Genetically grouped atmospheric transmissivity weighted multi-group wide-band k-distribution model for remote infrared imaging

In the prediction of remote infrared imaging for hydrocarbon-fueled vehicles and their exhaust plumes, conventional wide-band k-distribution methods encounter significant challenges due to the pronounced non-isothermal and inhomogeneous between combustion gases and the atmosphere, which severely violate the correlated-k assumption. To address this issue, an atmospheric transmissivity weighted multi-group wide-band k-distribution (ATWMGWB) model is proposed, where the atmospheric transmissivity is incorporated into the emission term of the radiative transfer equation. The deviation of correlated-k characteristics between two thermodynamic states is quantitatively evaluated and adopted as the objective function in a genetic algorithm to optimize the spectral grouping results of the proposed model. A set of 102 1-D cases is used to exhaustively search and optimize the model’s computational parameters based on a defined objective function. Results demonstrate that the optimized ATWMGWB model achieves superior accuracy and computational efficiency compared to the fictitious gas-based statistical narrow-band model and narrow band k-distribution model across the 2-2.5 μm, 3-5 μm, and 8-14 μm bands. In a 3-D nozzle fluid field calculated by large eddy simulation, the relative error of multi-band, multi-angle infrared radiance with respect to line-by-line calculations remains within $\pm 8\text{\%}$. With an imaging time comprising only 4.2% of the total LES computation, the ATWMGWB model was demonstrated strong potential for applications in turbulence–radiation interaction (TRI) research.