Missile thermal emission flow field under the synergistic effect of deep reinforcement learning and wireless sensor network

Abstract

AbstractThe vehicle-mounted missile vertical thermal emission not only has excellent maneuverability, randomness, and concealment, but also has a short response time, good versatility, and high reliability, so it is widely used. In order to explore the gas flow field of thermal emission of the vehicle-mounted missile, firstly, this study introduces the theoretical basis of deep reinforcement learning (DRL) algorithm and wireless sensor network (WSN). Secondly, combining WSN and DRL, a WSN technology based on DRL is proposed. Finally, the DRL-based WSN technology is applied to the vertical thermal emission of the vehicle-mounted missile. In addition, the flow field simulation software is employed to simulate and compare the influence of the single-side and double-side diversion schemes of the gas flow discharged by the diverter on the open ground field flat collection launcher, and the diversion characteristics of the two schemes are obtained. The results show that in the single-side diversion scheme, the impact and ablation area of the gas jet on the ground mainly appear at the rear side of the diversion device, and the ablation area of the gas jet on the launcher vehicle is mainly at its tail end. While the ablative site and shock of the double-side diversion scheme on the ground mainly appear on both sides of the diversion, and the ablative part of the gas jet to the launcher is mainly present at the bottom of the frame and the inside surface of the tire. The study of missile thermal emission flow fields based on DRL and WSN has certain theoretical significance for the flow field variation of missile launching.