用神经网络组合预测法估算反舰导弹研制费用  

Estimating the development expenses of anti-ship missiles in a joint forecasting way of artificial neural network

作　　者：姜晨[1] 徐宗昌[1] 肖国军[2] 

机构地区：[1]装甲兵工程学院,北京100072 [2]海军航空工程学院,山东烟台264001

出　　处：《系统工程与电子技术》2004年第3期348-349,372,共3页Systems Engineering and Electronics

摘　　要：运用BP神经网络组合预测法对反舰导弹研制费用进行估算。首先利用回归分析法建立三元线性回归模型对导弹研制费用进行估算,然后利用3层前馈型BP神经网络对导弹研制费用进行估算,最后利用人工神经网络中的BP网络对所得结果进行组合预测。计算实例表明,使用该组合预测方法所得的预测结果比单使用神经网络或回归分析方法所得结果的总体误差要小,因而该方法是可行而有效的。A joint forecasting way of artificial neural network (ANN) is utilized to estimate the development expenses of anti-ship missiles. Firstly the ANN and regressive analysis are separately used to estimate the development expenses of the anti-ship missile. Then the BP neural network is employed to forecast the expense based on the above two estimating results. The result shows that the total forecasting error of this method is smaller than ANN or regressive analysis alone. So the method is effective and feasible.

关 键 词：反舰导弹 神经网络 回归分析 组合预测 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]