Options
A filter calibration method for laser-scanned weld toe geometries
Citation Link: https://doi.org/10.15480/882.13696
Publikationstyp
Journal Article
Date Issued
2024-12-01
Sprache
English
TORE-DOI
Volume
20
Article Number
100200
Citation
Applications in Engineering Science 20: 100200 (2024)
Publisher DOI
Scopus ID
Publisher
Elsevier
The scanning of weld seams can be used to evaluate the local weld toe geometry for fatigue assessments. Laser scanned weld seam profiles often contain noise which complicates the accurate measurement of the weld toe geometry. For that reason, filtering of the scanned data is necessary. The issue at hand is that a filtering method can significantly affect the measurement results. Therefore, a calibration of the filter input parameters is needed. In this study, a calibration method for filtered laser-scanned weld profiles is presented by using artificial weld toe geometries. The adjustment of different filter functions is achieved by using an optimization method on predefined weld toes with an artificial noise. The resulting input data for the filter functions is tested on a real specimen to verify the method. Through the calibration method it is possible to achieve satisfactory measurement results with precisely set input parameters for the filter functions. The most suitable filter functions for the measurement of the weld toe are the Gaussian and the Lowpass filter. Both functions are adequate as a universally applicable filter. For the evaluation of the measurement results of the radii and angles, a tolerance range is introduced, which is defined by the theoretically minimum measurable radii and angles. Using an adjusted Lowpass filter and a point distance of 0.07 mm set by the laser scanner, a measurement within the tolerance range of 0.2 mm is achievable for the weld toe radius. For the weld toe angle, the tolerance range of 1.5° is achieved for the majority of measurements.
Subjects
Data smoothing
Filter calibration
Laser scanning
Universal filter
Weld toe measurement
DDC Class
620.1: Engineering Mechanics and Materials Science
519: Applied Mathematics, Probabilities
Publication version
publishedVersion
Loading...
Name
1-s2.0-S2666496824000268-main.pdf
Size
35.6 MB
Format
Adobe PDF