长江科学院院报 ›› 2018, Vol. 35 ›› Issue (7): 94-99.DOI: 10.11988/ckyyb.20161228

• 岩土工程 • 上一篇    下一篇

岩溶区公路隧道围岩分级专家系统设计与应用

江杰1a,1b,2, 蒲鸥1a, 欧孝夺1a,1b, 赵建刚3, 谢规球3   

  1. 1.广西大学 a.土木建筑工程学院; b.工程防灾与结构安全重点实验室,南宁 530004;
    2.桂林理工大学广西岩土力学与工程重点实验室,广西 桂林 541004;
    3.中交第四公路工程局有限公司,北京 100022
  • 收稿日期:2016-11-23 出版日期:2018-07-01 发布日期:2018-07-12
  • 通讯作者: 欧孝夺(1970-),男,广西来宾人,教授,博士,博士生导师,主要从事岩土地下工程的教学与科学研究。E-mail:ouxiaoduo@163.com
  • 作者简介:江 杰(1979-),男,湖北麻城人,研究员,博士,硕士生导师,主要从事岩土地下工程的教学与科学研究。E-mail:jie_jiang001@126.com
  • 基金资助:
    国家自然科学基金项目(51568006, 51768006);中国博士后科学基金项目(2017M612865);广西岩土力学与工程重点实验室资助课题(14-KF-03)

Expert System of Surrounding Rock Classification for Highway Tunnel in Karst Region: Design and Application

JIANG Jie1, 2, 3, PU Ou1, OU Xiao-duo1,2, ZHAO Jian-gang4, XIE Gui-qiu4   

  1. 1.College of Civil Engineering and Architecture, Guangxi University,Nanning 530004, China;
    2.Key Laboratory of Disaster Prevention and Structural Safety, Guangxi University,Nanning 530004, China;
    3.Guangxi Provincial Key Laboratory of Geomechanics and Geotechnical Engineering, Guilin University ofTechnology, Guilin 541004, China; 4.CCCC Fourth Highway Engineering Co.,Ltd.,Beijing 100022, China
  • Received:2016-11-23 Online:2018-07-01 Published:2018-07-12

摘要: 岩溶地质条件对公路隧道建设影响较大。隧道在建设过程中围岩分级除考虑常用指标外,还须考虑岩溶影响因素指标。以广西岩溶区实际工程为背景,提出了考虑多因素的岩溶状态指标分析方法,建立以岩石坚硬程度、岩体完整程度、地下水状态、结构面产状和岩溶状态5个指标的围岩分级模型。根据MatLab和C++的软件平台设计,植入分级指标参数归一化算法,结合广西瑶寨等3个隧道工程,建立了神经网络专家知识库。最后对专家系统软件进行工程应用,软件判别与实际勘察判别一致性达到80.89%,工程应用效果良好。

关键词: 岩溶, 公路隧道, 围岩分级, 专家系统, 神经网络

Abstract: As highway tunnel construction is greatly affected by karst geological condition, karst indexes should be considered in addition to conventional indicators in surrounding rock classification. In this research, a model of surrounding rock classification was set up, in which rock hardness, rock mass integrity, groundwater status, structural plane occurrence and karst status are included. The expert system based on three tunnel projects in Guangxi Province for surrounding rock classification was designed by embedding normalization algorithm for the indexes using MatLab and C++. The expert system software was applied to engineering and the consistence between software results and actual survey reached 80.89%.

Key words: karst, highway tunnel, classification of surrounding rock, expert system, neural network

中图分类号: