DEVELOPMENT OF A SYSTEM FOR OPERATIONAL CONTROL AND QUALITY ASSURANCE OF PRODUCTS PRODUCED BY ADDITIVE MANUFACTURING (DMLS) USING COMPUTER VISION AND MACHINE LEARNING
Keywords:
Additive manufacturing, DMLS, quality management, operational control, computer vision, machine learning, predictive qualityAbstract
With the increasing adoption of additive manufacturing, in particular direct laser sintering of metal (DMLS), in high-tech industries, ensuring stable and predictable product quality is becoming an important task. Existing control methods, mainly post-process, do not allow effective prevention of defects in real time. In this paper, the concept of an integrated system of operational quality control of DMLS products is proposed. The system is based on the collection of high-frequency sensory data (visual, thermal) during the formation of each layer, their subsequent intelligent processing using computer vision and machine learning to automatically detect anomalies, classify defects and predict quality. The potential of this approach for the transition to proactive quality management, defect reduction and optimization of production cycles is substantiated.
References
Луценко О.В. Об оценке эффективности внедрения метода «аутсорсинг» / О.В. Луценко, Д.А. Комаренко // Актуальные вопросы и перспективы развития науки, техники и технологии: материалы Международной научно-практической конференции. – Казань, 2020. – С. 103-109.
Wohlers T., Gornet P. Wohlers Report 2020: 3D Printing and Additive Manufacturing State of the Industry. – [Fort Collins, CO] : Wohlers Associates, 2020. – [21] с.
Additive manufacturing of metals / D. Herzog, M. Seyda, E. Wycisk, C. Emmelmann // Acta Materialia. – 2016. – Vol. 117. – P. 371–392.
Additive manufacturing of metals: With an eye toward the next frontier / T. DebRoy [и др.] // Journal of Manufacturing Science and Engineering. – 2018. – Vol. 140, no. 5. – P. 051011.
Artificial intelligence and digital twin for smart manufacturing quality control: A review / L. Wang, R. Li, H. Lv // International Journal of Production Research. – 2021. – Vol. 59, no. 12. – P. 3601–3618.
Review of in-situ process monitoring and in-situ metrology for metal additive manufacturing / S. K. Everton [и др.] // Materials & Design. – 2016. – Vol. 95. – P. 431–445.
Anomaly detection in melt pool imagery from additive manufacturing via a convolutional neural network / L. Scime, J. Beuth // The International Journal of Advanced Manufacturing Technology. – 2018. – Vol. 95. – P. 1229–1241.
Prediction of melt pool dimensions in selective laser melting via a random forest algorithm / G. Tapia, A. Elwany // Journal of Manufacturing Science and Engineering. – 2014. – Vol. 136, no. 6. – P. 061017.
Deep Learning for Industrial Visual Inspection: A Survey / Y. Luo, X. Li // IEEE Transactions on Industrial Informatics. – 2020. – Vol. 16, no. 2. – P. 779–792.
