基于GA-BPNN的铁路站房结构工程投资估测研究

摘要
图/表
参考文献
相关文章 (2)

全文: PDF (3257 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要研究目的:近年来，随着我国客运专线、城际铁路的迅速发展，一大批现代化的铁路客站也已建成或者在建中。为更好适应今后铁路客站发展趋势及客站建设工作需要，进一步加强铁路客站工程投资控制方面的研究，对新建铁路客站站房工程投资的估测进行研究是十分有必要的。

研究结论:本文通过对铁路客站站房结构工程投资的影响因素及估测模型进行研究，得到以下结论: (1)基于收集到的大量铁路客站站房结构及工程投资基础数据，分析得到铁路客站站房结构工程投资的主要影响因素为站房面积、基底面积、站房建筑高度、抗震设防烈度、楼盖柱的结构高度、楼盖结构投影面积，并将其作为输人向量建立了BP神经网络估测模型;(2)应用GA遗传算法对BP神经网络模型的初始权值阈值进行了优化，通过估测结果对比，估测模型的精度及泛化能力都得到了显著提高;(3)通过工程实例对模型进行验证，估测模型得到的结果同概算值相对误差为1.19%，符合工程项目前期估算的精度要求，进一步验证了铁路客站站房结构工程投资估测模型的工程实用性;(4)本文研究内容可为相关研究领域提供一定的借鉴，同时对铁路客站站房结构工程投资控制研究具有一定的参考意义。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	魏庆朝
	闫松涛
	时瑾
	杨娜

关键词 ：铁路站房, 工程投资, 估测研究, GA-BPNN

Abstract：Research purposes:Nowadays，as the fast development of china’s high speed railway，a large number of railway-stations are constructing. To meet the developing trend of railway station，and reinforce the theoretical research of the investment of the railway-station construction，there is an urgent demand for carrying on the research of the estimation of the investment of railway-station buildings.

Research conclusions:In this paper, we studied the influence factors and the estimating model of the investment of railway-station buildings，and got the following conclusions:(1)Based on the collected data of railway-station buildings，the main factors were obtained. BPNN model was built with the main factors as the input vector. (2)Genetic algorithm was used to optimize the initial weight and bias of BPNN model, the precision and generalization ability of the model was better through the comparison. (3)The practicability of GA-BPNN model was verified with the example of a new project. (4) The research of this paper provided the references to the similar research and the investment control of the railway-station buildings.

Key words： railway-station buildings construction investment estimating research GA-BPNN

收稿日期: 2014-01-15

PACS:

U291. 6

基金资助:铁道部科技研究开发计划项目(20106009 - F)

作者简介: 魏庆朝，1957年出生，男，教授

引用本文:

魏庆朝,闫松涛,时瑾,杨娜. 基于GA-BPNN的铁路站房结构工程投资估测研究[J]. 铁道工程学报, 2014, 31(5): 95-100.
WEI Qing-chao，YAN Song-tao，SHI Jin, YANG Na. Research on the Estimation of the Investment of Railway-station Buildings Based on GA-BPNN. Journal of Railway Engineering Society, 2014, 31(5): 95-100.

链接本文:

http://tdgcxb.crec.cn/CN/ 或 http://tdgcxb.crec.cn/CN/Y2014/V31/I5/95

[1 ]	段晓晨，余建星，等.基于CS, WLC, BPNN理论预测铁路工程造价的方法[J].铁道学报，2006(8) ;117-122.
[2]	Stoy Christian，Schalcher Hans - rudolf. Residential Building Projects :Building Cost and Indicators and Drivers [ J ]. Construction Engineering and Management. 2007(2) ：139.
[5]	翟宜峰，等.用遗传算法优化神经网络初始权重的方法 [J].吉林大学学报(工学版)，2003 ( 2 ) :45 - 50.
[6]	王述红，郝哲.岩体微裂隙注浆量预测分析的遗传神经网络方法[J].岩土工程学报，2001(5);572-575.
[7 ]	Hill T Marquezl，Connor M O，etc. Artificial Neural Network Model for Forecasting and Decision Making [J]. International Journal of Forecasting，1994(10)： 28-31.
[3]	Douglas J Ferry, Peter S Brandon，Jonathan D Ferry. Cost Planning of Buildings [ M ]. Landon ： Blackwell Science Ltd，1999.
	Wang Shuhong，Hao Zhe. The Genetic Algorithm - neural Network Method to Forecast the Miniature Crack Grouting in Rock Matrix[ J]. Chinese Journal of Geotechnical Engineering，2001 (5) ：572 -575.
	Duan Xiaochen，Yu Jianxing，etc. A Method of Estimating WLC of Scheduled Railway Projects Based on CS，WLC and BPNN Theorems [ J ]. Journal of the China Railway Society,2006(8) :117 -122.
	Zhai Yifeng， etc. Method for Optimizing Initial Weights of ANNs by Gas [ J ]. Journal of Jilin University (Engineering and Technology Edition)，2003 (2):45 -50.
[4 ]	Stoy Christian，Pollalis Spiro，Schalcher Hans - rudolf. Drivers for Cost Estimating in Early Design ： Case Study of Resi - dential Construction [ J ]. Constr - uction Engineering and Management，2008(1) :32.