Abstract
This paper introduces a new diagnostic approach for concurrent discrete event systems (DES) in automated production plants. The proposed diagnostic procedure is based on Petri net models of the DES. Discrepancies between the Petri net models and event sequences, which are observed during the operation of the production plant, are detected as potential faults. A case-based diagnosis is then carried out to identify fault types that are easy to interpret for humans. The proposed diagnostic procedure is evaluated using an industrial conveying system.
Zusammenfassung
In diesem Beitrag wird ein neuer Diagnoseansatz für nebenläufige diskrete Ereignissysteme (DES) in automatisierten Produktionsanlagen eingeführt. Das vorgeschlagene Diagnoseverfahren basiert auf einer Modellierung der DES mit Petri-Netzen. Diskrepanzen zwischen dem Verhalten der Petri-Netze und den Ereignisfolgen, die während des Betriebs der Produktionsanlage erfasst werden, werden als potentielle Fehler detektiert. Weiterhin wird eine Fall-basierte Diagnose durchgeführt, um für den Menschen leicht zu interpretierende Fehlertypen zu identifizieren. Das vorgeschlagene Diagnoseverfahren wird anhand eines industriellen Transportsystems evaluiert.
About the author
Stefan Windmann received the Dipl.-Ing. and Dipl.-Inf. degrees in electrical engineering and technical computer sciences from University of Paderborn, Germany, in 2004, where he received the Ph.D. degree in electrical engineering in 2008. He is currently employed as senior scientist at Fraunhofer IOSB-INA in Lemgo, Germany. His current research interests include machine learning algorithms and methods for diagnosis and optimization of automated production systems.
-
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
-
Research funding: None declared.
-
Conflict of interest statement: The authors declare no conflicts of interest regarding this article.
Appendix: Relationship table based fault detection
In this work, the RT-based fault detection approach proposed in [6] has been extended to detect not only incorrect event sequences but also timing errors. Therefore, the relationship table, which captures the sequence relations between the events, has been replaced by two matrices
| Algorithm 3: Learning the relationship tables |
|---|
| Given: Event sequences Γ = {E 1 … E m } |
|
Result: Relationship tables
|
| (0)
|
| (1) for i = 1 … m: |
| (2) ∀e ∈ E i : |
| (3) if σ −1(e, 1) ∈ E i : |
| (4)
|
| (5)
|
| (6)
|
| (7)
|
| (8)
|
All entries of the matrices
| Algorithm 4: RT-based fault detection |
|---|
|
Given: Event sequences Γ = {E
1 … E
m
} Relationship tables
|
| Result: “‘ok”’ or “‘failure”’ |
| (1) ∀i ∈ {1 … m} : ∀e ∈ E i : |
| (2) ∀l ∈ {1 … L}|σ −1(e, l) ∈ E i : |
| (3)
|
| (4)
|
|
|
| (5) return “‘failure”’ |
The basic principle is to check for each event e in the current event sequence E
i
whether the time constraints defined by the matrices
are allowed. In this work, the parameters L = 10, δ = 0.1, δ c = 1 have been used, which have been optimized in informal experiments.
References
[1] Z. Gao, C. Cecati, and S. X. Ding, “A survey of fault diagnosis and fault tolerant techniques,” IEEE Trans. Ind. Electron., vol. 62, no. 6, pp. 3757–3767, 2015. https://doi.org/10.1109/tie.2015.2417501.Search in Google Scholar
[2] Y. J. Park, S. K. S. Fan, and C. Y. Hsu, “A review on fault detection and process diagnostics in industrial processes,” Processes, vol. 8, no. 9, 2020, Art. no. 19. https://doi.org/10.3390/pr8091123.Search in Google Scholar
[3] A. Vodenčarević, H. Kleine Büning, O. Niggemann, and A. Maier, “Using behavior models for anomaly detection in hybrid systems,” in Proceedings of the 23rd International Symposium on Information, Communication and Automation Technologies-ICAT 2011, 2011.10.1109/ICAT.2011.6102093Search in Google Scholar
[4] A. Vodenčarević, H. Kleine Buning, O. Niggemann, and A. Maier, “Identifying behavior models for process plants,” in Emerging Technologies & Factory Automation (ETFA), IEEE 16th Conference on, 2011, pp. 1–8.10.1109/ETFA.2011.6059080Search in Google Scholar
[5] D. N. Pandalai and L. E. Holloway, “Template languages for fault monitoring of timed discrete event processes,” IEEE Trans. Autom. Control, vol. 45, no. 5, pp. 868–882, 2000. https://doi.org/10.1109/9.855548.Search in Google Scholar
[6] E. Laftchiev, D. Nikovski, X. Sun, and A. Dau, “Anomaly detection in discrete manufacturing systems using event relationship tables,” in Proceedings of The International Workshop on Principles of Diagnosis, Warsaw, CEUR-WS, 2018.Search in Google Scholar
[7] F. Basile, “Overview of fault diagnosis methods based on Petri net models,” in European Control Conference (ECC), 2014.10.1109/ECC.2014.6862631Search in Google Scholar
[8] G. Zhu, L. Feng, Z. Li, and N. Wu, “An efficient Fault Diagnosis approach based on integer linear programming for labeled Petri nets,” IEEE Trans. Autom. Control, vol. 66, pp. 2393–2398, 2021. https://doi.org/10.1109/tac.2020.3008712.Search in Google Scholar
[9] M. P. Cabasino, A. Giua, and C. Seatzu, “Fault detection for discrete event systems using Petri nets with unobservable transitions,” Automatica, vol. 46, pp. 1531–1539, 2010. https://doi.org/10.1016/j.automatica.2010.06.013.Search in Google Scholar
[10] Y. Wang, L. Yin, and G. Zhu, “Online Fault diagnosis of labeled Petri nets based on reachability graphs and topological sorting,” IEEE Access, vol. 8, pp. 162363–162372, 2020. https://doi.org/10.1109/access.2020.3021476.Search in Google Scholar
[11] F. Basile, M. P. Cabasino, and C. Seatzu, “State estimation and fault diagnosis of labeled time Petri net systems with unobservable transitions,” IEEE Trans. Autom. Control, vol. 4, pp. 997–1009, 2015. https://doi.org/10.1109/tac.2014.2363916.Search in Google Scholar
[12] X. Wang, C. Mahulea, and M. Silva, “Fault diagnosis graph of time Petri nets,” in Proc. Eur. Control Conf. (ECC 13), 2013.10.23919/ECC.2013.6669417Search in Google Scholar
[13] D. Lime and O. H. Roux, “Model checking of time Petri nets using thestate class timed automaton,” Discrete Event Dyn. Syst., vol. 16, pp. 179–205, 2006. https://doi.org/10.1007/s10626-006-8133-9.Search in Google Scholar
[14] G. Zhu, Z. Li, N. Wu, and A. Al-Ahmari, “Fault identification of discrete event systems modeled by Petri nets with unobservable transitions,” IEEE Trans. Syst. Man Cybern., vol. 49, pp. 333–345, 2019. https://doi.org/10.1109/tsmc.2017.2762823.Search in Google Scholar
[15] J. Ladiges, A. Fuelber, E. Arroyo, A. Fay, C. Haubeck, and W. Lamersdorf, “Learning material flow models for manufacturing plants from data traces,” in 2015 IEEE 13th International Conference on Industrial Informatics (INDIN), 2015, pp. 294–301.10.1109/INDIN.2015.7281750Search in Google Scholar
[16] S. Windmann, “Learning Petri net models from sensor data of conveying systems based on the merging of prefix and postfix trees,” AT-Automatisierungstechnik, vol. 70, pp. 142–153, 2022. https://doi.org/10.1515/auto-2021-0101.Search in Google Scholar
[17] A. Azarian, A. Siadat, and P. Martin, “Optimization of a knowledge-based system by a meta-heuristic approach for the automotive diagnosis,” in Industrial Engineering and Engineering Management (IEEM), 2011 IEEE International Conference on, 2011.10.1109/IEEM.2011.6118190Search in Google Scholar
[18] Y. Pan, Q. Loi, R. Zhang, L. Dong, and X. Zhang, “Test point optimization for model-based fault diagnosis expert system,” in Proceedings of The Automatic Control and Artificial Intelligence (ACAI 2012), Xiamen, Institution of Engineering and Technology (IET), 2012.Search in Google Scholar
[19] R. Guo, L. Cheng, J. Li, R. Hahn, and H. Liu, “A survey of learning causality with data: problems and methods,” ACM Comput. Surv., vol. 53, pp. 1–37, 2020. https://doi.org/10.1145/3397269.Search in Google Scholar
[20] C. Kühnert, Data-driven Methods for Fault Localization in Process Technology, Karlsruhe, KIT Scientific Publishing, 2013.Search in Google Scholar
[21] J. de Kleer, B. Janssen, D. G. Bobrow, T. K. B. Saha, N. R. Moore, and S. Sutharshana, “Fault augmented modelica models,” in Proceedings of The 24th International Workshop on Principles of Diagnosis, Jerusalem, CEUR-WS, 2013, pp. 71–78.Search in Google Scholar
[22] I. Roychoudhury, M. J. Daigle, G. Biswas, and X. Koutsoukos, “Efficient simulation of hybrid systems: a hybrid bond graph approach,” Simulation, vol. 87, pp. 467–498, 2011.10.1177/0037549710364478Search in Google Scholar
[23] A. Diedrich, A. Maier, and O. Niggemann, “Model-based diagnosis of hybrid systems using satisfiability modulo theory,” in AAAI Conference on Artificial Intelligence, 2019.10.1609/aaai.v33i01.33011452Search in Google Scholar
[24] X. Berjaga, A. Pallares, and J. Melendez, “A framework for case-based diagnosis of batch processes in the principal components space,” in Emerging Technologies Factory Automation, 2009. ETFA 2009. IEEE Conference on, 2009, pp. 1–9.10.1109/ETFA.2009.5347075Search in Google Scholar
[25] J. Gao, W. Xiao, Y. Xie, F. Gu, and B. Yao, “An intelligent fault diagnosis approach integrating cloud model and CBR,” in 6th International Conference on Computer Science and Network Technology, 2017.10.1109/ICCSNT.2017.8343705Search in Google Scholar
[26] K. Balzereit and O. Niggemann, “Sound and complete reconfiguration for a class of hybrid systems,” in Proceedings of The 32nd International Workshop on Principles of Diagnosis, Hamburg, CEUR-WS, 2021.Search in Google Scholar
[27] A. Feldman, G. Provan, and A. Van Gemund, “Approximate model-based diagnosis using greedy stochastic search,” J. Artif. Intell. Res., vol. 38, pp. 371–413, 2010. https://doi.org/10.1613/jair.3025.Search in Google Scholar
© 2023 Walter de Gruyter GmbH, Berlin/Boston
