小样本、贫信息下民用飞机费用估算模型及算法

发布时间：2018-05-19 23:17

本文选题：费用估算 + 估算模型　；参考：《系统仿真学报》2014年03期

【摘要】：费用估算是民用飞机成本管理的核心内容,直接关系到民用飞机项目的成败。针对民用飞机费用估算问题展开研究,把GM(0,N)模型与BP神经网络算法相结合,应用BP神经网络对建立的GM(O,N)模型估计值进行再次优化,二者的组合模型能有效提高估算精度。模型构建步骤如下:首先构建原始GM(0,N)模型并根据该模型求出样本的费用估算值;以GM(0,N)模型的参数变量、样本估计值为输入,实际值为期望输出建立BP神经网络并在mat lab里训练;利用训练后的组合模型对民用飞机的费用进行估算。以民用飞机费用估算为仿真实例,分别使用多元线性回归、传统的GM(O,N)模型,和该方法建立费用估算模型,对目标机种的费用做出估算,分析估算结果,表明该模型具有更好的拟合与估算效果。
[Abstract]:Cost estimation is the core of civil aircraft cost management and directly related to the success or failure of civil aircraft projects. In this paper, the cost estimation of civil aircraft is studied. The GM0N) model is combined with the BP neural network algorithm, and the BP neural network is applied to optimize the estimated value of the established GM0 / ONM model again. The combined model of the two models can effectively improve the estimation accuracy. The steps of constructing the model are as follows: firstly, the original GM0N) model is constructed and the estimated cost value of the sample is obtained according to the model; the parameter variable of the GM0N) model, the estimated value of the sample is input, the actual value is the expected output, and the BP neural network is established and trained in mat lab. The cost of civil aircraft is estimated by the combined model after training. Taking the cost estimation of civil aircraft as an example, the multivariate linear regression model and the traditional GMN model are used respectively, and the cost estimation model is established by this method. The cost of the target aircraft is estimated, and the estimated results are analyzed. It shows that the model has better fitting and estimating effect.
【作者单位】：南京航空航天大学经济与管理学院;
【基金】：国家自然科学基金项目(70901041,71171113) 教育部博士点基金项目(20093218120032)
【分类号】：F560.5;F224

【参考文献】