Industrial Application of Deep Neural Network for Aluminum Casting Defect Detection in Case of Unbalanced Dataset

• Autorzy: Awtoniuk Michał , Majerek Dariusz , Myziak Artur , Gajda Cyprian

Identyfikatory

DOI

10.12913/22998624/154963

• Języki publikacji: EN

Abstrakty

EN

We have developed a deep neural network for casting defect detection. The approach is original because it assumes the use of data related to the casting manufacturing process, i.e. measurement signals from the casting machine, rather than data describing the finished casting, e.g. images. The defects are related to the production of car engine heads made of silumin. In the current research we focused on the detection of defects related to the leakage of the casting. The data came from production plant in Poland. The dataset was unbalanced. It included nearly 38,500 observations, of which only 4% described a leak event. The work resulted in a deep network consisting of 22 layers. We assessed the classification accuracy using a ROC curve, an AUC index and a confusion matrix. The AUC value was 0.97 and 0.949 for the learning and testing dataset, respectively. The model allowed for an ex-post analysis of the casting process. The analysis was based on Shapley values. This makes it possible not only to detect the occurrence of a defect but also to give potential reasons for the appearance of a casting leak.

Słowa kluczowe

EN

machine learning; deep neural network; classification; casting defect; casting defect detection

• Wydawca: Lublin University of Technology ; Polish Society of Ecological Engineering (PTIE), Branch of PTIE in Lublin
• Czasopismo: Advances in Science and Technology. Research Journal
• Rocznik: 2022
• Tom: Vol. 16, no 5
• Strony: 120--128
• Opis fizyczny: Bibliogr. 26 poz., fig., tab.

Twórcy

autor

Awtoniuk Michał

• Institute of Mechanical Engineering, Warsaw University of Life Sciences, Nowoursynowska 166, 02-787 Warsaw, Poland

• https://orcid.org/0000-0002-3771-992X

autor

Majerek Dariusz

• Faculty of Fundamentals of Technology, Lublin University of Technology, Nadbystrzycka 38, 20-618 Lublin, Poland

• https://orcid.org/0000-0002-0035-7187

autor

Myziak Artur

• MyFee Sp. z o. o., Wojrowicka 49/1, 54-436 Wrocław, Poland

autor

Gajda Cyprian

• IoTSolution Sp. z o. o., Zana 39 A, 20-634 Lublin, Poland

Bibliografia

• 1. Salat R., Awtoniuk M. Black box modeling of PIDs implemented in PLCs without structural information: a support vector regression approach. Neural Computing and Applications 2015; 26: 723–34.
• 2. Zając K., Płatek K., Biskup P., Łatka L. Modelling of hardfacing layers deposition parameters using robust machine learning algorithms. Journal of Physics: Conference Series 2021; 2130:012016.
• 3. Kulisz M., Zagórski I., Matuszak J., Kłonica M. Properties of the Surface Layer After Trochoidal Milling and Brushing: Experimental Study and Artificial Neural Network Simulation. Applied Sciences 2020; 10:75.
• 4. Szala M, Łatka L, Awtoniuk M, Winnicki M, Michalak M. Neural Modelling of APS Thermal Spray Process Parameters for Optimizing the Hardness, Porosity and Cavitation Erosion Resistance of Al2O3-13 wt% TiO2 Coatings. Processes 2020; 8:1544.
• 5. Szeląg B., Suligowski R., De Paola F., Siwicki P., Majerek D., Łagód G. Influence of urban catchment characteristics and rainfall origins on the phenomenon of stormwater flooding: Case study. Environmental Modelling & Software 2022; 150:105335.
• 6. Pizoń J., Kulisz M., Lipski J. Matrix profile implementation perspective in Industrial Internet of Things production maintenance application. Journal of Physics: Conference Series 2021; 1736:012036.
• 7. Awtoniuk M., Nowakowski T., Chlebowski J., Świętochowski A., Dąbrowska M., Klonowski J., et al. Internet of Things as an element of the frost protection system in orchards. Journal of Physics: Conference Series 2021; 2130:012015.
• 8. He H., Garcia EA. Learning from Imbalanced Data. IEEE Transactions on Knowledge and Data Engineering 2009; 21:1263–84.
• 9. Mohammed R., Rawashdeh J., Abdullah M. Machine Learning with Oversampling and Undersampling Techniques: Overview Study and Experimental Results. In: Proc. of 11th International Conference on Information and Communication Systems (ICICS), Irbid, Jordan 2020, 243–248.
• 10. He H., Bai Y., Garcia EA., Li S. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. In: Proc. of IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence), Hong Kong 2008, 1322–1328.
• 11. Menardi G., Torelli N. Training and assessing classification rules with imbalanced data. Data Mining and Knowledge Discovery 2014; 28: 92–122.
• 12. Chawla NV., Bowyer KW., Hall LO., Kegelmeyer WP. SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research 2002; 16: 321–357.
• 13. Mery D. Aluminum Casting Inspection Using Deep Learning: A Method Based on Convolutional Neural Networks. Journal of Nondestructive Evaluation 2020; 39: 12.
• 14. Jiang L., Wang Y., Tang Z., Miao Y., Chen S.. Casting defect detection in X-ray images using convolutional neural networks and attention-guided data augmentation. Measurement 2021; 170: 108736.
• 15. Nikolić F., Štajduhar I., Čanađija M. Casting Defects Detection in Aluminum Alloys Using Deep Learning: a Classification Approach. International Journal of Metalcasting 2022.
• 16. Lin J., Ma L., Yao Y. Segmentation of casting defect regions for the extraction of microstructural properties. Engineering Applications of Artificial Intelligence 2019; 85: 150–163.
• 17. Du W., Shen H., Fu J., Zhang G., He Q. Approaches for improvement of the X-ray image defect detection of automobile casting aluminum parts based on deep learning. NDT & E International 2019; 107: 102144.
• 18. Yeo I-K., Johnson RA. A New Family of Power Transformations to Improve Normality or Symmetry. Biometrika 2000; 87: 954–959.
• 19. Zheng A., Casari A. Feature Engineering for Machine Learning: Principles and Techniques for Data Scientists. O’Reilly Media, 2018.
• 20. Chollet F., others. Keras. GitHub 2015. Available from: https://github.com/fchollet/keras.
• 21. R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing 2020. Available from: https://www.Rproject.org/.
• 22. Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters 2006; 27: 861–874.
• 23. Perkins NJ., Schisterman EF. The Inconsistency of “Optimal” Cutpoints Obtained using Two Criteria based on the Receiver Operating Characteristic Curve. American Journal of Epidemiology 2006; 163: 670–675.
• 24. Biecek P., Burzykowski T. Explanatory Model Analysis: Explore, Explain and Examine Predictive Models. Chapman and Hall/CRC, 2021.
• 25. Lundberg SM., Lee S-I. A Unified Approach to Interpreting Model Predictions. In: Proc. of 31st Conference on Neural Information Processing Systems (NIPS’17), Long Beach CA, USA 2017.
• 26. Dozat T. Incorporating Nesterov Momentum into Adam. In: Proc. of International Conference on Learning Representations (ICLR 2016), San Juan, Puerto Rico 2016.

Uwagi

Opracowanie rekordu ze środków MEiN, umowa nr SONP/SP/546092/2022 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2022-2023).