High voltage network pillar verticality analysis by means of point cloud processing

Egidij Egej Vencelj (2019) High voltage network pillar verticality analysis by means of point cloud processing. EngD thesis.

Preview

PDF Download (27Mb)

Abstract

Technical flawlessness of power grids is required for an uninterrupted flow of electricity, a key element of modern infrastructure. Due to their size, inspections of power grids are carried out with a flying vessel, which captures spatial and other data. Due to the large size of this spatial data, there is a need for fast and reliable processing, based on which we determine some of the network failures. The automatic determination of the characteristic points of the network pylons enables us to identify cases where a suspicion of a critical declination is present, ie. the failure of the network pylon. In this thesis we have developed a procedure for automatically determining the characteristic points of the pylons. The automatic determination of characteristic points was achieved by point cloud processing, using a combination of established methods in the field of computer vision. We compared the results of the procedure with manually determined points, and evaluated the success of the developed procedure on the basis of differences.

Actions (login required)