Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise
Castellani A, Schmitt S, Hammer B (2021)
In: Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I. Oliver N, Pérez-Cruz F, Kramer S, Read J, Lozano JA (Eds); Lecture Notes in Computer Science, 12975. Cham: Springer International Publishing: 469-484.
Konferenzbeitrag
| Veröffentlicht | Englisch
Download
Es wurden keine Dateien hochgeladen. Nur Publikationsnachweis!
Autor*in
Herausgeber*in
Oliver, Nuria;
Pérez-Cruz, Fernando;
Kramer, Stefan;
Read, Jesse;
Lozano, Jose A.
Einrichtung
Projekt
Abstract / Bemerkung
In complex industrial settings, it is common practice to monitor the operation of machines in order to detect undesired states, adjust maintenance schedules, optimize system performance or collect usage statistics of individual machines. In this work, we focus on estimating the power output of a Combined Heat and Power (CHP) machine of a medium-sized company facility by analyzing the total facility power consumption. We formulate the problem as a time-series classification problem, where the class label represents the CHP power output. As the facility is fully instrumented and sensor measurements from the CHP are available, we generate the training labels in an automated fashion from the CHP sensor readings. However, sensor failures result in mislabeled training data samples which are hard to detect and remove from the dataset. Therefore, we propose a novel multi-task deep learning approach that jointly trains a classifier and an autoencoder with a shared embedding representation. The proposed approach targets to gradually correct the mislabelled data samples during training in a self-supervised fashion, without any prior assumption on the amount of label noise. We benchmark our approach on several time-series classification datasets and find it to be comparable and sometimes better than state-of-the-art methods. On the real-world use-case of predicting the CHP power output, we thoroughly evaluate the architectural design choices and show that the final architecture considerably increases the robustness of the learning process and consistently beats other recent state-of-the-art algorithms in the presence of unstructured as well as structured label noise.
Erscheinungsjahr
2021
Titel des Konferenzbandes
Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I
Serien- oder Zeitschriftentitel
Lecture Notes in Computer Science
Band
12975
Seite(n)
469-484
Konferenz
ECML PKDD: Joint European Conference on Machine Learning and Knowledge Discovery in Databases
Konferenzort
Bilbao, Spain
Konferenzdatum
2021-09-13 – 2021-09-17
ISBN
978-3-030-86485-9
eISBN
978-3-030-86486-6
Page URI
https://pub.uni-bielefeld.de/record/2969237
Zitieren
Castellani A, Schmitt S, Hammer B. Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise. In: Oliver N, Pérez-Cruz F, Kramer S, Read J, Lozano JA, eds. Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I. Lecture Notes in Computer Science. Vol 12975. Cham: Springer International Publishing; 2021: 469-484.
Castellani, A., Schmitt, S., & Hammer, B. (2021). Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise. In N. Oliver, F. Pérez-Cruz, S. Kramer, J. Read, & J. A. Lozano (Eds.), Lecture Notes in Computer Science: Vol. 12975. Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I (pp. 469-484). Cham: Springer International Publishing. https://doi.org/10.1007/978-3-030-86486-6_29
Castellani, Andrea, Schmitt, Sebastian, and Hammer, Barbara. 2021. “Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise”. In Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I, ed. Nuria Oliver, Fernando Pérez-Cruz, Stefan Kramer, Jesse Read, and Jose A. Lozano, 12975:469-484. Lecture Notes in Computer Science. Cham: Springer International Publishing.
Castellani, A., Schmitt, S., and Hammer, B. (2021). “Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise” in Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I, Oliver, N., Pérez-Cruz, F., Kramer, S., Read, J., and Lozano, J. A. eds. Lecture Notes in Computer Science, vol. 12975, (Cham: Springer International Publishing), 469-484.
Castellani, A., Schmitt, S., & Hammer, B., 2021. Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise. In N. Oliver, et al., eds. Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I. Lecture Notes in Computer Science. no.12975 Cham: Springer International Publishing, pp. 469-484.
A. Castellani, S. Schmitt, and B. Hammer, “Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise”, Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I, N. Oliver, et al., eds., Lecture Notes in Computer Science, vol. 12975, Cham: Springer International Publishing, 2021, pp.469-484.
Castellani, A., Schmitt, S., Hammer, B.: Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise. In: Oliver, N., Pérez-Cruz, F., Kramer, S., Read, J., and Lozano, J.A. (eds.) Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I. Lecture Notes in Computer Science. 12975, p. 469-484. Springer International Publishing, Cham (2021).
Castellani, Andrea, Schmitt, Sebastian, and Hammer, Barbara. “Estimating the Electrical Power Output of Industrial Devices with End-to-End Time-Series Classification in the Presence of Label Noise”. Machine Learning and Knowledge Discovery in Databases. Research Track. European Conference, ECML PKDD 2021, Bilbao, Spain, September 13–17, 2021, Proceedings, Part I. Ed. Nuria Oliver, Fernando Pérez-Cruz, Stefan Kramer, Jesse Read, and Jose A. Lozano. Cham: Springer International Publishing, 2021.Vol. 12975. Lecture Notes in Computer Science. 469-484.
Link(s) zu Volltext(en)
Access Level
![Restricted access for author/reviewer only. Restricted](https://pub.uni-bielefeld.de/images/access_restricted.png)