Wavelet analysis-based expulsion identification in electrode force sensing of resistance spot welding
- 103 Downloads
Resistance spot welding is one of the most widely used processes in metal fabrication, but the expulsion affects the spot quality. To develop a precise and reliable weld quality assessment method for resistance spot welding, the welding force signal was measured and a novel wavelet transform based on multi indexes is proposed. The impulse and resultant damping vibration signal, which is the most obvious feature of expulsion, was extracted from the welding force waveform by 7-level wavelet transform with Daubechies5 wavelet. The detail signal in level 6 was chosen as the target signal, as it covers the dominant frequency of expulsion. To obtain the local characteristics of the expulsion, a multi indexes were calculated which includes the peak-to-peak value, kurtosis index, and pulse index. The peak-to-peak measures the range of a signal value, and kurtosis index and pulse index are sensitive to the impulse signal. According to experimental analysis, the peak-to-peak value of the expulsion signal is significantly higher than that of other conditions, while other two indexes are not that obvious but with same trend. The experimental data results show that the multi expulsion indexes are effective indicators in evaluating expulsion in resistance spot welding.
KeywordsQuality assessment Resistance spot welding Wavelet transform Expulsion
This research is supported financially by the National Natural Science Foundation of China (Grant No. 51775007) and the China Scholarship Council (CSC).
- 1.Mironov LG, Afanasev LK, Zaitsev VA (1979) Control of the quality of welds directly during resistance spot welding. Weld Prod 26:40–43Google Scholar
- 2.Yuan SX, Luo Z, BU XZ, Song KL, Ao SS (2008) Study of acoustic signal in the process of resistance spot welding based on array sensor system. China Weld 17:58–61Google Scholar
- 7.Livshits AG (1997) Universal quality assurance method for resistance spot welding based on dynamic resistance. Weld J 76:383s–390sGoogle Scholar
- 14.Doubov AA (2002) Express method of quality control of a spot resistance welding with usage of metal magnetic memory. Weld World Le Soudage Dans Le Monde 46:317–320Google Scholar
- 16.Podržaj, Primož, and Samo Simončič (2011) Resistance spot welding control based on fuzzy logic. Int J Adv Manuf Technol, 52: 959–967Google Scholar
- 18.Zhang ZD, Li DQ, Yin XH (2001) Study on spot welding quality monitoring models by linear regression theory. Trans China Weld Inst 22:31–35Google Scholar
- 19.Zhang ZD, Li DQ, Tang DJ, Li XJ (2004) Measures for decreasing errors of ANN models of spot welding quality monitor. Trans China Weld Inst 25:113–116Google Scholar
- 23.Matsuyama K. (1997). Nugget size sensing of spot weld based on neural network learning. Int Inst Weld, Doc. No.III- 1081-97 Google Scholar
- 24.Monari G et al. (1998) Prediction of spot welding diameter using neural networks. Int Inst Weld, Doc.III-1108-98Google Scholar
- 25.Luo Z, Shan P, Hu SS, Lian JR, Yi XL, Xue WF (2003) Application of the wavelet packet and its energy spectrum to identify nugget splash during the aluminum alloys spot welding. China Weld 12:98–102Google Scholar
- 26.Resnikoff HL, Wells RO (1998) The Mallat algorithm. In: Wavelet Analysis. Springer, New York, NYGoogle Scholar