DOI QR코드

DOI QR Code

다중 작업 학습 구조 기반 공정단계별 공정조건 및 성형품의 품질 특성을 반영한 사출성형품 품질 예측 신경망의 성능 개선에 대한 연구

A study on the performance improvement of the quality prediction neural network of injection molded products reflecting the process conditions and quality characteristics of molded products by process step based on multi-tasking learning structure

  • 이효은 (한국생산기술연구원 금형성형연구부문) ;
  • 이준한 (한국생산기술연구원 금형성형연구부문) ;
  • 김종선 (한국생산기술연구원 금형성형연구부문) ;
  • 조구영 (단국대학교 기계공학과)
  • Hyo-Eun Lee (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Jun-Han Lee (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Jong-Sun Kim (Department of Molding & Metal Forming R&D, Korea Institute of Industrial Technology) ;
  • Gu-Young Cho (Department of Mechanical Engineering, Dankook University)
  • 투고 : 2023.12.14
  • 심사 : 2023.12.31
  • 발행 : 2023.12.31

초록

Injection molding is a process widely used in various industries because of its high production speed and ease of mass production during the plastic manufacturing process, and the product is molded by injecting molten plastic into the mold at high speed and pressure. Since process conditions such as resin and mold temperature mutually affect the process and the quality of the molded product, it is difficult to accurately predict quality through mathematical or statistical methods. Recently, studies to predict the quality of injection molded products by applying artificial neural networks, which are known to be very useful for analyzing nonlinear types of problems, are actively underway. In this study, structural optimization of neural networks was conducted by applying multi-task learning techniques according to the characteristics of the input and output parameters of the artificial neural network. A structure reflecting the characteristics of each process step was applied to the input parameters, and a structure reflecting the quality characteristics of the injection molded part was applied to the output parameters using multi-tasking learning. Building an artificial neural network to predict the three qualities (mass, diameter, height) of injection-molded product under six process conditions (melt temperature, mold temperature, injection speed, packing pressure, pacing time, cooling time) and comparing its performance with the existing neural network, we observed enhancements in prediction accuracy for mass, diameter, and height by approximately 69.38%, 24.87%, and 39.87%, respectively.

키워드

과제정보

본 연구는 중소기업벤처부의 스마트제조혁신기술개발사업 (Project No.00140364, SE230069)이 지원한 것이다.

참고문헌

  1. Sung, S. M., Jun, J. S. A study on the motorcycle lear cowl injection molding by CAE analysis, Design & Manufacturing, Vol. 13:4 34-39. 2019.
  2. Yoo, M. J., Kim, K. A., Han, S. R., Optimization of feed system of base mold for washing machine using CAE, Design & Manufacturing, Vol. 15:1, 1-7, 2021.
  3. Hong, S. K., Implementation of an simulation-based digital twin for the plastic blow molding process, Design & Manufacturing, Vol. 17:3 1-7. 2023.
  4. Lee, D. W., Kim, J. S., Lee, H. H., Lee, S. H., A study on CAE and injection molding of automotive thick-walled light guide with micro-optical patterns Design & Manufacturing, Vol. 17:3, 8-14. 2023.
  5. Koh, B. W., Kim, J. S., Choi, H. G., Estimation of Process Parameters Using QFD and Neural Networks in Injection Molding, Ind. Eng. Interfaces, Vol. 21:2, 221-228, 2008.
  6. Rhee, B. O., Tae, J. S., Choi, J. H., Injection Mold Cooling Circuit Optimization by Back-Propagation Algorithm, J. Korean Soc. Manuf. Technol. Eng., Vol. 18:4, 430-435, 2009.
  7. Lee, J. H., Yang, D. C., Yoon, K. H., Kim, J. S., Effects of Input Parameter Range on the Accuracy of Artificial Neural Network Prediction for the Injection Molding Process. Polym., Vol 14, 1724, 2022.
  8. Li, L., Jamieson, K., DeSalvo, G., Rostamizadeh, A., Talwalkar, A., Hyperband: A novel bandit-based approach to hyper-parameter optimization. J. Mach. Learn. Res. Vol. 18, 6765-6816. 2017.
  9. Kingma, D. P., Ba, J., Adam: A method for stochastic optimization., arXiv, 2014.