Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning
Sensors are a key element of recent Industry 4.0 developments and currently further sophisticated functionality is embedded into them, leading to smart sensors. In a typical “Factory of the Future” (FoF) scenario, several smart sensors and different data acquisition units (DAQs) will be used to moni...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2022-08-01
|
Series: | Measurement: Sensors |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S2665917422000101 |
_version_ | 1811224055200612352 |
---|---|
author | Tanja Dorst Maximilian Gruber Benedikt Seeger Anupam Prasad Vedurmudi Tizian Schneider Sascha Eichstädt Andreas Schütze |
author_facet | Tanja Dorst Maximilian Gruber Benedikt Seeger Anupam Prasad Vedurmudi Tizian Schneider Sascha Eichstädt Andreas Schütze |
author_sort | Tanja Dorst |
collection | DOAJ |
description | Sensors are a key element of recent Industry 4.0 developments and currently further sophisticated functionality is embedded into them, leading to smart sensors. In a typical “Factory of the Future” (FoF) scenario, several smart sensors and different data acquisition units (DAQs) will be used to monitor the same process, e.g. the wear of a critical component, in this paper an electromechanical cylinder (EMC). If the use of machine learning (ML) applications is of interest, data of all sensors and DAQs need to be brought together in a consistent way. To enable quality information of the obtained ML results, decisions should also take the measurement uncertainty into account. This contribution shows an ML pipeline for time series data of calibrated Micro-Electro-Mechanical Systems (MEMS) sensors. Data from a lifetime test of an EMC from multiple DAQs is integrated by alignment, (different schemes of) interpolation and careful handling of data defects to feed an automated ML toolbox. In addition, uncertainty of the raw data is obtained from calibration information and is evaluated in all steps of the data processing pipeline. The results for the lifetime prognosis of the EMC are evaluated in the light of “fitness for purpose”. |
first_indexed | 2024-04-12T08:43:39Z |
format | Article |
id | doaj.art-30e297572b634c87ae70fc0159a62035 |
institution | Directory Open Access Journal |
issn | 2665-9174 |
language | English |
last_indexed | 2024-04-12T08:43:39Z |
publishDate | 2022-08-01 |
publisher | Elsevier |
record_format | Article |
series | Measurement: Sensors |
spelling | doaj.art-30e297572b634c87ae70fc0159a620352022-12-22T03:39:47ZengElsevierMeasurement: Sensors2665-91742022-08-0122100376Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learningTanja Dorst0Maximilian Gruber1Benedikt Seeger2Anupam Prasad Vedurmudi3Tizian Schneider4Sascha Eichstädt5Andreas Schütze6ZeMA – Center for Mechatronics and Automation Technology gGmbH, Saarbrücken, Germany; Lab for Measurement Technology, Department of Mechatronics, Saarland University, Saarbrücken, Germany; Corresponding author. Lab for Measurement Technology, Department of Mechatronics, Saarland University, Saarbrücken, Germany.Physikalisch-Technische Bundesanstalt, Braunschweig, Berlin, GermanyPhysikalisch-Technische Bundesanstalt, Braunschweig, Berlin, GermanyPhysikalisch-Technische Bundesanstalt, Braunschweig, Berlin, GermanyZeMA – Center for Mechatronics and Automation Technology gGmbH, Saarbrücken, Germany; Lab for Measurement Technology, Department of Mechatronics, Saarland University, Saarbrücken, GermanyPhysikalisch-Technische Bundesanstalt, Braunschweig, Berlin, GermanyZeMA – Center for Mechatronics and Automation Technology gGmbH, Saarbrücken, Germany; Lab for Measurement Technology, Department of Mechatronics, Saarland University, Saarbrücken, GermanySensors are a key element of recent Industry 4.0 developments and currently further sophisticated functionality is embedded into them, leading to smart sensors. In a typical “Factory of the Future” (FoF) scenario, several smart sensors and different data acquisition units (DAQs) will be used to monitor the same process, e.g. the wear of a critical component, in this paper an electromechanical cylinder (EMC). If the use of machine learning (ML) applications is of interest, data of all sensors and DAQs need to be brought together in a consistent way. To enable quality information of the obtained ML results, decisions should also take the measurement uncertainty into account. This contribution shows an ML pipeline for time series data of calibrated Micro-Electro-Mechanical Systems (MEMS) sensors. Data from a lifetime test of an EMC from multiple DAQs is integrated by alignment, (different schemes of) interpolation and careful handling of data defects to feed an automated ML toolbox. In addition, uncertainty of the raw data is obtained from calibration information and is evaluated in all steps of the data processing pipeline. The results for the lifetime prognosis of the EMC are evaluated in the light of “fitness for purpose”.http://www.sciencedirect.com/science/article/pii/S2665917422000101Machine learningDynamic measurement uncertaintyInterpolationTime seriesPredictive maintenanceLow cost sensor network |
spellingShingle | Tanja Dorst Maximilian Gruber Benedikt Seeger Anupam Prasad Vedurmudi Tizian Schneider Sascha Eichstädt Andreas Schütze Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning Measurement: Sensors Machine learning Dynamic measurement uncertainty Interpolation Time series Predictive maintenance Low cost sensor network |
title | Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning |
title_full | Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning |
title_fullStr | Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning |
title_full_unstemmed | Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning |
title_short | Uncertainty-aware data pipeline of calibrated MEMS sensors used for machine learning |
title_sort | uncertainty aware data pipeline of calibrated mems sensors used for machine learning |
topic | Machine learning Dynamic measurement uncertainty Interpolation Time series Predictive maintenance Low cost sensor network |
url | http://www.sciencedirect.com/science/article/pii/S2665917422000101 |
work_keys_str_mv | AT tanjadorst uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning AT maximiliangruber uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning AT benediktseeger uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning AT anupamprasadvedurmudi uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning AT tizianschneider uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning AT saschaeichstadt uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning AT andreasschutze uncertaintyawaredatapipelineofcalibratedmemssensorsusedformachinelearning |