Study On The Technology Of Workpiece Assembly Alignment In Video Frequency Image

The nozzle is an important part of the solid rocket engine, the nozzle and the engine main body is a discrete state, it requests the connection holes on the face of nozzle flange and engine flange align when assembling. At present, the solid rocket engine's nozzle assembly still selected the artificial alignment method, which already could not meet production department's need for consuming, low accuracy, instability, the eyes be easily weary and be restricted by the environment. So a kind of automatic precise alignment method is urgently awaited. Therefore, the study of workpiece assembly alignment has certain practical significance based on the video frequency image.Through analyzing and researching various existing principles and characteristics of assembly alignment technology, according to the production line structure, the overall project of assembly alignment system based on the video frequency image is designed. The alignment system examination includes the parallelism examination and the center examination. This article has established the plane parallelism examination mathematical model, elaborated the principle and the algorithm of the parallelism examination and the center examination in detail. The parallelism examination is completed by making use of the range finder sensor examining the distance between the range finder and nozzle and the engine flange to adjust two planes parallelism. The center examination is achieved through the CCD camera gathering circular connection hole image on the engine flange and nozzle flange, obtaining the connection hole edge data by carrying out a series of processing on the image such as filtering, the division to the image, extracting the edge and so on. The edge of the round hole is the important basis of central examination. Adopt the least squares fitting and the Hough transformation to detect the central coordinates and the diameter of the goal round holes, to align based on the data of central departure. The experimental result indicated that the design proposal is feasible, when the goal outline edge is incomplete, the examination error is big because of the limitation of the Hough transformation algorithm. The least squares fitting examination's uniformity is good, can meet the examination precision requirements.