基于改进支持向量机的工程造价预测模型  

Engineering Cost Prediction Model Based on Improved Support Vector Machine

作　　者：朱琳[1] 刘春[1] ZHU Lin;LIU Chun(Network Management Center,Sichuan College of Architectural Technology,Deyang 618000)

机构地区：[1]四川建筑职业技术学院网络管理中心

出　　处：《计算机与数字工程》2019年第12期3209-3213,共5页Computer & Digital Engineering

基　　金：四川建筑职业技术学院2015年院级科研项目“建筑工程造价的非线性建模与预测研究”(编号：2015KJ07)资助

摘　　要：工程造价预测是当前工程管理领域研究中的热点,针对当前工程造价预测模型存在的预测精度低、建模效率低等不足,提出了基于改进支持向量机的工程造价预测模型。首先收集工程造价历史数据,并对它们进行一定的预处理,然后采用改进支持向量机对工程造价建模的训练样本进行学习,并采用粒子群算法确定模型的参数,从而建立工程造价的预测模型,最后采用Matlab 2014R工具箱实现了工程造价预测的仿真对比实验,结果表明,改进支持向量机大幅度提高了工程造价的预测精度,而且工程造价整体预测性能要明显优于对比模型,具有更高的实际应用价值。The prediction of project cost is a hot topic in the field of engineering management.In view of the low prediction accuracy and low modeling efficiency of current engineering cost prediction model,a prediction model of engineering cost based on improved support vector machine is proposed.The first collection of historical data and the project cost,they must be preprocessed,and then the improved least squares support vector machine is used to the project cost modeling of training samples,and by using particle swarm algorithm,the parameters of the model are determined,so as to establish a prediction model of project cost,finally it is realized by the comparative experiments,simulation in project cost prediction Matlab toolbox of 2014 R.The results show that the improved support vector machine greatly improves the accuracy of prediction of engineering cost,engineering cost and overall prediction performance are significantly better than the contrast model,which has higher practical value.

关 键 词：工程造价 预测模型 改进支持向量机 粒子群优化算法 

分 类 号：U415.1[交通运输工程—道路与铁道工程]