基于空域划分的超视距空战态势威胁评估

Modeling air combat situation assessment based on combat area division

编队超视距空战(BVR,Beyond Visual Range)已成为现代空战的主要模式.在空战优势区域与劣势区域判断的基础上对整个空域进行划分,并给出4种特定空域态势.从空中态势和编队作战能力两方面对空战态势进行分析.使用主成分分析法选取输入变量分析编队作战能力,降低评估过程中收集数据的复杂度.应用遗传神经网络对影响BVR各因素进行效能评估,将遗传算法(GA,Genetic Algorithms)与多层前馈(BP,Back Propagation)网络结合,利用GA的全局搜索优化BP网络的结构参数,有效克服BP算法的局部收敛等问题.结果表明:该模型能在综合分析空战各指标后给出红蓝双发的态势评估指标,该模型可有效减少评估中的人为因素,使评估结果更为客观可信.

The beyond visual range（BVR） air combat has become one of the most important air modes of modern air combat.The whole airspace division was made base on advantages and disadvantages of regional area.Four specific airspace situations were put forward.A new model was set up combined situation assessment model and formation combat capacity model.Using principal component analysis（PCA） to select input variables of formation combat capacity model,which can reduce the complexity of collecting data.Combined neural network was used for effectiveness evaluation of BVR.Combine genetic algorithms（GA） with back propagation（BP） neural network,using GA＇s global to search optimized BP network structure parameters,overcome the local convergence and other issues of BP algorithm effectively.The result shows that the model can limit the artificial factors,making the solution more objective and creditable.