基于极限学习机的建筑工程成本预测  

Construction Cost Forecast Based on Extreme Learning Machine

作　　者：凌飞[1] LING Fei(School of Civil Engineering,Shaanxi Polytechnic Institute,Xianyang 712000,China)

机构地区：[1]陕西工业职业技术学院,土木工程学院,陕西咸阳712000

出　　处：《微型电脑应用》2022年第10期157-160,共4页Microcomputer Applications

摘　　要：建筑工程成本影响因素之间存在复杂的非线性关系,传统建筑工程成本预测方法存在精度低的问题,为此提出基于极限学习机的建筑工程成本预测方法。收集建筑工程成本的历史数据,采用极限学习机对历史数据进行拟合和挖掘,建立建筑工程成本预测模型,并引入粒子群算法搜索极限学习机方法的输入权值和偏置值,进行建筑工程成本预测的实例分析。实验结果表明,极限学习可以描述建筑工程成本变化特点,预测误差很低,获得理想的建筑工程成本预测结果。There is a complex non-linear relationship between the influencing factors of construction engineering cost,and the traditional construction engineering cost prediction method has the problem of low accuracy.Therefore,a construction engineering cost prediction method based on extreme learning machine is proposed.The historical data of construction engineering cost is collected.Extreme learning machine is used to fit and mine the historical data to establish the prediction model of construction engineering cost,and particle swarm optimization algorithm is introduced to search and optimize the input weight and bias value of extreme learning machine method.An example of construction engineering cost prediction is analyzed.The experimental results show that extreme learning can describe the change characteristics of construction engineering cost,the prediction error is very low,and the ideal prediction result of construction engineering cost is obtained.

关 键 词：极限学习机 建筑工程 成本预测 粒子群算法 全局搜索 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]