Anomaly Detection for Injection Molding Using Probabilistic Deep Learning

Ketonen, Vili; Blech, Jan Olaf

doi:10.1109/icps49255.2021.9468190

Cited by 7 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Savic et al [ 26 ] train two competing AEs, trained offline, stripped to run on edge devices, to classify abnormal behavior in of CPSs in cellular networks. Ketonen et al [ 27 ] facilitate a variational AE based on gated recurrent units (GRU) with static and dynamic time-based features to reason over outliers. Additionally, the authors rank the features in the reconstructed feature vector based on the contribution to the reconstruction error.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis

Kaupp

Humm

Nazemi

et al. 2022

Sensors

View full text Add to dashboard Cite

Smart factories are complex; with the increased complexity of employed cyber-physical systems, the complexity evolves further. Cyber-physical systems produce high amounts of data that are hard to capture and challenging to analyze. Real-time recording of all data is not possible due to limited network capabilities. Limited network capabilities are the reason for a chain of faults introduced via active surveillance during fault diagnosis. These introduced faults may slow down production or lead to an outage of the production line. Here, we present a novel approach to automatically select production-relevant shop floor parameters to decrease the number of surveyed variables and, at the same time, maintain quality in fault diagnosis without overloading the network. We were able to achieve higher throughput, mitigate communication losses and prevent the disruption of factory instructions. Our approach uses an autoencoder ensemble via minority voting to differentiate between normal—always on—variables and production variables that may yield a higher entropy. Our approach has been tested in a production-equal smart factory and was cross-validated by a domain expert.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Both Ketonen et al [ 27 ] and Balogh et al [ 28 ] contribute to explainable AI models. The former enables the explainability by hints in the ranked feature vector, and the latter uses a visualizable model to show broken constraints.…”

Section: Related Workmentioning

confidence: 99%