The Research Of Magnetic-control SAW Seam Tracking Based On The Empirical Wavelet Transform

With the development of the country’s economic and the progress of the science technology, high quality and efficient automatic welding technology has become a hot research and focus. To achieve automatic and intelligence welding, the key technology lies in sampling and distilling the accurate weld deviation signals, and this is an urgent problem in seam tracking system. The article mainly researched the seam tracking system of magnetic submerge-arc welding,optimized sensor structure of the seam tracking, and structural parameters were optimized designed a filtering method for tracking the signal, and proposed a signal analysis method based on the experience of the wavelet transform. Ultimately optimized the parameters of the sensor structure to improve the tracking accuracy of the entire system. The main research contents are as follows:1、Analyzed and described the various types of sensors for automatic tracking of submerged arc welding, and also summarized the main information processing method for current signal analysis in submerged arc welding.2、Optimized the structure of seam tracking sensor in magnetron submerged arc welding, enhanced the stability on collecting the initial voltage signal. For the characteristics of the initial voltage signals, designed a filtering method for seam tracking signal in magnetic submerged arc, namely a method combined hardware filtering with wavelet transform,extracted the main waveform of the seam tracking signal;3、Analysed the impacts of the structure parameters of the new sensor on the tracking signal by experiments. And response surface methodology in design methods BBD model of multi-parameter optimization, get a good track waveform best parameter values.4、Presented weld tracking signal analysis method based on the experiential wavelet transform, proved the EWT is superior in nonlinear signal analysis function than the EMD through the simulation experiment. Eventually the EWT method is used in real-time voltage signal analysis of magnetic control arc welding seam tracking, it’s confirmed that the method can accurately extract seam tracking signal waveform, and the experiment sinusoidal tracking signal waveform are extremely closing to the theory, meanwhile obtained other characteristics of the low-frequency jamming signal.5、Introduced the compositions of the magnetron submerged arc welding seam tracking system, focused on analyzing seam tracking signal, reselected the microcontroller chips which have faster reaction rate and more powerful functions; Through the tracking signal extraction and tracking effect contrast test between the original filtering signal and the experience wavelet transform, got the better tracking effect and weld forming quality, and finally proved the effectiveness and accuracy of the method.