Detection system of high voltage shell discharge insulation performance based on multimodal texture analysis

Abstract

The invention relates to the field of artificial intelligence, in particular to a high-voltage shell discharge insulation performance detection system based on multimodal texture analysis. The image acquisition module processes the collected thermal infrared images to obtain multiple real connected domains; the real connected domain division module calculates the temperature anomaly rate of the real connected domain; calculates the depth anomaly rate of the real connected domain, and divides the real connected domain into discharge connectivity Domain and undischarged connected domain; distribution coefficient calculation module, calculates the distribution coefficient of discharge connected domain and undischarged connected domain, discharge coefficient determination module, calculates discharge coefficient of high-voltage shell, prevention coefficient determination module, calculates high-voltage shell disaster prevention coefficient ; Evaluation module, establish a binary group to evaluate the discharge insulation performance. The present invention collects different image data of the high-voltage casing for multi-modal analysis, and can obtain accurate discharge insulation performance.

Description

technical field [0001] The invention relates to the field of artificial intelligence, in particular to a high-voltage shell discharge insulation performance detection system based on multimodal texture analysis. Background technique [0002] The high-voltage casing is one of the upper casings of the gas-insulated metal-enclosed switchgear (GIS), and it is an extremely important part. The parts work normally under high voltage without affecting each other, and realize the effective and safe connection between the functional components of the GIS equipment, so it is necessary to ensure their insulation performance. [0003] At present, the insulation performance testing of high-voltage housings is mainly performed manually or with instruments, but the efficiency of manual testing is low, and the testing instruments are generally more expensive and costly; the existing technology uses threshold segmentation and other technologies to detect surface defects. It is tested to dete...

AI Technical Summary

Problems solved by technology

[0005] The invention provides a high-voltage housing discharge insulation performance detection system based on multi-modal texture analysis to solve the existing problems, including: an image acquisition module, which processes the collected thermal infrared images to obtain multiple real connected domains; The real connected domain division module calculates the abnormal temperature rate of the real connected domain; calculates the abnormal depth rate of the real connected domain and divides the real connected domain into discharged connected domain and undischarged connected domain; the distribution coefficient calculation module calculates discharged connected domain and undischarged connected domain The distribution coefficient of the domain, the discharge coefficient determination module, calculates the discharge coefficient of the high-voltage shell, and the prevention coefficient determination module, calculates the high-voltage shell prevention coefficient; the evaluation module, establishes a binary group to evaluate the discharge insulation performance

Images