20kV变电站的多层数据智能三维调度方法研究

变电站数据存在空间拓扑结构，传统方法采用二维调度，导致维度空间信息消失，调度结果不可靠，影响调度人员对整个变电站状态的掌控。为此，提出一种新的20kV变电站多层数据智能三维调度方法。通过四叉树方法建立变电站多层数据空间索引，把最小外包矩形置于能够将其全部保存的最小矩形节点内，给出四叉树形成过程。按照不同层次与粒度，通过空间划分方法对20kV变电站多层次数据进行划分，得到连续分层层次细节模型，从下往上遍历四叉树，求出每层误差，将求出的误差作为多细节层LOD选取的参考值。给出保持区域更新与数据调度间动态平衡的方法。分析了视点区域，利用当前视点参数对LOD分布进行管理。利用索引表获取二维空间和三维空间的映射关系，在三维空间中确定三维多层数据。依据模型与视点的距离动态智能调度不同层次变电站数据，给出详细三维调度过程。实验结果表明，所提方法适于实际应用，可行性强。

Research on Multilayer Data Intelligent Three - Dimensional Scheduling Method for 20kV Substation

The substation data has spatial topology. The traditional method adopts two-dimensional scheduling, which leads to the disappearance of dimensional space information and the unreliable dispatching result, which affects the dispatcher''s control of the whole substation status. Therefore, a new multi-layer data intelligent three-dimensional scheduling method for 20kV substation is proposed. The quadtree method is used to establish the spatial index of multi-layer data in substations, and the minimum enclosing rectangle is placed in the smallest rectangular node that can save all of them. The quadtree formation process is given. According to the different levels and granularities, the multi-level data of 20kV substations are divided by space division method to get the continuous hierarchical level detail model, traversing the quadtree from the bottom to the top, calculating the error of each level, and taking the calculated error as the detail Layer LOD selected reference value. A method of keeping the dynamic balance between area update and data schedule is given. The viewpoint area is analyzed, and the LOD distribution is managed by using the current viewpoint parameters. The index table is used to obtain the mapping relationship between two-dimensional space and three-dimensional space, and the three-dimensional multi-layer data is determined in the three-dimensional space. Based on the distance between model and viewpoint, the data of substations at different levels are dynamically and intelligently dispatched, and the detailed three-dimensional scheduling process is given. Experimental results show that the proposed method is suitable for practical application and is feasible.