PCB 커넥터의 치수 데이터 기반 형상복원 및 품질분류 모델 개발

This study proposes a product shape reconstruction and quality classification model based on dimension data as a methodology for improving existing quality management and inspection techniques. Image data analysis of products is drawing attention in quality inspection, but the method involves various limitations in terms of cost and time. This paper was written to reduce these limitations. The PCB connector discussed in this paper is mainly produced through an injection molding process. The main causes of shape defects in PCB connectors a re the f latness and uniformity of pinholes alignment. To examine these factors, we define a baseline at the top and bottom of the product. We then label the quality by measuring two aspects: the angle between the baseline and the product"s bottom surface, and the distance between the baseline and each pinhole. Afterwards, a quality classification model is designed with injection molding process data based on these quality labels. In future work, we anticipate the development of a quality classification model that utilizes image analysis techniques. This model will be based on the use of reconstructed shape image files.