堆石崩积体的滑溃前兆特征及其快速探测判识方法研究

The rockfill colluvium is a common secondary geological disasters after earthquakes. It has the following characteristics: material composition heterogeneity, presence forms randomness, loose accumulation sub-stability, slide and collapse sudden. So those characteristics are extremely easy to bring hidden security danger to the near residential, transportation, and engineering activities. In this project, we firstly will take geomorphic characteristics, material composition, presence forms as starting points, and use the method of combining model test and sophisticated numerical simulation test to search the prognostic features laws of the rockfill colluvium sliding and collapsing in different geomorphic characteristics, material composition, presence forms, and induced load. The laws includes the law of the original disaster points' position change, the law of local displacements evolution, and the law of force chains' strength evolution among the stones. Secondly, based on the tests' results, we will put to find the original disaster points and judge the critical mutation state of catastrophe as targets, and use comprehensively those methods included nonlinear mathematics, intelligent science, space reasoning topology, data mining technology to study the mathematical physics model which describing the prognostic features laws of the rockfill colluvium sliding and collapsing. At last, according to the catastrophe recognition theories, we will combine with TLS, computer visual identity and analysis, three-dimensional sophisticated numerical simulation to research and develop rapid detecting and recogniting methods of the prognostic features laws of the rockfill colluvium sliding and collapsing. The research has important scientific significance and engineering value in geological disasters' security evaluation and effective disaster warning.

堆石崩积体是地震后常见次生地质灾害隐患，它具有物质组构分异性、堆积形态随机性、松散堆积亚稳定性、滑溃前兆突发性，极易对临近人居、交通、及工程活动带来安全隐患。本项目首先以其地貌特征、物质组构、堆积形态为切入点，采用模型试验与精细化数值仿真试验相结合，探求不同地貌特征、物质组构、堆积形态、诱发荷载下堆石崩积体滑溃前兆特征规律(灾源点位置变化、局部位移演化、块石间力链强度演化)；其次基于试验成果，以寻找滑溃灾源点与灾变判定临界突变状态为目标，综合非线性数学、智能科学、空间推理拓扑学、数据挖掘技术等方法，研究描述堆石崩积体滑溃前兆特征的数理模型；最后以灾源点搜索规则与灾变判定临界突变模型为灾变判识理论，融合TLS测绘技术、计算机视觉识别与分析技术、三维精细化数值仿真技术，研发堆石崩积体滑溃前兆特征的快速探测判识方法。上述研究对提升地质灾害安全评估与有效防灾预警具有重要科学意义与工程价值。

堆石崩积体是地震诱发岩坡崩塌所产生的次生地质体，它具有物质组构分异性、堆积形态随机性、工程安全亚稳定性、滑溃破坏突发性，对灾区人员、物资运输及灾后重建等救灾与工程活动带来巨大的安全隐患。经过四年的研究工作，基本上按计划完成了三大部分的研究内容：1、基于三维激光扫描技术、计算机视觉技术、非线性数学、数据挖掘技术，研发描述破碎危岩与堆石崩积体的地形环境、物质组构、堆积形态的三维地质环境建模与地质信息采集技术，研究了堆石崩积体的地形环境、物质组构、堆积形态等赋存形式特征规律；2、采用振动台模型试验与精细化数值试验相结合的技术手段，开展了不同地形环境、物质组构、堆积形态、诱发荷载下堆石崩积体滑溃破坏规律研究；3、以寻找滑溃灾源点与滑溃临界破坏状态为目标，研究了堆石崩积体滑溃前兆特征及其规律（灾源点位置变化、局部位移演化、块石间力链强度演化），研究了堆石崩积体滑溃破坏前兆特征的快速探测判识方法。最终，本项目开发了三维地质环境建模与地质信息采集技术软件平台VGM 1.0（Virtual Geological Mapping），实现了地形条件、地质环境、地质信息的全自动、非接触、数字化建模与信息化识别；总结了不同条件下堆石崩积体滑溃破坏的特征规律，划分了2类典型滑溃破坏模式，获得了不同地形条件、物质组构、堆积形态与3类典型滑溃破坏模式之间的对应关系；建立了基于复杂网络理论的灾源点搜索方法与基于极限状态几何表达的堆石崩积体滑溃临界破坏状态数学模型，实现融合三维激光扫描测量、计算机视觉识别、数值与数学分析、空间几何判断的堆石崩积体滑溃破坏前兆特征的快速探测判识方法。上述研究对提高震区灾后地质灾害安全评估与防灾预警的理论与技术水平具有重要科学意义与工程应用价值。本课题共发表科研论文10篇，其中SCI/EI检索1篇，期刊论文9篇，目前在投稿4篇；申请专利4项均已公开；参加国内外会议10次；培养硕士研究生6人。