The Signal Processing And Endpoint Detection Of The Discontinuity Fillet Seam Based On Rotating Arc Sensor

There are a lot of robots working at the same time on welding field, In order to improve the automation degree, robots need to be able to detect weld end point automatically。Laser visual sensor detect endpoint of drainage holes based on that laser rays is divided into two segments by endpoint of drainage holes, there has to be a vertical artifacts。 And there is not vertical artifacts at the end point of general weld, laser rays cannot be divided into two segments, so it unable to detect a hole weld end point。 This paper mainly includes the following parts:First, Construct robot mechanical structure and the slide block。 Get the movement characteristic data of the robot body and cross slide block by experiment research。 Fitting equation is obtained by means of least square method。Secondly, The robot manually control system were designed. The hardware of robotic automatic tracking system have been set up. complete the debugging of hardware system and software system of the robot。 the robot control box were designed based on virtual prototype。 All control hardware integrated in the control box, make robots waterproof、dustproof and performance more stable。Thirdly, the rotating arc signal denoising experiment was carried out In a variety of classical filtering methods, and found that denoising effect is not ideal by the experimental results。 Through theoretical analysis and experimental verification, Found that for rotating arc signal denoising, the method of wavelet decomposition reconstruction is better than the method of Combination denoising method and Adaptive soft threshold wavelet denoising method。Finally, Analysis of weld rotating arc signal waveform characteristics before the endpoint and after the endpoint, Weld deviation be used detect endpoint is put forward。Verified the correctness and superiority by compared with other methods through the experiment。...