Rail damage detection method, device and equipment

Abstract

The invention discloses a rail damage detection method. The rail damage detection method includes the steps that detection signals in the train driving process are collected, wherein the detection signals comprise vibration acceleration signals and / or impact signals; the detection signals are subjected to feature extraction to obtain a feature value; the feature value is compared with a preset threshold value, and whether the feature value is over than the preset threshold value or not is judged; if the feature value is over than the preset threshold value, the damage type of a rail is determined according to the feature value. According to the ail damage detection method, whether the rail is damaged or not can be accurately detected, the accuracy and the reliability of rail damage detection are guaranteed, and the rail safety and the train running safety are effectively improved. The invention also discloses a rail damage detection device and equipment, and the rail damage detection device and equipment have the technical effects.

Description

technical field [0001] The invention relates to the field of track management, in particular to a track damage detection method, and also relates to a track damage detection device and equipment. Background technique [0002] As a device for carrying the load of the train, the quality and state of the track directly affect the safety of the train. For this reason, it is necessary to detect the track, find damage in time, repair the damage, and avoid train operation accidents caused by track damage. At present, most of the ways of track inspection are to use rail inspection cars or flaw detectors for detection. However, the above detection methods need to be implemented during the period when there is no train running, and the detection is performed every once in a while, resulting in poor real-time detection. There is no way to ensure timely detection and repair of track damage. Or, the track image is obtained by the visual instrument on the train, and then the track image...

AI Technical Summary

Problems solved by technology

At present, most of the ways of track inspection are to use rail inspection cars or flaw detectors for detection. However, the above detection methods need to be implemented during the period when there is no train running, and the detection is performed every once in a while, resulting in poor real-time detection. Failure to ensure timely detection and repair of track damage

Or, the track image is obtained by the visual instrument on the train, and then the track image is analyzed to realize the track detection. Although this track detection method improves the real-time performance of the track detection to a certain extent, its detection cost is high and the detection accuracy is low. Low, for example, it is impossible to detect the internal defects of the track, the fine cracks on the surface, etc., so that the accuracy of the test results cannot be guaranteed, the quality of the track cannot be guaranteed, and there are potential safety hazards in the operation of the train

Images