通风管路基温度场神经网络预测模型.pdf

第 23 卷第 24 期岩石力学与工程学报 23(24):4131~4136
2004 年 12 月 Chinese Journal of Rock Mechanics and Engineering Dec.,2004



通风管路基温度场神经网络预测模型*

常斌 1 李宁 1、2 苏波 2 李国玉 2 全晓娟 2
(1 西安理工大学岩土工程研究所西安 710048) (2 中国科学院寒区旱区环境与工程研究所冻土工程国家重点实验室兰州 730000)

摘要针对青藏铁路沿线冻土特性,对冻土通风管路基温度场约束空间进行了分析、简化。基于系统通风管路基
温度场数值仿真分析结果,建立了面向工程设计人员的神经网络预测模型,并开发了相应的优化设计软件系统。
该预测模型具有对时间、空间、活动层、布设方案等参数的独立与综合分析能力。同时,对模型的预测效果进行
了检验。
关键词土力学,冻土,通风管路基,神经网络,温度场
分类号 TU 文献标识码 A 文章编号 1000-6915(2004)24-4131-06

ANN PREDICTIVE MODEL FOR PREDICTING TEMPERATURE FIELD
EVOLVEMENT WITHIN VENTIDUCT ROADBED

Chang Bin1,Li Ning1,2,Su Bo2,Li Guoyu2,Quan Xiaojuan2
(1Institute of Geotechnical Engineering,Xi′an University of Technology, Xi′an 710048 China)
(2State Key Laboratory of Frozen Soil Engineering,Cold and Arid Regions Environmental and Engineering Research Institute,
CAS, Lanzhou 730000 China)

Abstract Ventiduct roadbed is an active cooling foundation to resist thaw. Many numerical analyses have been
conducted to study the effectiveness of the ventiduct layout on the frozen soil foundation,with a lot of data on
temperature field evolvement. Obviously,these data are helpful for engineers to optimize the design of ventiduct
roadbed. For the maximum uses of the data,the artificial work (ANN) method is preferable. Thus,an
ANN model needs to be established to predict tem