矿山高边坡变形动态监测及稳定性预测

发布时间：2018-09-05 16:17

【摘要】：随着矿山开采的不断进行，越来越多的露天矿进入深凹开采阶段，所形成的高陡边坡稳定性也越来越差。此外，长期以来矿山企业对高陡边坡缺乏科学合理的监测手段，无法真正的掌握边坡的变形状态和发展规律，为矿山的安全生产埋下隐患。因此，愈来愈要求大范围、经常性和连续性地获取矿山边坡的变形参数，并对该参数进行及时有效的分析处理，，进而对边坡的稳定性作出科学的变形分析和预报研究。边坡的失稳变形是一个由微观到宏观的发展变化过程，只有借助精密的仪器及严密的监测方法才能获得目标点微小的形变和发展规律。针对矿山高陡边坡特殊的地质结构和观测环境，测量机器人的出现以最为经济可靠的方式满足了高陡边坡变形动态监测中对测量精度的要求。它集成步进马达和CCD影像传感器，采用ATR（Automatic Target Recognition）自动目标识别技术，实现了对目标的自动照准测量任务。同时，徕卡测量系统公司为自己生产的测量机器人提供了机载程序软件开发平台“GeoBasic”以及基于计算机在线控制软件的“GeoCOM”接口技术平台，用户可以根据不同需求开发上传到仪器中的机载程序，也可以开发外接设备上的应用程序。在边坡变形监测研究工作中，监测是手段、预测是目的。经多年研究，目前对变形数据的分析、处理，已形成了一套较为成熟的理论体系，建立了丰富的趋势拟合与预测预报模型，如回归分析模型，卡尔曼滤波模型、时间序列模型、小波理论、灰色模型以及人工神经网络模型等。石人沟铁矿是我国露天转地下开采较早的铁矿山，以石人沟铁矿的边坡变形监测系统为研究对象，运用回归模型和小波变换对原始监测数据进行去噪，还原变形的真实值。建立灰预测GM（1，1）模型对边坡的变形趋势进行预测，通过横向和纵向对比分析，证明了不同长度的数据序列对灰预测模型的精度影响明显，同时验证了小波变换结合灰理论能有效的提高模型的预测精度。
[Abstract]:With the development of mining, more and more open-pit mines enter the stage of deep pit mining, and the stability of high and steep slope becomes worse and worse. In addition, for a long time, mining enterprises lack scientific and reasonable monitoring means to high and steep slopes, and can not really grasp the deformation state and development law of the slopes, thus laying hidden dangers for the safe production of mines. Therefore, it is more and more necessary to obtain the deformation parameter of mine slope regularly and continuously, and to analyze and deal with the parameter in a timely and effective manner, and to make scientific deformation analysis and prediction study on the stability of the slope. The unstable deformation of slope is a developing process from micro to macro. Only by means of precise instruments and strict monitoring methods can the small deformation and development law of the target point be obtained. In view of the special geological structure and observation environment of high steep slope, the appearance of measuring robot meets the requirement of measuring precision in dynamic monitoring of high steep slope deformation in the most economical and reliable way. It integrates step motor and CCD image sensor, and uses ATR (Automatic Target Recognition) automatic target recognition technology to realize the task of automatic target alignment measurement. At the same time, Leica Measurement Systems Corporation provides the airborne software development platform "GeoBasic" and the "GeoCOM" interface technology platform based on the computer on-line control software for the measurement robot produced by Leica. Users can develop airborne programs uploaded to the instrument according to different requirements, and can also develop applications on external devices. In the study of slope deformation monitoring, monitoring is the means and prediction is the purpose. After many years of research and analysis and processing of deformation data, a set of relatively mature theoretical system has been formed, and rich trend fitting and forecasting models, such as regression analysis model, Kalman filter model, time series model, have been established. Wavelet theory, grey model and artificial neural network model. Shirengou Iron Mine is an early iron mine mountain in our country. Taking the slope deformation monitoring system of Shirengou Iron Mine as the research object, the original monitoring data are de-noised by regression model and wavelet transform, and the true value of deformation is reduced. The deformation trend of slope is predicted by using grey prediction GM (1 + 1) model. The results show that different length data series have obvious influence on the accuracy of grey prediction model. At the same time, it is verified that wavelet transform combined with grey theory can effectively improve the prediction accuracy of the model.
【学位授予单位】：河北联合大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TD854;TD76

【参考文献】