摘要
信息技术的发展对建筑行业产生了深远的影响。在工程造价管理领域,快速、准确预测工程项目造价总额能够为建设单位招投标决策提供科学的数据支撑。本文将BP神经网络与工程造价预测相结合,将工程造价中各类型工程特征量化,转化为看可输入BP神经网络预测分析模型的训练样本,通过确定参数和节点转换函数,构建了基于BP神经网络的建筑工程造价预测模型。经案例验证,本模型造价估算值与实际值误差仅为1.14%,能够满足工程造价预测应用需求。
The development of information technology has a far-reaching impact on the construction industry.In the field of project cost management,rapid and accurate prediction of total project cost can provide scientific data support for the bidding decision of the construction unit.In this paper,BP neural network is combined with engineering cost prediction,and various engineering characteristics in engineering cost are quantified and converted into training samples that can be input into BP neural network prediction and analysis model.By determining parameters and node conversion function,a construction engineering cost prediction model based on BP neural network is constructed.The case study proves that the deviation between the estimated cost and the actual value of this model is only 1.14%,which can meet the application demand of project cost prediction.
作者
滕凌云
Teng Lingyun(Fujian Jianlan Construction Technology Co.,Ltd.,Fujian 350400)
出处
《住宅产业》
2020年第12期110-113,共4页
Housing Industry
关键词
建筑工程
造价预测
BP神经网络
construction engineering
cost prediction
BP neural network
