在露天煤矿开采中,开采环境存在复杂多变等特点,矿山整体结构的改变导致的矿山边坡形变给安全生产带来重大隐患,也会给生态环境带来一定的破坏。当前对于矿山的变形监测虽逐渐趋向于自动化,但整个过程仍需依赖于人工或各种监测设备,且...在露天煤矿开采中,开采环境存在复杂多变等特点,矿山整体结构的改变导致的矿山边坡形变给安全生产带来重大隐患,也会给生态环境带来一定的破坏。当前对于矿山的变形监测虽逐渐趋向于自动化,但整个过程仍需依赖于人工或各种监测设备,且设备的维护比较困难,有的设备操作较为复杂、成本较高。为了更为简便直观地监测矿井边坡的形变信息,结合基于学习的点云配准方法实现了一种基于激光点云配准的矿井边坡变形监测方法。该方法首先提出了一种深度学习模型SA-RPE(Self-Attention with Relative Position Encoding)(相对位置编码的自注意力模型)在矿山数据集上实现了点云配准,并通过实验数据进行了验证;然后,根据配准的结果对矿井边坡进行了形变分析,并通过截取矿井边坡点云不同方向的断面进一步分析了各个断面的形变程度,结合二维断面图与三维点云渲染图的实验结果,表明深度学习模型SA-RPE能够比较准确地实现矿山激光点云的配准任务。通过分析深度学习模型预测的配准结果中旋转矩阵与平移向量的误差能够很好地掌握矿山的整体形变信息,而矿井边坡点云在不同方向上的断面图直观地展现了每一处形变的程度,计算不同时期断面点云对应点之间的平均距离能定量地描述各个断面的形变程度,通过阈值检测出来的异常值反映了断面上发生了较大形变的区域。实验结果体现了所提方法能够在满足实时性的基础上准确直观地表现出矿山边坡变形的信息。展开更多
The development of remote frequency transfer techniques,especially the appearance of optical clocks with unprecedented stability,has prompted geoscientists to study their applications in geodesy.Using remote frequency...The development of remote frequency transfer techniques,especially the appearance of optical clocks with unprecedented stability,has prompted geoscientists to study their applications in geodesy.Using remote frequency transfer technique,by frequency comparison of two optical clocks at two points P and Q connected by optical fibers,one can measure the signal’s frequency shift between them,and the geopotential difference between them can be determined based on the gravity frequency shift equation.Given the orthometric height of P,the orthometric height of Q can be determined.Since the present stability of the optical clock has achieved 1×10^(-18) or better and comparing the frequency transfer via optical fiber provides stability at 10^(-19) level,the optical clock network enables determining the orthometric height at centimeter-level.This study provides a formulation to determine the height diffe rence at one-centimeter level between two points on the ground based on the optical fiber frequency transfer technique.展开更多
文摘在露天煤矿开采中,开采环境存在复杂多变等特点,矿山整体结构的改变导致的矿山边坡形变给安全生产带来重大隐患,也会给生态环境带来一定的破坏。当前对于矿山的变形监测虽逐渐趋向于自动化,但整个过程仍需依赖于人工或各种监测设备,且设备的维护比较困难,有的设备操作较为复杂、成本较高。为了更为简便直观地监测矿井边坡的形变信息,结合基于学习的点云配准方法实现了一种基于激光点云配准的矿井边坡变形监测方法。该方法首先提出了一种深度学习模型SA-RPE(Self-Attention with Relative Position Encoding)(相对位置编码的自注意力模型)在矿山数据集上实现了点云配准,并通过实验数据进行了验证;然后,根据配准的结果对矿井边坡进行了形变分析,并通过截取矿井边坡点云不同方向的断面进一步分析了各个断面的形变程度,结合二维断面图与三维点云渲染图的实验结果,表明深度学习模型SA-RPE能够比较准确地实现矿山激光点云的配准任务。通过分析深度学习模型预测的配准结果中旋转矩阵与平移向量的误差能够很好地掌握矿山的整体形变信息,而矿井边坡点云在不同方向上的断面图直观地展现了每一处形变的程度,计算不同时期断面点云对应点之间的平均距离能定量地描述各个断面的形变程度,通过阈值检测出来的异常值反映了断面上发生了较大形变的区域。实验结果体现了所提方法能够在满足实时性的基础上准确直观地表现出矿山边坡变形的信息。
基金supported by the National Natural Science Foundations of China(Grant Nos.42030105,41721003,41804012,41631072,41874023)the Space Station Project(Grant No.2020-228)the Natural Science Foundation of Hubei Province of China(Grant No.2019CFB611)。
文摘The development of remote frequency transfer techniques,especially the appearance of optical clocks with unprecedented stability,has prompted geoscientists to study their applications in geodesy.Using remote frequency transfer technique,by frequency comparison of two optical clocks at two points P and Q connected by optical fibers,one can measure the signal’s frequency shift between them,and the geopotential difference between them can be determined based on the gravity frequency shift equation.Given the orthometric height of P,the orthometric height of Q can be determined.Since the present stability of the optical clock has achieved 1×10^(-18) or better and comparing the frequency transfer via optical fiber provides stability at 10^(-19) level,the optical clock network enables determining the orthometric height at centimeter-level.This study provides a formulation to determine the height diffe rence at one-centimeter level between two points on the ground based on the optical fiber frequency transfer technique.
