基于极限学习机算法的城建项目工程造价预测方法  

A Cost Prediction Method for Urban Construction Projects Based on Extreme Learning Machine Algorithm

作　　者：聂霞 NIE Xia(Xingyi City Planning and Design Institute,Xingyi Guizhou 562400,China)

机构地区：[1]兴义市规划设计院,贵州兴义562400

出　　处：《信息与电脑》2023年第21期29-31,共3页Information & Computer

摘　　要：城建项目工程的初步投资估算,在项目方案的比选和投资控制中起着至关重要的作用。城建项目决策阶段因信息量有限,对建设项目的工程造价进行预测十分必要。基于此,研究基于极限学习机算法的城建项目工程造价预测方法,首先在公开数据库中选择城建项目工程造价数据,采用缩放处理方法描述数据特征,其次按照不同的数据特征进行统一化处理,最后基于极限学习机算法构建网络模型,预测不同类型的城建项目工程造价,从而完成方法设计。以不同工程项目为测试对象的实验结果表明,新方法能够实现对不同工程单方造价的有效预测,具有应用价值。The preliminary investment estimation of urban construction projects plays a crucial role in the comparison and selection of project plans and investment control.The amount of information in the decision-making stage of urban construction projects is limited,and it is necessary to predict the engineering cost of construction projects.Therefore,a method for predicting the engineering cost of urban construction projects based on extreme learning machine algorithm is studied.Select the cost data of urban construction projects in the public database and use the scaling method to describe the data characteristics;Unify processing according to different data features,construct a network model based on extreme learning machine algorithm,predict the cost of different types of urban construction projects,and complete method design.The experimental results indicate that the new method can effectively predict the unit cost of different engineering projects using different engineering projects as test objects,with specific application value.

关 键 词：极限学习机 城建项目 工程造价 网络模型 

分 类 号：TM72[电气工程—电力系统及自动化]