Original data feature extraction and identification method based on multi-channel ultrasonic flaw detection

A technology of feature extraction and raw data, applied in the direction of processing the response signal of detection, etc., can solve the problems such as the lack of time domain analysis ability of Fourier transform, the lack of effective information, and the influence of spectrogram.
CN110967409AInactive Publication Date: 2020-04-07东莞灵虎智能科技有限公司
5 Cites 9 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
东莞灵虎智能科技有限公司
Publication Date
2020-04-07
Estimated Expiration
Not applicable · inactive patent

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses an original data feature extraction and identification method based on multi-channel ultrasonic flaw detection, and particularly relates to an extraction and identification method for ultrasonic flaw detection original data acquired by a steel rail. Non-blind area detection is carried out on the rail head and the rail waist of a steel rail by adopting a 10-channel compositeserial probe, and broadband ultrasonic waves are reflected by the steel rail and are received by a receiving probe of an array probe. The method mainly comprises the following steps: firstly, using awavelet transform algorithm based on an improved threshold function for denoising a collected steel rail ultrasonic echo A-display data signal; then extracting features of the ultrasonic echo A-display data signals by adopting a wavelet modulus maximum feature extraction method, finally, normalizing the extracted features, converting all feature data into [0, 1] data, and displaying the processed echo in a B-type display mode. The method can extract effective pulse signals, improves the flaw judgment accuracy, and can be widely applied to the field of steel rail ultrasonic flaw detection.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of non-destructive testing, is mainly applied in the field of rail flaw detection, and relates to a feature extraction and recognition method based on multi-channel ultrasonic flaw detection raw data. Background technique

[0002] The rapid development of my country's high-speed railways has brought challenges to the state maintenance and safety assurance of railway rail infrastructure, especially the multiple speed increases have also put forward higher requirements for the quality of the rails. Rails are prone to cracks and other damages that affect and limit the performance of rails during use, and under the action of wheel load and temperature force, rail damage will be more serious. Rail fractures caused by rail stress, fatigue, and defect failures can lead to major accidents such as train derailment and overturning, resulting in casualties and huge property losses.

[0003] Ultrasonic flaw detection i...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More