An approach for fault monitoring of insulators based on image tracking

In order to achieve unmanned monitoring in the smart substation, the video monitoring technology has been introduced. However, it is still lacking of real-time image processing in its actual use. In this paper, a system based on the image tracking technology for fault monitoring of insulators in the substation is developed. In this system, an unmanned aerial vehicle (UAV) moving in a special trace in the substation is used to carry on an infrared camera to shoot insulators and the images are transmitted to the PC end by the image transmission module on the UAV. At the PC end, images are processed in real time. When faults are detected in touring, the UAV will hover and make the camera shoot the fault insulator to track the development of the fault and the system will warn operators on duty to handle it. There are three steps in the image processing: the image segmentation, the fault diagnosis, and the fault tracking. For the image segmentation, the temperature value is transformed to the gray value and then a new approach to choosing segmentation threshold based on the histogram envelope is proposed. Eventually, insulator objective is extracted successfully. For the fault diagnosis, a relative temperature difference (RTD) method namely that the temperature difference between the average temperature of the high-temperature area and the other area of an insulator accounts for the temperature difference between the average temperature of the high-temperature area and the ambient temperature is adopted. With lots of tests, 30% is chosen as the threshold of small fault and 50% as large. In fault tracking, the change curve of the maximum temperature (MT) of the insulator and average temperature (AT) and RTD are showed on the screen of PC end. In the end, a piece of software on MATLAB is developed. This system can also be used for the power line touring by UAV.