Monitoring Technology for Detecting the Cobwebbing-Foreign-Matter Entering the Gap Between Molds

抄録

In recent years, the industrial sector has increasingly relied on technology for monitoring the operating status of production and processing equipment as a measure to address labor shortages. In the plastic molding industry, which is the subject of this study, efforts have been made to track operational conditions by installing a variety of sensors capable of detecting different types of problems. Furthermore, it is evident that so-called DX manufacturing, in which sensor-derived information is utilized to achieve efficient production through digital technologies such as AI and the Internet of Things, will become mainstream in the future. In this study, attention was directed to the cobwebbing phenomenon, a defect observed in injection molding, with a particular focus on its monitoring technology. Although the diameter of the threads pinched during the cobwebbing phenomenon is approximately 0.1 mm, early detection is critical because the defect can cause mold damage and deterioration. To address this issue, strain data were obtained from a mold equipped with a strain gauge, and discrimination technology was investigated by comparing the normal state with the pinched state using the Mahalanobis distance, a pattern recognition method. The results indicated that it may be possible to detect a thread with a diameter of about 0.05 mm if it becomes trapped during molding, and certain mold features that enhance discrimination accuracy were identified. Therefore, this study reports an example of monitoring technology for the entrapment of cobwebbing, specifically foreign matter entering during the molding process.