부산대학교 도서관

The complexity of welding process signal brings trouble to monitoring and evaluation. This paper presents a novel approach of welding condition monitoring based on the actual collected welding pressure signals, which combines the ideas of signal coarsening and discrete Fréchet distance, and takes the similarity and mapping path between the pressure signals of a single welding process as the evaluation criteria. The coarse-grained signal is used to keep the original change trend of the signal and reduce the influence of random noise. The discrete Fréchet distance is used to quantitatively evaluate the distance between different unequal length signals. The experimental results show that the proposed solution can give the quantitative standard of similarity between different welding pressure signals and the signal mapping relationship. This research provides a new perspective for condition monitoring of welding process.