Intelligent vision detection system and state detection method of transformer substation device

Abstract

The invention discloses a routing inspection robot vision system and a state detection method of a transformer substation. The routing inspection robot vision system comprises a motion control module, wherein a holder control module is arranged on the motion control module and connected with an image acquisition module through a vision servo module, the image acquisition module is connected with an image intelligent analyzing system through an image transmission module, the image intelligent analyzing system comprises a device classifying module, a disconnecting link state recognition module, a high-voltage breaker recognition module, an instrument reading module and an appearance abnormal recognition module, and modules of the image intelligent analyzing system are connected with a comprehensive analyzing module which is connected with a transformer substation information control system interface module. The routing inspection robot vision system and the state detection method of the transformer substation can automatically analyze the current device states including all detection contents of the daily patrol working of the transformer substation, achieve conversion from a cycle patrol mode to the patrol mode based on the states, complete automatic recognition and recording of the device states of the transformer substation and greatly reduce engagement of working staff.

Description

[0001] technical field [0002] The invention relates to an intelligent visual detection system and a detection method. In particular, it relates to an intelligent visual detection system and a state detection method for substation equipment. Background technique [0003] In recent years, with the rapid and continuous development of China's national economy, electricity is the basis for the development of various industries, and the length of transmission lines and the number of substations have followed a sharp increase. The traditional power equipment inspection work relies on manual inspection of equipment status on a regular basis to ensure the daily safe operation of substations. With the increasing power consumption and the number of substations, the contradictions such as low efficiency of manual inspection, scattered detection quality, and no connection with the existing information system are becoming more and more obvious. Obviously, manual inspection can no l...

AI Technical Summary

Problems solved by technology

Obviously, the state detection of the above-mentioned special equipment cannot satisfy the detection of the state of all equipment in the substation, and then cannot completely replace the function of manual inspection.

[0009] (2) Judging the status of the device only by the detection result of a single image, the reliability is poor

The existing visual inspection module only gives the detection result of the equipment status of a single image, and does not refer to the inspection method of confirming the equipment status from multiple angles during manual inspection

Therefore, the results of image detection directly affect the reliability of robot state detection, and due to various factors such as image quality and lighting effects in outdoor environments, it is difficult to achieve complete reliability of image detection

Therefore, relying solely on the results of single image processing, the reliability of the existing visual inspection system cannot be guaranteed

[0010] (3) The image information collection method is highly dependent on the positioning accuracy of the robot

Most of the existing visual detection methods use fixed point detection, that is, the robot will stop at a fixed position every time it inspects, and set the pan / tilt to a fixed position to collect equipment images. and other factors, the target device cannot be guaranteed to appear accurately within the field of view of the image, which is an important factor leading to the failure of visual inspection

[0011] (4) Lack of intelligent analysis function

Most of the existing visual inspection systems use status inspection. After the status inspection is completed, it is still necessary to manually judge whether the status is correct based on the work ticket information and historical records, and the degree of integration with the substation information management system is not high, and the on-site inspection results cannot be sent to the information management system. automatic update of

[0012] Due to the above-mentioned problems of the existing visual detection system, it is difficult to realize reliable automatic detection of the state of all equipment in the substation, and the detection accuracy and stability need to be improved, so that the reliability of the detection of the state of the substation equipment by robots cannot be guaranteed, thus affecting the substation Hidden dangers in safe operation

Images