Identification method and system of abnormal power equipment images

Abstract

The invention discloses an identification method of abnormal power equipment images, and is applied to identifying the abnormal power equipment images from power equipment images. The method comprises the following steps: 1, obtaining the power equipment images; 2, extracting power equipment features from the power equipment images; 3, based on the power equipment features, identifying power equipment and image areas where the power equipment is located; and 4, comparing images of the image areas with corresponding comparison images to determine whether abnormal change exists in the images of the image areas so as to identify the abnormal power equipment images, wherein the abnormal change is correlated with operation state abnormities of the power equipment. Correspondingly, the invention also brings forward an identification system of abnormal power equipment images, for identifying abnormal power equipment images from power equipment images.

Description

technical field [0001] The invention relates to a method and system for identifying abnormal images, in particular to a method and system for identifying images of abnormal electric equipment. Background technique [0002] With the continuous expansion of the scale of my country's power network, the safe and reliable operation of power equipment such as substations is crucial, and the operating status of power equipment is one of the key factors that determine its safe and stable operation. [0003] Image monitoring of power equipment through video monitoring system is a way of monitoring the operation status of power equipment. Existing video monitoring systems only have video monitoring and video recording functions, and cannot perform intelligent active identification and analysis of monitoring targets. They only transmit a large number of images to the dispatching end, requiring operators to observe and analyze images at all times, which virtually increases the number of...

AI Technical Summary

Problems solved by technology

Existing video monitoring systems only have video monitoring and video recording functions, and cannot perform intelligent active identification and analysis of monitoring targets. They only transmit a large number of images to the dispatching end, requiring operators to observe and analyze images at all times, which virtually increases the number of operations. At the same time, the weakness of human eyes and the subjectivity of manual judgment have seriously affected the further improvement of the automation of power equipment operation status monitoring; in addition, the operation status of many high-voltage equipment is difficult to convert into electrical signals. The process of conversion and transmission is easily affected by strong electromagnetic fields; the operating parameters of important equipment need to be monitored in real time, and manual inspection is difficult to meet the real-time requirements, and the inspector's sense of responsibility, work attitude and mental state seriously affect the detection results; and , it is difficult for the human eye to distinguish the grayscale changes of subtle images, and it is difficult to objectively judge the degree of surface defects of electrical equipment

[0004] For example, the power grid dispatching automation system (SCADA) high-voltage equipment insulation monitoring, relay protection and other devices play an important role in the safe and reliable operation of substations, but relatively speaking, the power equipment operation status monitoring system is not perfect, many of which are the source of problems Due to the existence of high voltage and strong electromagnetic field, the acquisition of many key parameters is limited by the measured, environment and measurement methods, and it is necessary to consider too many factors such as system safety, insulation and weak signal transmission; in addition, some operating parameters and fault symptoms Signals are difficult to convert into electrical signals through contact measurement, and even cannot be obtained by computer monitoring, such as transformer oil leakage. Such important non-electrical signals can be obtained more realistically by image monitoring

Images