Construction of a 3D Distribution Network Environment Based on Multi-source Data Fusion

Abstract

Traditional distribution networks are designed using two-dimensional drawings. Its expression is abstract, the accuracy of manually measured elevation data could be higher, and complex sections are difficult to measure. To solve this problem, a three-dimensional environment for digital planning and design of the distribution network is proposed by combining laser point cloud and oblique photography modelling. Firstly, a rough registration method based on the main material plane is proposed to improve the registration efficiency. The GICP method is used to accurately register the point cloud and build a refined model of distribution equipment; Then, the UAV tilt photography technology is used to model the distribution line corridor, and the elevation model is obtained at the same time. The experimental results show that the model error after precise registration is 3.75 cm, which meets the requirements of high-precision modelling. The average error between the elevation model obtained by the tilt photography of the distribution line corridor and the measured elevation is 0.112 m, which meets the accuracy requirements of mapping.