2019
Schmidt, Carsten; Hocke, Tristan; Denkena, Berend
Artificial intelligence for non-destructive testing of CFRP prepreg materials Artikel
In: Production Engineering, S. 1-10, 2019.
Abstract | Links | BibTeX | Schlagwörter: Artificial Intelligence, Automated Fiber Placement, Composite Manufacturing, Composite Structures, Defects, Prepreg, Quality Assurance, Thermal Imaging
@article{Schmidt2019,
title = {Artificial intelligence for non-destructive testing of CFRP prepreg materials},
author = {Carsten Schmidt and Tristan Hocke and Berend Denkena},
url = {https://link.springer.com/article/10.1007%2Fs11740-019-00913-3},
doi = {https://doi.org/10.1007/s11740-019-00913-3},
year = {2019},
date = {2019-07-02},
journal = {Production Engineering},
pages = {1-10},
abstract = {This paper presents a concept of the quality assurance for CFRP prepreg materials and focusses on the classification of thermographic images using convolution neural networks (CNNs). The method for non-destructive testing of CFRP prepreg materials combines a laser-triangulation sensor and an infrared camera to monitor both, the geometry and the impregnation of the prepreg material. The aim is to ensure a high material quality excluding any defective material in an early stage of the process chain of the production of CFRP components. As a result, the reliability of Automated-Fiber-Placement processes utilizing this previously tested material increases. Therefore, an artificial intelligence is set up to classify the thermal images of the CFRP material. Two different architectures of CNN are trained and validated with data sets consisting of thermal images of several prepreg materials and different material defects, such as geometric deviations and varying fiber-matrix-ratios caused by an incorrect impregnation. The CNNs are able to differentiate prepreg materials and to detect and classify certain material-independent defects for known and trained prepreg materials.},
keywords = {Artificial Intelligence, Automated Fiber Placement, Composite Manufacturing, Composite Structures, Defects, Prepreg, Quality Assurance, Thermal Imaging},
pubstate = {published},
tppubtype = {article}
}
This paper presents a concept of the quality assurance for CFRP prepreg materials and focusses on the classification of thermographic images using convolution neural networks (CNNs). The method for non-destructive testing of CFRP prepreg materials combines a laser-triangulation sensor and an infrared camera to monitor both, the geometry and the impregnation of the prepreg material. The aim is to ensure a high material quality excluding any defective material in an early stage of the process chain of the production of CFRP components. As a result, the reliability of Automated-Fiber-Placement processes utilizing this previously tested material increases. Therefore, an artificial intelligence is set up to classify the thermal images of the CFRP material. Two different architectures of CNN are trained and validated with data sets consisting of thermal images of several prepreg materials and different material defects, such as geometric deviations and varying fiber-matrix-ratios caused by an incorrect impregnation. The CNNs are able to differentiate prepreg materials and to detect and classify certain material-independent defects for known and trained prepreg materials.
Schmidt, Carsten; Hocke, Tristan; Denkena, Berend
Deep learning-based classification of production defects in automated-fiber-placement processes Artikel
In: Production Engineering, Bd. 13, 3-4, S. 501-509, 2019.
Abstract | Links | BibTeX | Schlagwörter: Automated Fiber Placement, Industry 4.0, Thermal Imaging
@article{Schmidt2019,
title = {Deep learning-based classification of production defects in automated-fiber-placement processes},
author = {Carsten Schmidt and Tristan Hocke and Berend Denkena},
url = {http://link.springer.com/article/10.1007/s11740-019-00893-4},
doi = {https://doi.org/10.1007/s11740-019-00893-4},
year = {2019},
date = {2019-03-15},
journal = {Production Engineering},
volume = {13, 3-4},
pages = {501-509},
abstract = {This paper presents a deep learning-based approach for the detection and classification of production defects that comple-
ments an existing thermographic online monitoring system for Automated-Fiber-Placement (AFP) processes. The detection
and classification procedure is performed in two stages. In the first stage, the system monitors each tow individually and
classifies its process status. Furthermore, it detects and classifies production defects that affect individual tows such as a
tow-twist. In the second stage, the system monitors the total width of the faultless tows. In this stage, production defects
effecting multiple tows, for example gaps or overlaps, are detected and classified. Twelve different deep convolution neural
networks (CNN) with three various architectures are learned supervised relating to different data sets. The performance of
both identification stages is explored separately before the entire system will be set up. Therefore, the thermal images of the
data sets are superimposed by noise to test the performance of the selected CNN.},
keywords = {Automated Fiber Placement, Industry 4.0, Thermal Imaging},
pubstate = {published},
tppubtype = {article}
}
This paper presents a deep learning-based approach for the detection and classification of production defects that comple-
ments an existing thermographic online monitoring system for Automated-Fiber-Placement (AFP) processes. The detection
and classification procedure is performed in two stages. In the first stage, the system monitors each tow individually and
classifies its process status. Furthermore, it detects and classifies production defects that affect individual tows such as a
tow-twist. In the second stage, the system monitors the total width of the faultless tows. In this stage, production defects
effecting multiple tows, for example gaps or overlaps, are detected and classified. Twelve different deep convolution neural
networks (CNN) with three various architectures are learned supervised relating to different data sets. The performance of
both identification stages is explored separately before the entire system will be set up. Therefore, the thermal images of the
data sets are superimposed by noise to test the performance of the selected CNN.
ments an existing thermographic online monitoring system for Automated-Fiber-Placement (AFP) processes. The detection
and classification procedure is performed in two stages. In the first stage, the system monitors each tow individually and
classifies its process status. Furthermore, it detects and classifies production defects that affect individual tows such as a
tow-twist. In the second stage, the system monitors the total width of the faultless tows. In this stage, production defects
effecting multiple tows, for example gaps or overlaps, are detected and classified. Twelve different deep convolution neural
networks (CNN) with three various architectures are learned supervised relating to different data sets. The performance of
both identification stages is explored separately before the entire system will be set up. Therefore, the thermal images of the
data sets are superimposed by noise to test the performance of the selected CNN.
2017
Schmidt, Carsten; Denkena, Berend; Hocke, Tristan; Völtzer, Klaas
1st CIRP Conference on Composite Materials Parts Manufacturing, 2017.
Links | BibTeX | Schlagwörter: Automated Fiber Placement, Process Monitoring, Quality Assurance, Thermal Imaging
@conference{Schmidt2017b,
title = {Influence of AFP process parameters on the temperature distribution used for thermal in-process monitoring},
author = {Carsten Schmidt and Berend Denkena and Tristan Hocke and Klaas Völtzer},
editor = {Procedia CIRP 66},
doi = {10.1016/j.procir.2017.03.220},
year = {2017},
date = {2017-06-07},
booktitle = {1st CIRP Conference on Composite Materials Parts Manufacturing},
pages = {68-73},
keywords = {Automated Fiber Placement, Process Monitoring, Quality Assurance, Thermal Imaging},
pubstate = {published},
tppubtype = {conference}
}
Schmidt, Carsten; Denkena, Berend; Hocke, Tristan; Völtzer, Klaas
Thermal imaging as a solution for reliable monitoring of AFP processes Konferenz
3rd ACM Automated Composites Manufacturing, Montreal, Canada, 2017.
BibTeX | Schlagwörter: Automated Fiber Placement, Industry 4.0, Manufacturing Quality, Process Monitoring, Thermal Imaging
@conference{Schmidt2017,
title = {Thermal imaging as a solution for reliable monitoring of AFP processes},
author = {Carsten Schmidt and Berend Denkena and Tristan Hocke and Klaas Völtzer},
year = {2017},
date = {2017-04-20},
booktitle = {3rd ACM Automated Composites Manufacturing},
address = {Montreal, Canada},
keywords = {Automated Fiber Placement, Industry 4.0, Manufacturing Quality, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {conference}
}
Brüning, Jan; Schmidt, Carsten; Denkena, Berend
Erhöhung der Prozesssicherheit von Automated-Fiber-Placement-Prozessen Artikel
In: Ingenieurspiegel, Bd. 1, S. 29-31, 2017.
Abstract | BibTeX | Schlagwörter: Composite Structures, Process Monitoring, Thermal Imaging
@article{Brüning2017,
title = {Erhöhung der Prozesssicherheit von Automated-Fiber-Placement-Prozessen},
author = {Jan Brüning and Carsten Schmidt and Berend Denkena},
year = {2017},
date = {2017-01-16},
journal = {Ingenieurspiegel},
volume = {1},
pages = {29-31},
abstract = {In der Luftfahrtindustrie ist Automated Fiber Placement (AFP) eine der führenden Fertigungstechnologien für die kosteneffektive Serienproduktion von hochqualitativen Leichtbaustrukturen. Allerdings bieten sowohl die automatisierten Fertigungssysteme als auch die Prozessplanungssysteme noch ungenutzte Verbesserungspotenziale hinsichtlich Zuverlässigkeit und Effizienz.},
keywords = {Composite Structures, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {article}
}
In der Luftfahrtindustrie ist Automated Fiber Placement (AFP) eine der führenden Fertigungstechnologien für die kosteneffektive Serienproduktion von hochqualitativen Leichtbaustrukturen. Allerdings bieten sowohl die automatisierten Fertigungssysteme als auch die Prozessplanungssysteme noch ungenutzte Verbesserungspotenziale hinsichtlich Zuverlässigkeit und Effizienz.
2016
Schmidt, Carsten; Völtzer, Klaas; Hocke, Tristan; Brüning, Jan
Bahnplanung für Automated-Fiber-Placement-Prozesse Artikel
In: Lightweight Design, Bd. 4, 2016.
BibTeX | Schlagwörter: Automated Fiber Placement, Path Planning, Process Monitoring, Thermal Imaging
@article{Schmidt2016b,
title = {Bahnplanung für Automated-Fiber-Placement-Prozesse},
author = {Carsten Schmidt and Klaas Völtzer and Tristan Hocke and Jan Brüning},
year = {2016},
date = {2016-09-15},
journal = {Lightweight Design},
volume = {4},
keywords = {Automated Fiber Placement, Path Planning, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {article}
}
Denkena, Berend; Schmidt, Carsten; Völtzer, Klaas; Hocke, Tristan
POTENTIAL OF THERMAL INLINE MONITORING IN AUTOMATED FIBER PLACEMENT PROCESS FOR AEROSPACE APLICATIONS Konferenz
7th International Symposium on Composites Manufacturing for High Performance Applications - ISCM, Braunschweig, 2016.
BibTeX | Schlagwörter: Automated Fiber Placement, Process Monitoring, Thermal Imaging
@conference{Denkena2016b,
title = { POTENTIAL OF THERMAL INLINE MONITORING IN AUTOMATED FIBER PLACEMENT PROCESS FOR AEROSPACE APLICATIONS},
author = {Berend Denkena and Carsten Schmidt and Klaas Völtzer and Tristan Hocke},
year = {2016},
date = {2016-08-25},
booktitle = {7th International Symposium on Composites Manufacturing for High Performance Applications - ISCM},
address = {Braunschweig},
keywords = {Automated Fiber Placement, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {conference}
}
Schmidt, Carsten; Denkena, Berend; Völtzer, Klaas; Hocke, Tristan
Thermal Image-Based Monitoring for the Automated Fiber Placement Process Konferenz
10th CIRP Conference on Intelligent Computation in Manufacturing Engineering, Neapel, Italien, 2016.
BibTeX | Schlagwörter: Automated Fiber Placement, Process Monitoring, Thermal Imaging
@conference{Schmidt2016b,
title = {Thermal Image-Based Monitoring for the Automated Fiber Placement Process},
author = {Carsten Schmidt and Berend Denkena and Klaas Völtzer and Tristan Hocke},
year = {2016},
date = {2016-07-20},
booktitle = {10th CIRP Conference on Intelligent Computation in Manufacturing Engineering},
address = {Neapel, Italien},
keywords = {Automated Fiber Placement, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {conference}
}
Denkena, Berend; Schmidt, Carsten; Völtzer, Klaas; Hocke, Tristan
Thermographic online monitoring system for Automated Fiber Placement processes Artikel
In: Composite Part B, Bd. Vol. 97, S. 239-243, 2016.
Abstract | Links | BibTeX | Schlagwörter: Automated Fiber Placement, Process Monitoring, Thermal Imaging
@article{Denkena2016,
title = {Thermographic online monitoring system for Automated Fiber Placement processes},
author = {Berend Denkena and Carsten Schmidt and Klaas Völtzer and Tristan Hocke},
doi = {10.1016/j.compositesb.2016.04.076},
year = {2016},
date = {2016-07-17},
journal = {Composite Part B},
volume = {Vol. 97},
pages = {239-243},
abstract = {Automated Fiber Placement processes are commonly used to manufacture lightweight structures e. g. for highly demanding aerospace applications. In general, quality inspection is usually carried out manually and it is considerably time consuming in terms of high lot sizes and huge part dimensions. An online AFP process monitoring based on thermal camera combined with process depending image processing is presented. The visible temperature difference between the laid-up tow and its surface underneath is analyzed and information of lay-up defects such as overlaps, gaps, twisted tows and bridging derived. This monitoring system can reduce the efforts in quality inspection and will help to increase process reliability significantly.},
keywords = {Automated Fiber Placement, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {article}
}
Automated Fiber Placement processes are commonly used to manufacture lightweight structures e. g. for highly demanding aerospace applications. In general, quality inspection is usually carried out manually and it is considerably time consuming in terms of high lot sizes and huge part dimensions. An online AFP process monitoring based on thermal camera combined with process depending image processing is presented. The visible temperature difference between the laid-up tow and its surface underneath is analyzed and information of lay-up defects such as overlaps, gaps, twisted tows and bridging derived. This monitoring system can reduce the efforts in quality inspection and will help to increase process reliability significantly.
Schmidt, Carsten; Völtzer, Klaas; Hocke, Tristan; Windels, Lars
Automated path planning and thermographic monitoring for automated fibre placement Artikel
In: JEC Composites Magazine, Bd. No. 104, 2016.
BibTeX | Schlagwörter: Automated Fiber Placement, Process Monitoring, Thermal Imaging
@article{Schmidt2016b,
title = {Automated path planning and thermographic monitoring for automated fibre placement},
author = {Carsten Schmidt and Klaas Völtzer and Tristan Hocke and Lars Windels},
year = {2016},
date = {2016-04-01},
journal = {JEC Composites Magazine},
volume = {No. 104},
keywords = {Automated Fiber Placement, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {article}
}
2015
Denkena, Berend; Schmidt, Carsten; Völtzer, Klaas
Online Monitoring of Automated Fiber Placement Processes by Thermal Imaging Konferenzbeitrag
In: ACM - Automated Composite Manufacturing, Montreal, 2015.
Abstract | BibTeX | Schlagwörter: Automated Fiber Placement, Process Monitoring, Thermal Imaging
@inproceedings{Denkena2015,
title = {Online Monitoring of Automated Fiber Placement Processes by Thermal Imaging},
author = {Berend Denkena and Carsten Schmidt and Klaas Völtzer},
year = {2015},
date = {2015-04-24},
booktitle = {ACM - Automated Composite Manufacturing},
address = {Montreal},
abstract = {Automated fiber placement processes are commonly used to manufacture lightweight structures e.g. for highly demanding aerospace applications. Quality inspection usually is done manually and in terms of high part numbers and huge part sizes it is a very time consuming process. An online AFP-process monitoring through a thermal camera combined with process depending image processing is presented.
Thermoset prepreg tows need to be kept cool inside the fiber placement head to prevent fouling. A mayor quality aspect of the placed tow is the tack between tow and surface. To ensure a good tack, surface and tows are heated right before the placement and compacted under specific pressure. Right behind the compaction-roller, the surface temperature of the placed tow shows a lower temperature than the surrounding surfaces. This temperature difference can be detected by a thermal camera. It is shown that by evaluating specific regions behind the roller and by applying an edge detection algorithm the tow geometry and position can be extracted and monitored. This information reveals many placing faults like overlaps, gaps, twisted tows and spliced tows. Furthermore, the bond influences heat transfer from the substrate to the tow surface. Therefore, the temperature of the placed tow can be used as a quality indicator of the bond.
The temperature of the heated surface behind the compaction-roller depends on the heat absorption, the heat transfer into the tooling, and the convection. A homogeneous temperature distribution is expected. If there are inhomogeneous sections in the placed laminate, their temperature differs from the surrounding surface temperature. These hot or cold spots indicate anomalies like bridging faults or foreign objects. Knowing the normal temperature picture of the actual process, hot and cold spots can be detected by dynamic thresholds depending on the actual process.
The presented results add to reduce above-mentioned efforts in quality inspection and will help to increase process reliability significantly.},
keywords = {Automated Fiber Placement, Process Monitoring, Thermal Imaging},
pubstate = {published},
tppubtype = {inproceedings}
}
Automated fiber placement processes are commonly used to manufacture lightweight structures e.g. for highly demanding aerospace applications. Quality inspection usually is done manually and in terms of high part numbers and huge part sizes it is a very time consuming process. An online AFP-process monitoring through a thermal camera combined with process depending image processing is presented.
Thermoset prepreg tows need to be kept cool inside the fiber placement head to prevent fouling. A mayor quality aspect of the placed tow is the tack between tow and surface. To ensure a good tack, surface and tows are heated right before the placement and compacted under specific pressure. Right behind the compaction-roller, the surface temperature of the placed tow shows a lower temperature than the surrounding surfaces. This temperature difference can be detected by a thermal camera. It is shown that by evaluating specific regions behind the roller and by applying an edge detection algorithm the tow geometry and position can be extracted and monitored. This information reveals many placing faults like overlaps, gaps, twisted tows and spliced tows. Furthermore, the bond influences heat transfer from the substrate to the tow surface. Therefore, the temperature of the placed tow can be used as a quality indicator of the bond.
The temperature of the heated surface behind the compaction-roller depends on the heat absorption, the heat transfer into the tooling, and the convection. A homogeneous temperature distribution is expected. If there are inhomogeneous sections in the placed laminate, their temperature differs from the surrounding surface temperature. These hot or cold spots indicate anomalies like bridging faults or foreign objects. Knowing the normal temperature picture of the actual process, hot and cold spots can be detected by dynamic thresholds depending on the actual process.
The presented results add to reduce above-mentioned efforts in quality inspection and will help to increase process reliability significantly.
Thermoset prepreg tows need to be kept cool inside the fiber placement head to prevent fouling. A mayor quality aspect of the placed tow is the tack between tow and surface. To ensure a good tack, surface and tows are heated right before the placement and compacted under specific pressure. Right behind the compaction-roller, the surface temperature of the placed tow shows a lower temperature than the surrounding surfaces. This temperature difference can be detected by a thermal camera. It is shown that by evaluating specific regions behind the roller and by applying an edge detection algorithm the tow geometry and position can be extracted and monitored. This information reveals many placing faults like overlaps, gaps, twisted tows and spliced tows. Furthermore, the bond influences heat transfer from the substrate to the tow surface. Therefore, the temperature of the placed tow can be used as a quality indicator of the bond.
The temperature of the heated surface behind the compaction-roller depends on the heat absorption, the heat transfer into the tooling, and the convection. A homogeneous temperature distribution is expected. If there are inhomogeneous sections in the placed laminate, their temperature differs from the surrounding surface temperature. These hot or cold spots indicate anomalies like bridging faults or foreign objects. Knowing the normal temperature picture of the actual process, hot and cold spots can be detected by dynamic thresholds depending on the actual process.
The presented results add to reduce above-mentioned efforts in quality inspection and will help to increase process reliability significantly.