Font Size: a A A

Study Of Welding Seam Detection And Welding Torch Servo Mechanism

Posted on:2009-09-04Degree:MasterType:Thesis
Country:ChinaCandidate:H T WangFull Text:PDF
GTID:2178360278475859Subject:Materials Processing Engineering
Abstract/Summary:PDF Full Text Request
Welding seam tracking by vision sensor has always been an active study task in the welding seam tracking field. With the rapid development of microelectronics and the very large scale integrated circuit technology, it is initially significant to apply advanced embedded processor in the upgrading and reconstructing of the existing welding seam tracking equipment as a whole, such as DSP, FPGA and CPLD. In the development of this new welding seam tracker, the development of welding seam information processing hardware system is a long cycle, so how to study welding seam detection algorithms effectively is a noticeable problem, before the accomplishment of the hardware system development. On the other hand, the welding torch servo mechanism is part of the new welding seam tracker, thus the servo motor must be elaborately picked up, and the controller must also be properly designed, so that special requests of the mechanism can be met.Three standards have been put forward in the study of welding seam detection algorithms development and selection, on PC but orienting to DSP/FPGA platform: manageable portability, validity, and time and space complexity. Intel's Open Source Computer Vision Library, OpenCV, is an ideal development tool for welding seam detection algorithms, and the introduction of OpenCV into our study properly serves the manageable portability of cross-platform algorithms and programs, and eases our algorithms development. In the fast development environment which is built by VC++ and OpenCV, real welding seam images have been processed step by step. Different algorithms have been analyzed and compared according to validity standard and time and space standard in each step. Finally a set of portable valid algorithms that have low time and space complexity have been developed, and these algorithms apply powerful codes support for the welding seam information processing platform based on DSP/FPGA. In the welding torch servo mechanism, the brushless DC motor has been selected as the servo motor of welding torch for its brushless-ness, well reliability and controllability. A DSP+CPLD controller scheme has been designed, and their functions have been determined according to their own characteristics. The startup, positive and negative turn, speed calculation and PWM speed regulation of the brushless DC servo motor has been realized, through power drive circuit design, CPLD digital circuit design, DSP interface circuit design and DSP program design. It is proved by experiments that the DSP+CPLD controller is an advanced and reliable scheme, and sets a well basis for further study of the welding torch servo mechanism.
Keywords/Search Tags:Welding seam detection algorithm, OpenCV, Brushless DC motor, DSP, CPLD
PDF Full Text Request
Related items