【作者】 周勇；

【导师】 柳建新；

【作者基本信息】 中南大学 ， 地质资源与地质工程， 2012， 博士

【摘要】 多条高速公路分布在湖南省的西部地区,并且这些滑坡类型繁多,地质条件以及气候条件复杂,因此滑坡灾害对于人们生活和生产的影响最为直接、危害最大,高速公路滑坡灾害的监测预报研究是湖南省交通建设中急需解决的重要课题之一。本文以湖南省西部地区地质背景为基础,充分研究了高速公路滑坡灾害的具体特点,结合高速公路实际需要,以监测预报的完整过程为主线开展研究,采用现场试验、理论研究等手段,建立滑坡灾害在高速公路中的监测预报完整技术,包括了以下的研究内容和成果：(1)全面系统地研究了湘西区域滑坡灾害,特别是高速公路滑坡灾害的地层、构造、地形地貌,综合分析了开挖诱发型滑坡的共同特点。形成了湘西高速公路滑坡监测的主要关键技术。包括主要监测方法、常用仪器以及数据的获取技术。总结、对比了适合湘西特点的高速公路勘察设计、现场施工、通车运营阶段的滑坡监测设计要点,可以推广应用于其它相似特点区域的生产实践。(2)在实际应用中开发出了集成GPS的多传感器滑坡立体自动监测系统。数据采集仪自动采集监测仪器传感器的数据之后存储于数采仪的存储单元内,数采仪与DTU传输单元连接后借助于GSM卡GPRS功能将存储单元内的监测数据通过Internet网络远程传输到数据处理中心并存入计算机内,实现了实时监控的目的。(3)提出了一套在西部地区较为适用的高速公路滑坡灾害危险性区划的基本理论和方法,包括评价范围和评价单元的确定、评价指标体系的建立、评价模型的选择、评价结果的检验和可靠性分析等,为西部地区高速公路滑坡灾害危险性区划与预测提供了基础和依据。提出通过“灾害类型及成因机制分析→主控因子分析和定性类比分析推理→数学模型分析→综合评价”的思路,对滑坡灾害危险性进行综合分析和评价。(4)构建了基于卡尔曼滤波法的监测数据分析模型和程序,可以有效降低监测数据噪声对监测结果的影响。将位移监测数据划分为六种类型：“D”型、“B”型、“r”型、钟摆型、“V”型和复合型；分析了每种曲线类型所对应的滑坡特征；用曲线拟合技术对监测数据定量描述。构建了基于灰色系统理论的预测模型,在少数据量的变形预测模型方面具有较大的优势,并成功应用于怀通高速滑坡监测。(5)构建了“基于GIS的高速公路滑坡灾害监测预报系统”平台,实现了滑坡灾害安全管理、监测图表处理、预测预报分析、滑坡灾害区划及灾害信息的(实时)检索与查询,并应用ArcGIS的ArcServer/ArcIMS技术发布到互联网上实现了动态网页浏览。更多还原

【Abstract】 The western region of Hunan province includes many freeway, the landslides are distributed in wide range with various types, and due to the special nature of geological, geographical conditions and climate conditions, landslide disaster has the most direct and harmful impact on people’s life and production. The study on the monitoring prection of highway landslide disaster is one of the most urgent and important topic in Hunan province traffic construction.In this paper, the western region of Hunan Province is taken as the geological background, a full study of the specific features of highway landslide disaster is done, combined with the actual needs. The whole procedure of monitoring prection is the main line in the study, the field tests and theoretical research is taken as the study means. Finally, the complete sets of technology for highway landslide disaster monitoring and forecasting are founded. The main contents and results are as follows,(1) a comprehensive and systematic study of the Xiangxi regional landslide hazard is done, especially on the highway landslide stratum, tectonics, landform. Comprehensive analysis of the excavation induced landslides is done for common features. The main key technology for the Xiangxi Expressway Landslide Monitoring is formed, including the main monitoring method, apparatus and data acquisition technology. Summary, contrast for the landslide monitoring design points in Xiangxi Highway Survey, design, construction, and traffic operation stage are done, which can be applied to other similar characteristics of regional production practice.(2) in the practical application, the GPS integrated multisensor landslide three-dimensional automatic monitoring system is developed. Data acquisition instrument automatic collection apparatus for monitoring sensor data is stored in the data acquisition instrument within the storage unit, a data acquisition instrument and DTU transmission unit is connected by GSM card GPRS function, and the storedage unit within the monitoring data through Internet network will be transmitted to the data processing center and stored in the computer, to realize the real-time monitoring purposes.(3) a set of the basic theories and methods relatively suitable in the western region highway landslide hazard zonation is proposed, including the appraisal scope and appraisal unit method, the establishment of the evaluation index system, evaluation model selection, evaluation results of testing and reliability analysis, which can provide the basis and guidance for western area highway landslide hazard zonation. The idea of " disaster types and formation mechanism,→main controlling factor analysis and qualitative analogy reasoning,→analysis of mathematical model,→integrated evaluation" is put forward, to give the comprehensive analysis and evaluation for landslide hazard.(4) based on Calman filter method in the analysis of monitoring data model and program, the effect of the noise monitoring data to the monitoring results can be effectively reduced. The displacement monitoring data are divided into six types:type "D"," B"," R" type, the pendulum type,"V" type and the compound type; analysis of each kind of type curve corresponding to the landslide characteristics is done; curve fitting technique is used for quantitative description of monitoring data. The prediction model is founded based on the grey system theory, which shows great advantages in the field of deformation prediction model with small amount of data, whose results are applied successfully in Yu Huaitong high speed landslide monitoring.(5) the construction of the platform of "GIS based highway landslide disaster monitoring and prediction system" is done, and it is realized the landslide hazard safety management, monitoring chart processing, prediction analysis, landslide hazard zonation and disaster information (real-time) retrieval and query, and the application of ArcGIS ArcServer/ArcIMS technology is published to the Internet to achieve the dynamic webpage browsing.更多还原