基于SVM和LS-SVM的住宅工程造价预测研究  

Forecasting the costs of residential construction based on support vector machine and least squares-support vector machine

作　　者：秦中伏[1] 雷小龙[1] 翟东[1] 金灵志[2] 

机构地区：[1]浙江大学建筑工程学院,浙江杭州310058 [2]杭州市发展规划研究院,浙江杭州310006

出　　处：《浙江大学学报（理学版）》2016年第3期357-363,共7页Journal of Zhejiang University（Science Edition）

基　　金：国网浙江省电力公司经济技术研究院资助项目(12-513205-007;名称:输电线路工程造价预测快速实现)

摘　　要：为在方案设计初期与工程造价相关信息很少的条件下,准确快速地预测住宅工程造价,在分析既往相关理论和方法优劣的基础上,选取支持向量机构建住宅工程造价预测模型,并通过主成分分析对原始数据进行降噪处理.选取住宅工程造价预测指标集与样本,对输入指标的数据进行主成分分析,消除指标相关性的同时对原始数据降维,将处理后的数据分别导入到"标准支持向量机"和"最小二乘支持向量机"模型中进行训练和预测,并对预测结果进行对比分析,选取较为合理的预测模型,通过参数寻优进一步优化预测效果.所构建预测模型的相对误差均控制在±7%以内,预测精度较高,结果稳定.To forecast the costs of a residential construction rapidly and accurately at the initial stage of construction that lacks relevant information,in view of the strengths and weaknesses of previous approaches,we choose support vector machine（SVM）and principal component analysis（PCA）.Firstly,a residential project cost forecasting index set is selected;The data of the input index is then analyzed and the correlation is eliminated by PCA;Thirdly,the processed data are imported into the standard support vector machine and trained by the least squares support vector machine model.The prediction results are compared and analyzed,and then a more reasonable prediction model is adopted;Finally,the prediction result of the model is optimized by model parameter optimization.Experiments show that the relative error of the prediction model is controlled within ±7%,and the result is stable.

关 键 词：造价预测 主成分分析 支持向量机 最小二乘支持向量机 

分 类 号：TU9[建筑科学]