|
CSME 2020/12
Volume 41 No.6
:
799-806
|
|
Condition Monitoring Technique for Injection Molding Process byWaveform-AnalysisMethod
Yoshio Fukushima a, Zhao Zongyang , Akiko Sakamoto b and Naoki Kawada a
aDepartment of Mechanical Engineering, Saitama Institute of Engineering, Fukaya Saitama, Japan.
bAdvanced Science Lab., Saitama Institute of Engineering, Fukaya Saitama, Japan.
|
Abstract:
IoT technology has spread rapidly all over the world in recent years. Many companies and engineers are paying attention to introduce IoT technology to a practical production process. To avoid some defects of injection molding easily and efficiently, the production process implemented by IoT technology can be useful. Most modern injection machines include the function that can measure some process information, such as pressure, temperature, clamping force, etc. This process information helps us to achieve a stable production system. In this study, in addition to the process information, the data of the acceleration sensor mounted to the mold has also been measured. By utilizing this information, process information and acceleration data, the waveform analysis has been carried out. Also, by using pattern recognition technology called MT-system, the method of discriminating defects has come to be realized. The important value called MD, Mahalanobis Distance, in MT-system was applied. In conclusion, it has been clarified that the change of the waveform and the quality of the molded parts have a relation to each other (e.g., acceleration, clamping force of the mold). From these things, it is found that the condition monitoring that can detect the injection process failure and the molded parts defects is possible by using MT-System.
|
|
Keywords: injection molding process, condition monitoring, waveform analysis, failure discrimination
|
Download PDF
|
|
*Corresponding author; e-mail:
|
©
2020
CSME , ISSN 0257-9731
|