The analysis of axle box vibration signal based on frequency slice wavelet transform

Wang, Fuge; Xing, Zongyi; Qin, Yong; Li, Jianwei; Chen, Yuejian

doi:10.1109/iccacs.2015.7361324

Cited by 1 publication

(1 citation statement)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, noise reduction method is necessary. Common signal decomposition algorithms include Wigner-Ville distribution, Fri1, and wavelet transform [31][32][33]. However, there is cross interference with the Wigner-Ville distribution.…”

Section: Introductionmentioning

confidence: 99%

Quantitative Imaging Detection of Additive Manufactured Parts Using Laser Ultrasonic Testing

et al. 2020

View full text Add to dashboard Cite

Selective laser melting (SLM) is an important method of additive manufacturing, however it has some disadvantages such as poor surface qualities of the formed parts and the appearances of small surface defects. In this paper, 360L stainless steel with artificial defects in different lengths are processed by SLM processing. A full noncontact laser ultrasonic-based B-scan detection system is built to detect the surface defects. The interaction between the scattered surface wave and the surface defect is verified through 3-dimension (3D) finite element simulation. In order to improve the signal-to-noise ratio (SNR) of laser ultrasonic signal in AM 316L part, Variational Mode Decomposition (VMD) algorithm based particle swarm optimization (PSO) is applied to denoise signals. Meanwhile, wavelet transform (WT) algorithm is used to compare SNR with VMD algorithm. Then, the time and amplitude parameters of different positions are extracted to realize B-scan imaging, and the lengths of defects are further accurately quantified through the time-amplitude-position imaging images. Otherwise, we find that the size of laser spot affects the precise quantification of defects. The smaller the spot is, the more precise the quantitative effect is. The results show that this method can quantitatively detect surface defects of AM 316L parts.

show abstract

Section: Introductionmentioning

confidence: 99%