基于神经网络预测模型的再入落点覆盖区快速计算方法

Rapid generation of re-entry landing footprint based on neural network predictive model

针对高超声速飞行器的再入落点覆盖区计算问题,提出了一种基于神经网络预测模型的快速计算方法。首先,在虚拟目标点法框架的基础上构建训练了快速预测覆盖区边界的神经网络模型,通过逼近给定再入条件下覆盖区边界横、纵程间存在的非线性关系实现对覆盖区边界的快速预测。然后,由快速预测得到的覆盖区边界对虚拟目标点法寻优参数的初值进行优化,使其分布在最优解附近以缩短对最优解的搜索时间。最后,进行了仿真验证。仿真结果表明,所提方法在保证最优性的前提下大幅提升了计算效率,具备良好的可行性与高效性。

To address the issue of the computation of re-entry landing footprint for hypersonic vehicles,this paper proposed a rapid computation method based on a neural network prediction model.Firstly,on the basis of virtual target method,a neural network model was constructed and trained to predict the landing footprint quickly.This model approximated the nonlinear relationship between the down-range and cross-range of the footprint boundary under given re-entry conditions,thereby realizing fast prediction of the landing footprint.Subsequently,the predicted footprint boundary obtained from the fast prediction neural network model was used to optimize the initial values of the virtual target method’s search parameters,enabling them to be distributed near the optimal solution and thereby shortening the search time for the optimal solution.Finally,the proposed method was validated through simulations.The simulation results show that the method significantly improves computational efficiency while maintaining optimality,indicating its feasibility and efficiency.