| When a high-speed vehicle with an optical imaging detection guidance system flies in the atmosphere,a complex flow field is formed between the optical cowl and the incoming flow,causing heat,thermal radiation and image transmission interference to the optical imaging detection guidance system,causing target image drift,jitter and blurring,an effect known as aero-optical effects.The quantitative calculation and analysis of aero-optical effects is an important issue for modern flying vehicles.This paper focuses on the modeling and simulation method of a blunt-headed vehicle,the effect of 0-15° angle of attack of a blunt-headed vehicle on aero-optical imaging deviation and the effect of different altitudes on 0-15° angle of attack of aero-optical imaging deviation,and the construction of an aero-optical imaging deviation prediction model based on optimization algorithm optimization neural network.The main research results of this paper are as follows.The simulation part uses software to model and mesh the head of the vehicle and simulates the head of the vehicle based on computational fluid dynamics using Fluent to obtain the density distribution of the surrounding flow field under different operating conditions of the vehicle.The density distribution around the vehicle is converted into the corresponding refractive index distribution by the Gladstone-Dale formula.The ray transmission equation is solved using the fourth-order Runge-Kutta methods,and the imaging deviation data are obtained using the reverse ray tracing and stopping criterion.The effect of 0-15° angle of attack of the blunt-headed vehicle on the imaging deviation of aero-optics was analyzed,and the following conclusions were obtained:The smaller the angle of attack,the flatter the refractive index distribution on the propagation path.The flatter the refractive index distribution on the light propagation path,the smaller the imaging deviation.In other words,as the angle of attack increases,the imaging deviation increases.0-15 km,the imaging deviation is sensitive to the change from 0-5°.20-25 km,the aero-optical imaging deviation slopes down first and then increases,but the changing trend is not the same.At 20 km,the imaging deviation variation in the 0-5° range is greater than that of 10-15°,however,at 25 km,the imaging deviation variation of 10-15° is greater than that of 0-5°.The effect of different altitudes on the imaging deviation of 0-15° angle of attack pneumatic optics was investigated,and the results showed that the larger the altitude,the flatter the refractive index distribution on the light propagation path and the smaller the imaging deviation.The slope of the imaging deviation increases in the negative direction of 0 with increasing angle of attack and approaches 0 with increasing altitude,which indicates that the imaging deviation appears to change significantly at large angles of attack and low altitudes.To address the problems of slow convergence and the problem of falling into local optima at the late stage of the optimization search,the slime mould algorithm is improved by adding Tent chaos and an improved random opposition-based learning.The improved slime mould algorithm is optimized for Elman neural network.Four performance evaluation metrics are used for comparison,and the results show that the prediction evaluation metrics of the ISMA-Elman model are better than those of the SMA-Elman model and the Elman model. |
