We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:


Current browse context:


Change to browse by:

References & Citations

DBLP - CS Bibliography


(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: A general anomaly detection framework for fleet-based condition monitoring of machines

Abstract: Machine failures decrease up-time and can lead to extra repair costs or even to human casualties and environmental pollution. Recent condition monitoring techniques use artificial intelligence in an effort to avoid time-consuming manual analysis and handcrafted feature extraction. Many of these only analyze a single machine and require a large historical data set. In practice, this can be difficult and expensive to collect. However, some industrial condition monitoring applications involve a fleet of similar operating machines. In most of these applications, it is safe to assume healthy conditions for the majority of machines. Deviating machine behavior is then an indicator for a machine fault. This work proposes an unsupervised, generic, anomaly detection framework for fleet-based condition monitoring. It uses generic building blocks and offers three key advantages. First, a historical data set is not required due to online fleet-based comparisons. Second, it allows incorporating domain expertise by user-defined comparison measures. Finally, contrary to most black-box artificial intelligence techniques, easy interpretability allows a domain expert to validate the predictions made by the framework. Two use-cases on an electrical machine fleet demonstrate the applicability of the framework to detect a voltage unbalance by means of electrical and vibration signatures.
Comments: Accepted in Mechanical Systems and Signal Processing, SI: Machine Diagnostics by AI
Subjects: Machine Learning (cs.LG); Signal Processing (eess.SP); Machine Learning (stat.ML)
MSC classes: I.5.3, I.5.4, J.2, I.2.1
ACM classes: I.5.3; I.5.4; J.2; I.2.1
DOI: 10.1016/j.ymssp.2019.106585
Cite as: arXiv:1912.12941 [cs.LG]
  (or arXiv:1912.12941v3 [cs.LG] for this version)

Submission history

From: Kilian Hendrickx [view email]
[v1] Mon, 30 Dec 2019 14:35:45 GMT (7469kb,D)
[v2] Mon, 6 Jan 2020 11:10:06 GMT (7469kb,D)
[v3] Tue, 7 Jan 2020 11:06:06 GMT (7475kb,D)

Link back to: arXiv, form interface, contact.