基于GA-BP的敌空袭兵力出动量预测  

Forecasting of air-raid weapon quantity of the opposed side based on GA-BP Algorithm

作　　者：路建伟[1] 唐松洁[1] 胡建辉[1] 刘己斌[1] 

机构地区：[1]防空兵指挥学院,郑州450052

出　　处：《电光与控制》2007年第1期120-123,共4页Electronics Optics & Control

摘　　要：在防空兵力资源有限的情况下,为取得最大的对敌作战效能,有必要对敌空袭兵力出动量进行预测。在遗传算法与误差反向传播神经网络结构模型相结合的基础上,设计了用遗传算法训练神经网络权重的新方法,并对敌对地面目标群空袭的兵力出动量进行预测,同时与BP算法和灰色系统理论模型进行了比较。经检验,计算值与实际值接近,并优于BP算法和灰色理论模型,具有良好的预测效果。In order to acquire optimum operational efficiency when the air - defence force resources are limited, it is needed to forecast the quantity of the opposed side used in air - raid operation. We put forward a new method for training the weights of neural network by genetic algorithm based on the combination of genetic algorithm（GA） and back propagation（BP） neural network models. We used it for forecasting the quantity of the weapons enemy may used for attacking ground object groups. Then we compared the method with the back propagation algorithm and gray system theory models. The result showed that the value calculated out was close to the real value, and the method proposed here is better than BP algorithm and gray system theory model.

关 键 词：地面目标群 出动量 空袭兵器 遗传算法 BP神经网络 预测 

分 类 号：V271.4[航空宇航科学与技术—飞行器设计]