DeeplabV3+ Image Processing Based Steel Components Corrosion Analysis of Power Transformation Line Towers

Abstract

As the essential infrastructure for the power industry, the safety and reliability of the transmission system are directly related to the safe operation of the power grid. However, the corrosion failure of transmission and transformation equipment has increasingly become a bottleneck that restricts the safe operation of power grids. If effective anti-corrosion measures are not taken in a timely manner, transmission and transformation equipment served in various complicated atmospheric environments will suffer serious corrosion damage in a relatively short period of time, which will endanger the safe utilization of transmission and transformation equipment together with the safety of grid operation. Therefore, this paper establishes a standard corrosion spectrum grading software based on image recognition technology, which is based on the DeeplabV3+ algorithm for image segmentation. Experiments have proven that the software can identify the parts that are corroding and will corrode using mobile devices so that the corrosion status of transmission and transformation equipment can be analyzed and evaluated. The utilization of the software can dramatically reduce the safety accidents and economic losses caused by neglecting corrosion, which is of great significance to the safe operation of the power grid.