Publikationen_1920x250_Detail
In: 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), IEEE

Including Sparse Production Knowledge into Variational Autoencoders to Increase Anomaly Detection Reliability

Tom Hammerbacher , Gorden Platz and Markus Lange-Hegermann,
Aug 2021

Digitalization leads to data transparency for production systems that we can benefit from with data-driven analysis methods like neural networks. For example, automated anomaly detection enables saving resources and optimizing the production. We study using rarely occurring information about labeled anomalies into Variational Autoencoder neural network structures to overcome information deficits of supervised and unsupervised approaches. This method outperforms all other models in terms of accuracy, precision, and recall. We evaluate the following methods: Principal Component Analysis, Isolation Forest, Classifying Neural Networks, and Variational Autoencoders on seven time series datasets to find the best performing detection methods. We extend this idea to include more infrequently occurring meta information about production processes. This use of sparse labels, both of anomalies or production data, allows to harness any additional information available for increasing anomaly detection performance.

Literature procurement: 2021 IEEE 17th International Conference on Automation Science and Engineering (CASE), IEEE
@inproceedings{2314,
author= {Hammerbacher, Tom and Platz, Gorden and Lange-Hegermann, Markus},
title= {Including Sparse Production Knowledge into Variational Autoencoders to Increase Anomaly Detection Reliability},
booktitle= {2021 IEEE 17th International Conference on Automation Science and Engineering (CASE)},
year= {2021},
editor= {},
volume= {},
series= {},
pages= {1262-1267},
address= {},
month= {Aug},
organisation= {},
publisher= {IEEE},
note= {},
}