Machine learning in composites manufacturing: A case study of Automated Fiber Placement inspection
Source abstract
Abstract
Abstract The large-scale adoption of composite materials in industry has allowed for a greater freedom in design and function of structures and their respective components. However, the freedom of material choice has resulted in increased complexity in manufacturing. Machine learning (ML) and Artificial Intelligence (AI) are currently being explored for a number of advanced manufacturing applications, and their applicability has begun to extend into the composites manufacturing realm. In this document, a comprehensive overview of machine learning applications in composites manufacturing will be presented with discussions on a novel inspection software developed for the Automated Fiber Placement (AFP) process at the University of South Carolina utilizing an ML vision system. This vision system allows for defect data to be fully integrated into the manufacturing process, allowing for the ML inspection system to influence several chains in the composites product lifecycle management.
Source: openaire · Quality: candidate unverified · Source record ↗
Research context
OpenAlex topics: Industrial Vision Systems and Defect Detection · Infrastructure Maintenance and Monitoring · Mineral Processing and Grinding
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (65)
Show source-provided references
1. Harik R, Saidy C, Williams SJ, Gurdal Z, Grimsley B. Automated fiber placement defect identity cards: cause, anticipation, existence, significance, and progression; 2018.
2. Cemenska J, Rudberg T, Henscheid M. Automated in-process inspection system for AFP machines. SAE Int J Aerosp. doi: 10.4271/2015-01-2608.10.4271/2015-01-2608 ↗
3. Multilayer feedforward networks are universal approximators10.1016/0893-6080(89)90020-8 ↗
4. Arbib M, Ballard D, Bower J, Orban G. Neural networks algorithms, applications; 1994. doi: 10.1016/0893-6080(94)90073-6, arXiv: arXiv:1312.6199v4.10.1016/0893-6080(94)90073-6 ↗
5. Jain AK, Mao J, Mohiuddin KM, Artificial neural networks: a tutorial; 1996. doi: 10.1109/2.485891, arXiv:arXiv:1411.3159v1.10.1109/2.485891 ↗
6. Schmidhuber J. Deep Learning in neural networks: an overview; 2015. doi: 10.1016/j.neunet.2014.09.003, arXiv:1404.7828.10.1016/j.neunet.2014.09.003 ↗
7. Support-vector networks10.1023/A:1022627411411 ↗
8. Defect characterization in infrared non-destructive testing with learning machines10.1016/j.ndteint.2009.05.004 ↗
9. Alpaydin E. Introduction to machine learning. 2nd ed.; 2010. doi: 10.1007/978-1-62703-748-8_7, arXiv:0904.3664.10.1007/978-1-62703-748-8_7 ↗
10. A genetic algorithm approach to the machine-component grouping problem with multiple objectives10.1016/0360-8352(92)90022-C ↗
11. Improved genetic algorithm for optimization design of a three-dimensional braided composite joint10.1016/j.compstruct.2018.08.051 ↗
12. Whitley D. A genetic algorithm tutorial. Stat Comput. doi: 10.1007/BF00175354.10.1007/BF00175354 ↗
13. García-Martínez C, Rodriguez FJ, Lozano M. Genetic algorithms. doi: 10.1007/978-3-319-07124-4_28.10.1007/978-3-319-07124-4_28 ↗
14. Schmitt LM. Theory of genetic algorithms. Theor Comput Sci. doi: 10.1016/s0304-3975(00)00406-0.10.1016/S0304-3975(00)00406-0 ↗
15. Srinivas M, Patnaik LM. Genetic algorithms: a survey, Computer doi: 10.1109/2.294849.10.1109/2.294849 ↗
16. ML-KNN: A lazy learning approach to multi-label learning10.1016/j.patcog.2006.12.019 ↗
17. The self-organizing map10.1016/S0925-2312(98)00030-7 ↗
18. Anomaly detection in composite elements using Lamb waves and soft computing methods10.1016/j.prostr.2017.07.080 ↗
19. Intelligent health monitoring of aerospace composite structures based on dynamic strain measurements10.1016/j.eswa.2012.01.179 ↗
20. Extreme learning machines: theory and application10.1016/j.neucom.2005.12.126 ↗
21. Imagenet classification with deep convolutional neural networks
22. Systematic evaluation of convolution neural network advances on the Imagenet10.1016/j.cviu.2017.05.007 ↗
23. Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition; 2014. p. 1–14, arXiv:1409.1556. URL: http://arxiv.org/abs/1409.1556.
24. LeCun Y, Bottou L, Bengio Y, Haffner P.Gradient-based learning applied to document recognition. Proc IEEE. doi: 10.1109/5.726791.10.1109/5.726791 ↗
25. Zeng G, He Y, Yu Z, Yang X, Yang R, Zhang L. InceptionNet/GoogLeNet – going deeper with convolutions. Cvpr 91(8):2016;2322–30. doi: 10.1002/jctb.4820, arXiv:1409.4842.10.1002/jctb.4820 ↗
26. He K, Zhang X, Ren S, Sun J. Identity mappings in deep residual networks. doi: 10.1007/978-3-319-46493-0_38.10.1007/978-3-319-46493-0_38 ↗
27. Deep residual learning for image steganalysis
28. Large-scale experimental evaluation of GPU strategies for evolutionary machine learning10.1016/j.ins.2015.10.025 ↗
29. GPU-accelerated extreme learning machines for imbalanced data streams with Concept Drift10.1016/j.procs.2016.05.509 ↗
30. Ma Y, Suda N, Cao Y, Vrudhula S, sun Seo J. ALAMO: FPGA acceleration of deep learning algorithms with a modularized RTL compiler. Integration 62(August 2017):2018;14–23. doi: 10.1016/j.vlsi.2017.12.00910.1016/j.vlsi.2017.12.009 ↗
31. FP-BNN: Binarized neural network on FPGA10.1016/j.neucom.2017.09.046 ↗
32. Posewsky T, Ziener D. Throughput optimizations for FPGA-based deep neural network inference. Microprocessors Microsystems 60(June 2017):2018;151–61.10.1016/j.micpro.2018.04.004 ↗
33. De novo composite design based on machine learning algorithm10.1016/j.eml.2017.10.001 ↗
34. Optimum tooling design for resin transfer molding with virtual manufacturing and artificial intelligence10.1016/S1359-835X(00)00147-0 ↗
35. Pressure based approach for Automated Fiber Placement (AFP) with sensor based feedback loop and flexible component in the effector10.1016/j.ifacol.2017.08.511 ↗
36. Delamination analysis in high speed drilling of carbon fiber reinforced plastics (cfrp) using artificial neural network model10.1016/j.matdes.2008.03.014 ↗
37. Delamination prediction in drilling of cfrp composites using artificial neural network
38. Experimental analysis and ann modelling of haz in laser cutting of glass fibre reinforced plastic composites10.1016/j.protcy.2016.03.044 ↗
39. Quality assessment of porous CFRP specimens using X-ray computed tomography data and artificial neural networks10.1016/j.compstruct.2018.02.096 ↗
40. Artificial neural network (ann) prediction of compressive strength of vartm processed polymer composites10.1016/j.commatsci.2004.11.001 ↗
41. CAggiano A, Rimpault X, Teti R, Balazinski M, Chatelain J, Nele L. Machine learning approach based on fractal analysis for optimal tool life exploitation in cfrp composite drilling for aeronautical assembly. CIRP Ann; 2018. p. 1–4. doi: 10.1016/j.cirp.2018.04.035.10.1016/j.cirp.2018.04.035 ↗
42. Integrated product and process design for resin transfer molded parts10.1177/002199839803201301 ↗
43. Production of low cost carbon-fiber through energy optimization of stabilization process10.3390/ma11030385 ↗
44. Towards pultrusion process optimisation using artificial neural networks10.1016/S0924-0136(98)00052-1 ↗
45. Optimisation of manufacturing process parameters using deep neural networks as surrogate models10.1016/j.procir.2018.03.046 ↗
46. A review of optimisation techniques used in the composite recycling area: state-of-the-art and steps towards a research agenda10.1016/j.jclepro.2016.08.038 ↗
47. Applying case-based reasoning to manufacturing systems design
48. Tension control system design of a filament winding structure based on fuzzy neural network
49. Application of a neural network to improve an automated thermoplastic tow-placement process10.1016/S0959-1524(00)00064-0 ↗
50. Sharp M, Ak R, Hedberg T. A survey of the advancing use and development of machine learning in smart manufacturing. J Manuf Syst. doi: 10.1016/j.jmsy.2018.02.004.10.1016/j.jmsy.2018.02.004 ↗
51. Sparse principal component thermography for subsurface defect detection in composite products
52. Progress in automated ply inspection of AFP layups10.1016/j.repl.2015.05.002 ↗
53. Gregory ED, Juarez PD. In-situ thermography of automated fiber placement parts. In: AIP Conference Proceedings; 2018. doi: 10.1063/1.5031551.10.1063/1.5031551 ↗
54. Juarez PD, Gregory ED, In situ thermal inspection of automated fiber placement manufacturing, In: AIP Conference Proceedings; 2019. doi: 10.1063/1.5099847.10.1063/1.5099847 ↗
55. Denkena B, Schmidt C, Völtzer K, Hocke T. Thermographic online monitoring system for automated fiber placement processes; 2016. doi: 10.1016/j.compositesb.2016.04.076.10.1016/j.compositesb.2016.04.076 ↗
56. Multisensorial self-learning systems for quality monitoring of carbon fiber composites in aircraft production10.1016/j.procir.2013.09.019 ↗
57. Machine learning approach for optimization of automated fiber placement processes10.1016/j.procir.2017.03.295 ↗
58. Long J, Shelhamer E, Darrell T, Fully convolutional networks for semantic segmentation. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition 07-12-June, 2015. p. 3431–40. doi: 10.1109/CVPR.2015.7298965, arXiv:1411.4038.10.1109/CVPR.2015.7298965 ↗
59. Maple C. Geometric design and space planning using the marching squares and marching cube algorithms. doi: 10.1109/GMAG.2003.1219671.10.1109/GMAG.2003.1219671 ↗
60. Albazzan MA, Harik R, Tatting BF, Gürdal Z, Blom-Schieber AW, Rassaian M, et al. Optimization of cylinders with holes under bending using nonconventional laminates. In: AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, 2018. p. 210049. doi: 10.2514/6.2018-1377.10.2514/6.2018-1377 ↗
61. Albazzan MA, Harik R, Tatting BF, Gürdal Z. Efficient design optimization of nonconventional laminated composites using lamination parameters: a state of the art. Compos Struct 209:2018;362–74. doi: 10.1016/j.compstruct.2018.10.095.10.1016/j.compstruct.2018.10.095 ↗
62. Wehbe R, Harik R, Gürdal Z. In-plane tow deformations due to steering in automated fiber placement. AIAA Scitech 2019 Forum. 2019. p. 1-13. doi: 10.2514/6.2019-1271.10.2514/6.2019-1271 ↗
63. Wehbe R, Tatting B, Harik R., Gurdal Z, Halbritter A, Wanthal A. Tow-path based modeling of wrinkling during the automated fiber placement process. In: The composites and advanced materials expo 2017 (CAMX 2017), September 11–14, Florida, United States. 2017.
64. Rajan S, Sutton MA, Wehbe R, Tatting B, Gürdal Z, Kidane A, Harik R. Experimental investigation of prepreg slit tape wrinkling during automated fiber placement process using StereoDIC. Compos Part B Eng 160(July 2018):2019;546–57. doi: 10.1016/j.compositesb.2018.12.017.10.1016/j.compositesb.2018.12.017 ↗
65. Halbritter J, Saidy C, Noevere A, Grimsley B. Automation of AFP process planning functions: importance and ranking. In: SAMPE 2019 conference & exhibition, Charlotte, North Carolina, US, 20–23 May 2019. 2019. doi: 10.33599/nasampe/s.19.1592.10.33599/nasampe/s.19.1592 ↗