Time-Domain Characteristics Analysis of Penetration Arc Sound Signal in MIG Welding

Abstract

Many important time-domain characteristic parameters are extracted through the short-time analysis of arc sound signal in MIG butt welding with spray transfer, which can be used for the diagnosis on the weld penetration status. And the short-time autocorrelation function and short-time average amplitude difference function are adopted to pitch estimation. The analysis results show that the penetration status can be accurately recognized via the short-time energy, average magnitude, average zero-crossing and zero-to-energy and so on. Meanwhile, the pitch estimation of arc sound signal in experiments is at 220 points, that is 5 ms or 200 Hz in cycle. The methods and results provide a foundation for the diagnosis on penetration based on analysis of arc sound signal and have great theoretical meaning and practical value.