透明工作面多属性动态建模技术

Multi-attribute dynamic modeling technique for transparent working face

  • 摘要: 为满足煤矿智能化开采对高精度地质模型的需求,提出透明工作面多属性动态建模方法,探讨了工作面综合探测多源异构数据特征、多属性数据融合算法、动态可视化建模技术,并进行实例应用。利用综合探测技术对工作面进行逐级综合探测,可在不同阶段获得多属性、多维度和多精度的多源异构探测数据,按照数据产生的阶段和频度,将多源异构探测数据划分为静态、动态和实时数据;通过数据配准实现多源异构探测数据量纲和尺度统一;通过交叉验证实现多源异构探测数据相互验证和补充;通过井震、震电和多参数联合反演,实现煤层厚度、地层波速、电阻率及其他属性参数预测;利用局部搜索、内插和网格化等动态可视化建模技术,实现模型的局部快速更新;采用局部渲染和CUDA实时绘制技术,实现模型高效渲染和实时呈现。结果表明:多属性融合能够将多源异构探测数据统一到同一地质空间中,交叉验证可提高探测数据解释精度,联合反演可实现多源异构探测数据多属性融合,提高探测数据空间分辨率并丰富属性信息;动态可视化建模以静态数据和工作面精细探测动态数据构建的初始工作面地质模型为基础,融合回采过程中获取的动态探测数据和实时数据,可实现回采工作面前方煤层顶底板和构造等信息快速局部更新和可视化,模型精度逐步得到提高,推采前方逐步地质透明化。透明工作面动态建模技术以采前、采中产生的多源异构数据为基础,与开采系统循环互馈,可为智能工作面开采提供高精度地质导航。

     

    Abstract: In order to provide a high-precision geological model for intelligent mining,a multi-attribute dynamic model- ing method for a transparent working face was put forward. Features of multi-source heterogeneous data from the com- prehensive detection of working face,the multi-attribute data fusion algorithm and dynamic visualization modeling tech- nique were studied and applied. Multi-attribute, multi-dimension and multi-precision data can be acquired through some comprehensive detection technologies at different production stages. Based on phases and frequencies of modeling data generation,the multi-source heterogeneous detection data were divided to static data,dynamic data and real-time data. Dimension and scales of the multi-source heterogeneous data can be unified through data registration and can be verified through cross validation. Coal thickness,stratum velocity,resistivity and other attribute parameters can be ob- tained by log-seismic joint inversion,seismic-resistivity joint inversion and multi-parameters joint inversion. The geo- logical model can be rapidly updated by using local data search,interpolation and grid technologies. Efficient and real- time rendering was realized through locally rendering and CUDA drawing methods. Results show that the multi-source heterogeneous detection data can be unified in the same geological space. Data interpretation accuracy can be raised by cross validation. Spatial resolution can be raised and attributes can be added by joint inversion. Based on the primary geological model built by the static data and dynamic data generated by fine detection in working face,coal seam roof, floor and structure information in front of the mining face can be updated based on dynamic data and real-time data to improve the model precision continuously. Transparent working face dynamic model was built on the basis of multi-at- tribute heterogeneous data generated during mining process. Intelligent cutting plan can be calculated through constant interaction between the dynamic model and mining system.

     

/

返回文章
返回