A method for detecting equipotential line faults of catenary suspension devices for high-speed railways

Abstract

The invention discloses a fault detection method for an equipotential line of a catenary suspension device of a high-speed railway. The positive and negative samples required for training the classifier are intercepted from the global and local images of the catenary obtained; the positive and negative sample images are extracted using the HOG algorithm, and the extracted features are input into the Adaboost cascade classifier and the SVM support vector machine Carry out training, use the trained classifier to identify and locate the target image to be detected from the entire catenary suspension device image; perform image preprocessing on the identified target image, and use the level set RSF model to extract the edge of the image target; calculate the fault The connected domain of the equipotential line image, the number and area of ​​the connected domain are calculated as the criterion for judging whether the equipotential line has a loose strand fault, and the fault judgment is carried out for the equipotential line identified in the catenary suspension device image. The invention better solves the efficient and rapid detection of the equipotential line fault of the high-speed catenary suspension device, can meet the real-time and accuracy requirements of the online detection of the system, and has broad application prospects.

Description

technical field [0001] The invention relates to the field of on-line detection of high-speed railway catenary, in particular to the technical field of real-time on-line detection and high-precision fault detection. Background technique [0002] The pantograph-catenary system is a key component of the traction power supply system. The pantograph-catenary system must ensure that the traction network transmits electric energy to the electric locomotive to ensure the normal operation of the electric locomotive. Due to the very complex mechanical and electrical interaction between the pantograph and the catenary, its failure rate has been in a high state. Therefore, efficient real-time detection of the pantograph-catenary system is required to grasp the work of the pantograph-catenary system in real time It is of great significance to detect the bad state in time and repair and maintain the bad state. [0003] The traditional pantograph-catenary detection is mainly manual work, ...

AI Technical Summary

Problems solved by technology

[0004] The existing non-contact network image detection device mainly relies on the linear array CCD camera installed on the top of the detection vehicle to take detection images, and mainly relies on human eye observation in fault identification, which has a low degree of automation and slow processing speed

Images