing infrared temperature measurement models. To address this issue, this paper proposes an infrared temperature measurement system based on non-Gaussian rough surfaces and dual-scale corrected Monte Carlo (NG-PCMC). Firstly, the fast Fourier transform is combined with the Johnson transformation system to construct an aluminum alloy rough surface model that can simultaneously regulate standard deviation, skewness, kurtosis, and autocorrelation length. Then, based on the surface model, the s/p polarization effect is introduced into the Monte Carlo ray tracing framework. And through dual-scale power spectral density decomposition, geometric optics is adopted at low frequencies and perturbation correction is introduced at high frequencies, to precisely calculate the emissivity of rough surfaces. After that, the multi-source background radiation was quantitatively corrected by combining Planck’s law and the view factor method. Finally, in order to verify the effectiveness of the proposed NG-PCMC method, emissivity measurement experiments were carried out on K465 alloy and high-purity iron samples, and temperature measurement experiments were conducted during the hot forming process of aluminum alloys. The experimental results show that this method can not only accurately obtain the emissivity of rough surfaces, but also significantly improve the accuracy and anti-interference performance of infrared temperature measurement.