中国安全科学学报 ›› 2018, Vol. 28 ›› Issue (1): 149-154.doi: 10.16265/j.cnki.issn1003-3033.2018.01.025

• 安全社会工程 • 上一篇    下一篇

装配式建筑安全文明施工费RS-LSSVM预测方法

刘名强1, 李英攀1 副教授, 陈晓2 经济师, 王芳1,3 工程师, 李瑞格1, 李晓喆1   

  1. 1 武汉理工大学 土木工程与建筑学院,湖北 武汉 430070
    2 中建三局绿色产业投资有限公司,湖北 武汉 430040
    3 中建三局第二建设工程有限责任公司,湖北 武汉 430074
  • 收稿日期:2017-10-15 出版日期:2018-01-28 发布日期:2020-09-28
  • 作者简介:刘名强 (1994—),男,江西萍乡人,硕士研究生,研究方向为建设项目管理及其信息化、建筑产业化、PPP模式等。E-mail:liuyangmai@whut.edu.cn。李英攀 (1978—),女,江西上饶人,博士,副教授,硕士生导师,从事建设项目管理及其信息化、建筑产业化、项目投融资管理与PPP模式等方面的研究。E-mail:liyingpan@whut.edu.cn。
  • 基金资助:
    湖北省自然科学基金资助(2013CFB346);中央高校基本科研业务费专项资金资助(WUT:2014-IV-122)。

Prediction of safety-civilized measure cost for fabricated building project based on RS-LSSVM

LIU Mingqiang1, LI Yingpan1, CHENXiao2, WANG Fang1,3, LI Ruige1, LI Xiaozhe1   

  1. 1 School of Civil Engineering and Architecture, Wuhan University of Technology,Wuhan Hubei 430070, China
    2 Green Industry Investment Co., Ltd of China Construction Third Engineering Bureau,Wuhan Hubei 430040, China
    3 The Second Construction Co., Ltd of China Construction Third Engineering Bureau,Wuhan Hubei 430074, China
  • Received:2017-10-15 Online:2018-01-28 Published:2020-09-28

摘要: 为准确测算装配式建筑安全文明施工费,开发一种基于粗糙集(RS)-最小二乘支持向量机(LSSVM)模型的预测方法。根据装配式建筑作业空间并行多维且以吊装施工为主的特点,分析影响费用的主要因素并通过RS属性约简算法确定其测算因子;引入LSSVM,构建装配式建筑项目安全文明施工费测算模型,给出计算方法以及模型流程;以某城市群部分装配式项目的相关数据进行模型学习训练和仿真测算,以此为例完成实证检验和分析。结果表明:在样本数据较少、指标成多维非线性关系的情况下,用该方法测算所得结果与实际情况符合较好(平均相对误差为4.92%),比BP模型和回归分析等2种传统方法(11.78%和17.67%)测算结果更准确,效率更高。

关键词: 装配式建筑, 安全文明施工费, 粗糙集(RS), 最小二乘支持向量机(LSSVM), 预测

Abstract: To get accurate prediction of the safety-civilized measure cost for fabricated building project, a method based on the combination of RS and LSSVM was developed. For developing the method, the measurement factors were ascertained by attribute reduction algorithm in RS according to the features of fabricated building project. A LSSVM was introduced and a RS-LSSVM model was built. The method based on RS-LSSVM was applied to a number of projects in pilot urban agglomeration as an example. The data on these projects were input into the model for training and simulation to verify the method. As the case study shows, under the condition of small-sample case and multi-dimensional nonlinear factors, in comparison with the conventional methods such as the multivariant linear regression and BP neural network method, the proposed method works more efficiently, and can give calculation results more closely conforming to the reality.

Key words: fabricated building, safety-civilized measure cost, rough set (RS), least squares support vector machine(LSSVM), prediction

中图分类号: