基于机载LiDAR点云数据的电力线安全距离检测

发布时间：2018-06-24 05:51

本文选题：Li + DAR　；参考：《兰州交通大学》2015年硕士论文

【摘要】：长期以来,输电线走廊的巡检、维护主要依赖于人工到现场勘查来完成,该方法不仅效率低,而且安全性差,已经不能满足现代化电网的发展和安全运行需求,因此电力部门急需一种高效、安全、先进的电力巡检方式。机载激光雷达测量技术的问世带来了新的电力巡检途径。与传统测量系统相比,机载Li DAR测量系统有着独特的技术优势,它能直接获取电力线走廊的激光点云,并且可以得到高精度的三维空间信息,进而获得三维电线走廊的全貌信息,如线路设施设备,地物的精确三维空间信息等等。现阶段,利用Li DAR测量技术在电力线路安全巡检中得到了广泛应用,如基于机载激光点云数据的电力线安全距离自动检测技术就是电力巡线中最重要的一项内容,通过检测可以直接确定有缺陷的区域,不仅节省了大量的时间,同时也消除了安全隐患。该内容即为本文的研究目标。本文首先探讨了不同的电力线提取方法,分析了各方法的利弊。在此基础上,选择了K-近邻聚类电力线提取方法对电力线进行了提取,具体的研究内容及提取的过程为:采用半自动化的方法对电力线点和大部分杆塔以外的点进行提取;通过量取塔顶中心的三维坐标对杆塔进行剔除;利用基于K-近邻聚类方法对电力线进行提取,采用最小二乘法原理对电力线点云进行拟合,并取得了良好的效果。在上述工作的基础上,利用已经提取好的地物点和电力线点进行了电力线安全距离检测的研究,其主要研究内容和成果如下:本文在分析对比利用低空航拍数据对电力线安全距离进行检测后,提出了一种利用机载Li DAR点云数据自动化检测树障的方法。通过对地物点规则格网化求取最高点,利用最高点做缓冲区,并将缓冲区内的所有电力线点与最高点进行距离的计算,找到所有小于安全距离的点。实验证明,本方法可以有效的找出存在隐患的地方。将林业中估算树木生长模型与电力线安全距离检测相结合,通过已知数据利用树木生长灰色代数曲线模型(Grey Algebraical curve Model,以下简称GAM模型)对最近两年树木的增长量进行估测,将估计结果重新代入到电力线走廊的地物点中,进一步找到存在隐患的地方,进行预警预测。
[Abstract]:For a long time, the inspection and maintenance of transmission line corridors mainly depend on the artificial on-site survey. This method is not only inefficient, but also has poor security, and can no longer meet the needs of the development and safe operation of modern power grids. Therefore, the electric power department is in urgent need of an efficient, safe and advanced power inspection method. The advent of airborne lidar measurement technology has brought a new way of power inspection. Compared with the traditional measurement system, the airborne Li Dar measurement system has a unique technical advantage, it can directly obtain the laser point cloud of the power line corridor, and can obtain high precision three-dimensional spatial information. Then, the complete information of three-dimensional wire corridor is obtained, such as the line facilities and equipment, the accurate three-dimensional spatial information of the features and so on. At present, Li Dar measurement technology has been widely used in power line safety inspection. For example, the automatic detection technology of power line safety distance based on airborne laser point cloud data is the most important content of power line inspection. The defect area can be directly determined by detection, which not only saves a lot of time, but also eliminates the hidden danger of safety. This content is the research goal of this paper. In this paper, different methods of power line extraction are discussed, and the advantages and disadvantages of each method are analyzed. On this basis, K-nearest neighbor clustering power line extraction method is selected to extract the power line. The specific research content and extraction process are as follows: the semi-automatic method is used to extract the power line points and the points outside most towers; The pole tower is eliminated by taking the three-dimensional coordinates of the top center of the tower in excess, and the power line is extracted by using K-nearest neighbor clustering method, and the power line point cloud is fitted by the principle of least square method, and good results are obtained. On the basis of the above work, the safe distance detection of power line is studied by using the ground objects and power line points that have been extracted. The main research contents and results are as follows: after analyzing and comparing the low altitude aerial photograph data to detect the safe distance of power line, a method of automatic detecting tree barrier using airborne Li Dar point cloud data is proposed in this paper. The highest point is obtained by the regular grid of ground objects, the highest point is used as buffer zone, and the distance between all the power line points in the buffer zone and the highest point is calculated, and all points less than the safe distance are found. Experiments show that the method can find out the hidden trouble effectively. In this paper, the tree growth model in forestry is combined with the safe distance detection of power line, and the growth of trees in the last two years is estimated by using the known data using the grey Algebraical curve model. The estimation result is replaced into the ground point of the power line corridor, and the hidden trouble is found further, and the prediction is carried out.
【学位授予单位】：兰州交通大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM75

【相似文献】