Abstract
Machine learning and artificial intelligence provide methods and algorithms to take advantage of sensor and actuator data in automated production systems. Product quality monitoring is one of the promising applications available for data-driven modeling, particularly in cases where the quality parameters cannot be measured with reasonable effort. This is the case for defects such as cracks in workpieces of hydraulic metal powder presses. However, the variety of shapes produced at a powder press requires training of individual models based on a minimal sample size of unlabeled data to adapt to changing settings. Therefore, this article proposes an unsupervised product quality monitoring approach based on dynamic time warping and non-linear regression to detect anomalies in unlabeled sensor and actuator data. A preprocessing step that isolates only the relevant intervals of the process is further introduced, facilitating efficient product quality monitoring. The evaluation on an industrial dataset with 37 samples, generated in test runs, shows a true-positive rate for detected product quality defects of 100% while preserving an acceptable accuracy. Moreover, the approach achieves the output within less than 10 seconds, assuring that the result is available before the next workpiece is processed. In this way, efficient product quality management is possible, reducing time- and cost-intensive quality inspections.
- B. Vogel-Heuser and D. Hess. 2016. Guest editorial industry 4.0–Prerequisites and visions. IEEE Transactions on Automation Science and Engineering 13, 2 (2016), 411–413. DOI:10.1109/TASE.2016.2523639Google Scholar
Cross Ref
- ASM International. 2002. Powder Metal Technologies and Applications (10th ed). ASM International, Materials Park, OH.Google Scholar
- G. Wang, A. Ledwoch, R. M. Hasani, R. Grosu, and A. Brintrup. 2019. A generative neural network model for the quality prediction of work in progress products. Applied Soft Computing 85 (2019), 106683. DOI:10.1016/j.asoc.2019.105683Google Scholar
Digital Library
- R. D. Labati, A. Genovese, E. Muñoz, V. Piuri, and F. Scotti. 2018. Applications of computational intelligence in industrial and environmental scenarios. In Learning Systems: From Theory to Practice. Studies in Computational Intelligence, Vol. 756. Springer, 29–46.Google Scholar
- I. Kirchen, B. Vogel-Heuser, O. Hildenbrand, R. Schulte, M. Vogel, M. Lechner, and M. Merklein. 2017. Data-driven model development for quality prediction in forming technology. In Proceedings of the 2017 IEEE 15th International Conference on Industrial Informatics (INDIN’17). 775–780. DOI:10.1109/INDIN.2017.8104871Google Scholar
- L. Yao and Z. Ge. 2019. Nonlinear Gaussian mixture regression for multimode quality prediction with partially labeled data. IEEE Transactions on Industrial Informatics 15, 7 (2019), 4044–4053. DOI:10.1109/TII.2018.2885363Google Scholar
Cross Ref
- X. Li, F. Wu, R. Zhang, and F. Gao. 2019. Nonlinear multivariate quality prediction based on OSC-SVM-PLS. Industrial & Engineering Chemistry Research 58, 19 (2019), 8154–8161. DOI:10.1021/acs.iecr.8b06079Google Scholar
Cross Ref
- S. Lalam, P. K. Tiwari, S. Sahoo, and A. K. Dalal. 2019. Online prediction and monitoring of mechanical properties of industrial galvanised steel coils using neural networks. Ironmaking & Steelmaking 46, 1 (2019), 89–96. DOI:10.1080/03019233.2017.1342424Google Scholar
Cross Ref
- C. Gonzalez-Val, A. Pallas, V. Panadeiro, and A. Rodriguez. 2020. A convolutional approach to quality monitoring for laser manufacturing. Journal of Intelligent Manufacturing 31, 3 (2020), 789–795. DOI:10.1007/s10845-019-01495-8Google Scholar
Cross Ref
- S. A. Shevchik, T. L. Quang, F. V. Farahani, F. Neige, B. Meylan, S. Zanoli, and K. Wasmer. 2019. Laser welding quality monitoring via graph support vector machine with data adaptive kernel. IEEE Access 7 (2019), 93108–93122. DOI:10.1109/ACCESS.2019.2927661Google Scholar
Cross Ref
- Y. Oh, M. Busogi, K. Ransikarbum, D. Shin, D. Kwon, and N. Kim. 2019. Real-time quality monitoring and control system using an integrated cost effective support vector machine. Journal of Mechanical Science and Technology 33, 12 (2019), 6009–6020. DOI:10.1007/s12206-019-1145-9Google Scholar
Cross Ref
- J. P. Liu, O. F. Beyca, P. K. Rao, Z. J. Kong, and S. T. S. Bukkapatnam. 2017. Dirichlet process Gaussian mixture models for real-time monitoring and their application to chemical mechanical planarization. IEEE Transactions on Automation Science and Engineering 14, 1 (2017), 208–221. DOI:10.1109/TASE.2016.2599436Google Scholar
Cross Ref
- Y. Tan, C. Niu, H. Tian, L. Hou, and J. Zhang. 2019. A one-class SVM based approach for condition-based maintenance of a naval propulsion plant with limited labeled data. Ocean Engineering 193 (2019), 106592. DOI:10.1016/j.oceaneng.2019.106592.Google Scholar
Cross Ref
- Y. Xiao, H. Wang, W. Xu, and J. Zhou. 2016. Robust one-class SVM for fault detection. Chemometrics and Intelligent Laboratory Systems 151 (2016), 15–25. DOI:10.1016/j.chemolab.2015.11.010Google Scholar
Cross Ref
- H. Y. Yu, J. L. Tang, and Y. Tang. 2015. Research on 800 MN hydraulic press monitoring system. Materials Research Innovations 19, S6 (2015), S6-181–S6-183. DOI:10.1179/1432891715Z.0000000001474Google Scholar
Cross Ref
- H. Chen, J. P. Tan, J. L. Gong, Z. Q. Shu, X. Y. Cao, and J. N. Zhou. 2011. Design of new condition monitoring system and data analysis in large forging die hydraulic press. Advanced Materials Research 311-313 (2011), 1546–1550. DOI:10.4028/www.scientific.net/AMR.311-313.1546Google Scholar
- D. Sowmiya. 2013. Monitoring and control of a PLC based VFD fed three phase induction motor for powder compacting press machine. In Proceedings of the 2013 7th International Conference on Intelligent Systems and Control (ISCO’13). 90–92.Google Scholar
Cross Ref
- J. A. Ferreira, P. Sun, and J. J. Gracio. 2006. Design and control of a hydraulic press. In Proceedings of the IEEE International Symposium on Computer-Aided Control Systems Design. 814–819.Google Scholar
- M. I. Rȋpanu, G. Nagîț, V. Merticaru, M. A. Mihalache, M. Boca, and V. Huşanu. 2014. An optimized methodology for process quality analysis and monitoring activities in case of sheet metal bearing cages stamping. Applied Mechanics and Materials 657 (2014), 183–187. DOI:10.4028/www.scientific.net/AMM.657.183Google Scholar
Cross Ref
- H. Du, B. Huang, L. Wang, and S. M. Chen. 2014. The design of monitoring system in large hydraulic press aiming at the precise closed-loop control. Applied Mechanics and Materials 989–994 (2014), 3062–3069. DOI:10.4028/www.scientific.net/AMR.989-994.3062Google Scholar
- B. Vogel-Heuser, V. Karaseva, J. Folmer, and I. Kirchen. 2017. Operator knowledge inclusion in data-mining approaches for product quality assurance using cause-effect graphs. In Proceedings of the 20th IFAC World Congress. 1358–1365. DOI:10.1016/j.ifacol.2017.08.233Google Scholar
- D. Pantförder, J. Schaupp, and B. Vogel-Heuser. 2017. Making implicit knowledge explicit—Acquisition of plant staff's mental models as a basis for developing a decision support system. In Proceedings of HCI International 2017: Posters’ Extended Abstracts. 358–365.Google Scholar
- E. I. Papageorgiou. 2012. Learning algorithms for fuzzy cognitive maps—A review study. IEEE Transactions on Systems, Man, and Cybernetics: Systems 42, 2 (2012), 150–163. DOI:10.1109/TSMCC.2011.2138694 Google Scholar
Digital Library
- M. Anisheh and H. Hassanpour. 2009. Adaptive segmentation with optimal window length scheme using fractal dimension and wavelet transform. International Journal of Engineering 22, 3 (2009), 257–268. http://www.ije.ir/article_71799.html.Google Scholar
- H. Azami, K. Mohammadi, and B. Bozorgtabar. 2012. An improved signal segmentation using moving average and Savitzky-Golay filter. Journal of Signal and Information Processing 3, 1 (2012), 39–44. DOI:10.4236/jsip.2012.31006Google Scholar
Cross Ref
- R. Esteller, G. Vachtsevanos, J. Echauz, and B. Litt. 2001. A comparison of waveform fractal dimension algorithms. IEEE Transactions on Circuits and Systems I 48, 2 (2001), 177–183. DOI:10.1109/81.904882Google Scholar
- M. Riedl, A. Müller, and N. Wessel. 2013. Practical considerations of permutation entropy. European Physical Journal Special Topics 222, 2 (2013), 249–262. DOI:10.1140/epjst/e2013-01862-7Google Scholar
Cross Ref
- H. Sakoe and S. Chiba. 1978. Dynamic programming algorithm optimization for spoken word recognition. IEEE Transactions on Acoustics, Speech, and Signal Processing 26, 1 (1978), 43–49. DOI:10.1109/TASSP.1978.1163055Google Scholar
Cross Ref
- H. Ding, G. Trajcevski, P. Scheuermann, X. Wang, and E. Keogh. 2008. Querying and mining of time series data. Proceedings of the VLDB Endowment 1, 2 (2008), 1542–1552. Google Scholar
Digital Library
- R. L. Nuzzo. 2016. The box plots alternative for visualizing quantitative data. PM & R: The Journal of Injury, Function, and Rehabilitation 8, 3 (2016), 268–272. DOI:10.1016/j.pmrj.2016.02.001Google Scholar
Cross Ref
Index Terms
Product Quality Monitoring in Hydraulic Presses Using a Minimal Sample of Sensor and Actuator Data






Comments