基于数值计算和空间数据库的露天矿边坡安全预警系统研究

发布时间：2018-11-20 16:10

【摘要】：边坡变形监测与稳定性预测预报研究是岩石力学领域的热点研究问题，并取得了丰硕的研究成果，但随着GPS技术和我国北斗系统的日臻完善，对于边坡变形监测与预测预报智能化集成的研究成果略显不足。深入开展无人值守的露天矿山边坡变形实时、连续监测与稳定性智能预测预报研究对于露天矿山边坡安全管理和包括北斗系统在内的卫星导航技术在边坡监测与预测预报领域的推广应用具有非常重大的理论意义和实用价值。课题以“基于GPS的矿区重点区域实时变形监测预警系统”项目为依托，针对金堆城露天矿边帮断裂带规模巨大、内部结构复杂、岩性软弱的地质特点，从安全监测的时效性出发，通过数值分析对边坡监测点位进行了优化、通过信号处理和理论分析建立了预测预报模型，并运用空间内插加密算法，，实现了变形监测和预测预报的智能化集成。取得了以下主要成果：（1）运用数值分析确定了金堆成露天矿山重点边坡拟选用的监测点位。并利用Leica GNSS QC软件对监测点数据的周跳、多路径、数据完整性以及数据连续性进行分析，优化了点位布置，提出了一种新的监测点布设的原则。（2）引入了小波阈值收缩法对数据进行降噪处理。在对比分析GM（1,1）等时距和非等时距模型位移预测精度的基础上，建立了改进的灰色-变维分形模型，并采用Matlab数值仿真模拟软件对该模型进行验证分析，对滑坡体的水平和垂直以及整体位移进行求解，得到该模型对GPS数据具有很好的适应性，提高了预测精度。（3）运用空间内插加密算法，通过建立实时动态的真三维维矿山安全状况演变模型，再现与仿真矿山开采活动及给安全、环境带来的相关影响；基于历史监测数据和时间序列模型，及时准确预测潜在的地质灾害发生地点、时间和影响范围；实时动态发布矿山安全和地质灾害等可视化信息和短信预警。实现了“监测”、“评价”、“预测”、“报警”、“防盗”五为一体的智能化集成，实现了对边坡的无人化、可视化实时在线监测与预测预报。课题成果在金钼股份金堆城露天矿一年多的生产运行情况表明，该系统实现了全方位、全天候、立体化地对矿区地表形变的相关信息的无人化实时采集、快速安全传输、自动解算处理、监测信息处理和可视化预警，给矿山边坡安全管理提供了一个智能化辅助决策支持，对GPS和北斗卫星导航系统在边坡监测中的推广具有借鉴价值。
[Abstract]:The research of slope deformation monitoring and stability prediction is a hot research problem in the field of rock mechanics, and a lot of research results have been obtained. However, with the development of GPS technology and Beidou system in China, The research on the intelligent integration of slope deformation monitoring and prediction is a little insufficient. Carry out the real-time deformation of unattended open pit slope in depth, The research on continuous monitoring and intelligent prediction and prediction of stability has very important theory for slope safety management and satellite navigation technology including Beidou system in the field of slope monitoring and prediction. Meaning and practical value Based on the project of "real-time deformation monitoring and warning system in key mining area based on GPS", the paper aims at the geological characteristics of the side fault zone of Jinduicheng opencast mine with large scale, complex internal structure and weak lithology. In view of the timeliness of safety monitoring, the slope monitoring points are optimized by numerical analysis, the prediction model is established by signal processing and theoretical analysis, and the spatial interpolation encryption algorithm is used. The intelligent integration of deformation monitoring and prediction is realized. The main achievements are as follows: (1) using numerical analysis to determine the monitoring points to be selected for the key slope of Jinduicheng opencast mine. The cycle slip, multipath, data integrity and data continuity of monitoring data are analyzed by using Leica GNSS QC software, and the point position arrangement is optimized, and a new principle of monitoring point layout is put forward. (2) the wavelet threshold shrinkage method is introduced to reduce the noise of the data. On the basis of comparing and analyzing the displacement prediction accuracy of GM (1 / 1) equal time interval model and non-equal time interval model, an improved grey variable dimension fractal model is established, and the model is verified and analyzed by Matlab numerical simulation software. By solving the horizontal, vertical and global displacement of the landslide, it is found that the model has a good adaptability to the GPS data and improves the prediction accuracy. (3) by using the spatial interpolation encryption algorithm, through the establishment of the real time dynamic evolution model of the safety situation of the true 3D mine, the paper reproduces and simulates the mining activities of the mine and the related influence to the safety and environment; Based on the historical monitoring data and time series model, the location, time and influence range of potential geological hazards are predicted in time and accurately, and visual information and short message warning such as mine safety and geological hazards are published dynamically and in real time. The intelligent integration of "monitoring", "evaluation", "prediction", "alarm" and "anti-theft" is realized, and the real-time on-line monitoring and prediction of slope is realized. The operation of the project in Jinduicheng opencast mine for more than a year shows that the system has realized the real-time collection of the information related to the surface deformation of the mining area in an all-round, all-weather and three-dimensional manner, and the rapid and safe transmission. Automatic calculation processing monitoring information processing and visual early warning provide an intelligent auxiliary decision support for slope safety management. It has reference value for the popularization of GPS and Beidou satellite navigation system in slope monitoring.
【学位授予单位】：武汉理工大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TD76

【参考文献】