改进BP网络在装备维修费用预测中的应用

Application of Improved BP Neural Network in Forecasting Equipment Maintenance Cost

提出了一种BP神经网络理论的改进算法,在此基础上建立了时间序列对象的预测模型,根据预测模型,对某型装备使用维修费用进行预测,并与实际使用维修费用进行了对比。结果表明此模型准确,并具有应用泛化能力,该算法为费用预测提供了一条新途径。

The improved BP neural network was put forward in this article, on this basis, a forecasting model of time sequence object was founded. A equipment maintenance cost has been studied by using this model, and the calculated results were compared with the practical cost. It showed that the model is correct, and has extensive application. The method provides a new way to forecast equipment maintenance cost.