Please use this identifier to cite or link to this item:
Publication type: Conference paper
Type of review: Peer review (publication)
Title: Uncertainty informed anomaly scores with deep learning : robust fault detection with limited data
Authors: Zgraggen, Jannik
Pizza, Gianmarco
Goren Huber, Lilach
et. al: No
DOI: 10.36001/phme.2022.v7i1.3342
Proceedings: Proceedings of the 7th European Conference of the Prognostics and Health Management Society 2022
Editors of the parent work: Do, Phuc
Michau, Gabriel
Ezhilarasu, Cordelia
Volume(Issue): 7
Issue: 1
Page(s): 530
Pages to: 540
Conference details: 7th European PHM, Turin, Italy, 6-8 July 2022
Issue Date: 4-Jul-2022
Series: PHM Society European Conference
Series volume: 7
Publisher / Ed. Institution: PHM Society
Publisher / Ed. Institution: State College
ISBN: 978-1-936263-36-3
Language: English
Subjects: Deep learning; Anomaly detection; Predictive maintenance; Wind turbines; Uncertainty quantification; Condition based maintenance; Machine learning
Subject (DDC): 006: Special computer methods
Abstract: Quantifying the predictive uncertainty of a model is an important ingredient in data-driven decision making. Uncertainty quantification has been gaining interest especially for deep learning models, which are often hard to justify or explain. Various techniques for deep learning based uncertainty estimates have been developed primarily for image classification and segmentation, but also for regression and forecasting tasks. Uncertainty quantification for anomaly detection tasks is still rather limited for image data and has not yet been demonstrated for machine fault detection in PHM applications. In this paper we suggest an approach to derive an uncertainty-informed anomaly score for regression models trained with normal data only. The score is derived using a deep ensemble of probabilistic neural networks for uncertainty quantification. Using an example of wind-turbine fault detection, we demonstrate the superiority of the uncertainty-informed anomaly score over the conventional score. The advantage is particularly clear in an "out-of-distribution" scenario, in which the model is trained with limited data which does not represent all normal regimes that are observed during model deployment.
Further description: Best Paper Award Lizenzangabe: CC BY 3.0 United States
Fulltext version: Published version
License (according to publishing contract): CC BY 3.0: Attribution 3.0 Unported
Departement: School of Engineering
Organisational Unit: Institute of Data Analysis and Process Design (IDP)
Published as part of the ZHAW project: Machine Learning Based Fault Detection for Wind Turbines
Appears in collections:Publikationen School of Engineering

Files in This Item:
File Description SizeFormat 
2022_Zgraggen-Pizza-Goren-Huber_Uncertainty-informed_ProPHME22.pdf3.56 MBAdobe PDFThumbnail

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.