水库滑坡演化进程多维诊断与稳定性研究

The reservoir landslide is characterized by the information evolution of multi-field with the primary influence of the periodic seepage field variation. Usually, the prediction of landslide stability only depending on the displacement-time curve is not satisfied. Therefore, it is urgent to carry out the study on the multi-dimensional diagnosis of evolution process and stability of reservoir landslide. With the method incorporating the modern geology and information technology, the systematic research on relevance mechanism of multi-dimensional information fusion and monitoring system of reservoir landslide can be performed. The typical landslides in the Three Gorges Reservoir Region are presented as the study cases. The interaction response regularity between seepage field and stress field and its relationship with multi-dimensional information evolution can be examined to reveal the evolution mechanism of reservoir landslide. The multi-dimensional information characteristics and the main control factors during the evolution are investigated, and a new monitoring method of multi-dimensional characteristic variables with multi-sensor technology can be put forward. According to the association rules and evidence theory, the data mining and fusion integration can be conducted. In order to achieve the evolution process diagnosis for reservoir landslide, the generalized model for reservoir landslide can be established, and the corresponding evolution stages threshold value of reservoir landslide can be determined. The multi-dimensional comprehensive monitoring system based on the evolution mechanism and evolution process will be established. Based on the research of the control law and key control factors of characteristic information during the landslide evolution process, the optimization control method for reservoir landslide can be proposed. The research achievements depending on the evolution process can provide the theoretical basis for prediction and control of the reservoir landslide.

水库滑坡具有渗流场周期变化主导的多场信息演化特征，单纯依据位移-时间曲线进行滑坡预测不尽如人意，亟需开展水库滑坡演化进程多维诊断与稳定性研究。本项目应用现代地质学和信息技术等多学科交叉融合的研究方法，系统开展水库滑坡多维信息融合关联机理与监测体系研究。以三峡库区典型滑坡为例，研究水库滑坡应力场与渗流场互响应规律及其与多场信息演变的关联性，揭示水库滑坡演化机理；研究水库滑坡多维信息特征与演化进程中的主控因素，提出基于多传感器技术的多维特征变量监测方法。依据关联规则和证据理论对滑坡多场信息进行数据挖掘与融合，提出水库滑坡演化进程概化模型，确定不同演化阶段阈值，实现水库滑坡演化进程诊断。构建基于水库滑坡演化机理与演化进程的多场信息监测新体系；研究水库滑坡演化进程特征信息的控制律和关键控制因素，提出水库滑坡优化防控方法，实现基于演化进程的水库滑坡预测预报，为水库滑坡预测预报与防控提供理论依据。

本项目紧密围绕“水库滑坡多场演化机理”和“水库滑坡多维信息融合关联机理”关键科学问题，以系统论和信息论为指导，通过现代地质学和信息技术的多学科交叉融合，以三峡库区地质灾害大型野外试验场为基地，基于水库滑坡演化机理，应用多传感器监测技术，研究水库滑坡多维信息特征与采集系统，研发水库滑坡多维数据挖掘和数据融合相集成的信息处理技术，提出水库滑坡多维诊断方法，构建水库滑坡多场信息综合监测新体系，指导水库滑坡科学合理监测。项目的研究目标是，提出基于多传感器技术的水库滑坡多维信息处理与滑坡演化进程多维诊断方法，构建水库滑坡多场信息综合监测新体系，指导水库滑坡科学合理监测。围绕项目研究目标，通过系统研究，圆满实现了上述预期目标。在基于多传感器技术的水库滑坡多维信息处理、水库滑坡演化进程多维诊断方法、水库滑坡多场信息综合监测新体系、水库滑坡演化模式、水库大型综合试验场建立、水库滑坡预测模型等方面取得了创新研究成果。项目系统开展了水库滑坡预测与治理应用。. 项目发表论文55篇，其中SCI收录27篇，EI收录25篇；授权国家发明专利授权15项；出版专著3部；主编地质灾害防治行业标准3部；获湖北省科技进步一等奖1项；主编行业规范3部、参编3部。培养博士后4人、博士研究生16人。