Identification of Defects in Ultrasonic Inspection of Overhead Distribution Lines Based on Bayesian Classification Algorithm

In China, overhead distribution lines are widely used in the power grids of rural and suburban areas. They take on the role of power delivery. The safe and stable operation of overhead distribution lines is of great significance to the reliability of power supply. In recent years, power companies in China have begun to use ultrasonic testing device to inspect defects on overhead distribution lines and have achieved good results. However, there are still some problems. The camera of the existing ultrasonic testing device has a low resolution and it can only judge whether there is a defect in a certain area. In addition, there are often many equipment in the picture. So it is difficult to identify the specific defective equipment. Therefore, it is often necessary to combine the cooperation of a high-definition camera, which increases the work cost and reduces the inspection efficiency. In order to solve these problems, this paper conducts a detailed statistical analysis of ultrasonic testing data of Shaanxi Electric Power Company from 2017 to 2020. Then a defect classification model for overhead distribution lines based on Bayesian classification algorithm is established. It is only necessary to obtain the testing data of existing testing device to calculate the specific equipment which the defect occurred. The results of sample test and field test show that the Bayesian classification model established in this paper can effectively identify the specific defective equipment. In the actual inspection process, the combination of ultrasonic testing device and Bayesian classification model can save cost and improve work efficiency.