Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis
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 netwo...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-10-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/21/8259 |
| _version_ | 1797466507972706304 |
|---|---|
| author | Lukas Kaupp Bernhard Humm Kawa Nazemi Stephan Simons |
| author_facet | Lukas Kaupp Bernhard Humm Kawa Nazemi Stephan Simons |
| author_sort | Lukas Kaupp |
| collection | DOAJ |
| description | 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. |
| first_indexed | 2024-03-09T18:39:50Z |
| format | Article |
| id | doaj.art-1f274b7e2c1045f8ae6a224a84807042 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T18:39:50Z |
| publishDate | 2022-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-1f274b7e2c1045f8ae6a224a848070422023-11-24T06:45:14ZengMDPI AGSensors1424-82202022-10-012221825910.3390/s22218259Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault DiagnosisLukas Kaupp0Bernhard Humm1Kawa Nazemi2Stephan Simons3Faculty of Computer Science, Darmstadt University of Applied Sciences, Haardtring 100, 64295 Darmstadt, GermanyFaculty of Computer Science, Darmstadt University of Applied Sciences, Haardtring 100, 64295 Darmstadt, GermanyResearch Group Human-Computer Interaction and Visual Analytics, Faculty of Media, Darmstadt University of Applied Sciences, Haardtring 100, 64295 Darmstadt, GermanyFaculty of Electrical Engineering and Information Technology, Darmstadt University of Applied Sciences, Haardtring 100, 64295 Darmstadt, GermanySmart 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.https://www.mdpi.com/1424-8220/22/21/8259context-aware diagnosisoutlier detectioncyber-physical systemspre-training variable selectionautoencoder ensembleproduction-relevant variables |
| spellingShingle | Lukas Kaupp Bernhard Humm Kawa Nazemi Stephan Simons Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis Sensors context-aware diagnosis outlier detection cyber-physical systems pre-training variable selection autoencoder ensemble production-relevant variables |
| title | Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis |
| title_full | Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis |
| title_fullStr | Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis |
| title_full_unstemmed | Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis |
| title_short | Autoencoder-Ensemble-Based Unsupervised Selection of Production-Relevant Variables for Context-Aware Fault Diagnosis |
| title_sort | autoencoder ensemble based unsupervised selection of production relevant variables for context aware fault diagnosis |
| topic | context-aware diagnosis outlier detection cyber-physical systems pre-training variable selection autoencoder ensemble production-relevant variables |
| url | https://www.mdpi.com/1424-8220/22/21/8259 |
| work_keys_str_mv | AT lukaskaupp autoencoderensemblebasedunsupervisedselectionofproductionrelevantvariablesforcontextawarefaultdiagnosis AT bernhardhumm autoencoderensemblebasedunsupervisedselectionofproductionrelevantvariablesforcontextawarefaultdiagnosis AT kawanazemi autoencoderensemblebasedunsupervisedselectionofproductionrelevantvariablesforcontextawarefaultdiagnosis AT stephansimons autoencoderensemblebasedunsupervisedselectionofproductionrelevantvariablesforcontextawarefaultdiagnosis |