Weld defect identification in friction stir welding using power spectral density

International Conference on Recent Advances in Materials & Manufacturing Technologies

Das, Bipul^1 ; Pal, Sukhomay^1 ; Bag, Swarup^1

Department of Mechanical Engineering, Indian Institute of Technology Guwahati, Guwahai, Assam

781039, India^1

关键词: Friction stir;    Internal defects;    Periodograms;    Preliminary information;    Process signals;    Transverse force;    Vertical force;    Welch periodogram;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/346/1/012049/pdf
DOI  :  10.1088/1757-899X/346/1/012049

来源: IOP

【 摘 要 】

Power spectral density estimates are powerful in extraction of useful information retained in signal. In the current research work classical periodogram and Welch periodogram algorithms are used for the estimation of power spectral density for vertical force signal and transverse force signal acquired during friction stir welding process. The estimated spectral densities reveal notable insight in identification of defects in friction stir welded samples. It was observed that higher spectral density against each process signals is a key indication in identifying the presence of possible internal defects in the welded samples. The developed methodology can offer preliminary information regarding presence of internal defects in friction stir welded samples can be best accepted as first level of safeguard in monitoring the friction stir welding process.