【作者】 冯夏庭； 张治强； 王泳嘉；

【Author】 Feng Xiating~(1,2),Zhang Zhiqiang~2,Wang Yongjia~2 (1 Institute of Rock and Soil Mechanics,The Chinese Academy of Sciences,Wuhan 430071 China)(2 Northeastern University,Shenyang 110006 China)

【机构】 中国科学院岩土力学研究所LRSM开放研究实验室； 东北大学资源与土木工程学院；

【摘要】 针对复杂条件下岩石工程问题,讨论的三种智能岩石力学分析方法:以基于工程经验和现场监测数据的知识自学习来建立分析预报模型的方法;从力学观点的出发识别具有开挖卸荷效应的岩体参数方法,以及从非线性动力学系统的动态演化的观点出发,采用进化力学分析方法对系统的未来变形特性进行识别的方法。给出大深度开采诱发岩爆的风险估计结果和三峡永久船闸区岩体开挖卸荷效应和未来一段时间内变形行为识别的结果。更多还原

【Abstract】 Three kinds of modeling methods are proposed for solving extremely complex problems.One is that knowledge learned from a lot of engineering cases and monitoring data and used to build an intelligent model.The second is to recognize unloading effect generated by excavation of rock mass from view of mechanics analysis.The third starts from view of nonlinear dynamic evolution.The behaviors of rock mass at future can be estimated by the model obtained by using evolutionary-neural-network algorithm.As examples,the results of assessing rockburst risks induced by mining at great depths and excavation unloading effect and its future deformation behaviors of permanent shiplock at Three Gorges Project are respectively given for each method.更多还原