0
Research Papers: Design for Manufacture and the Life Cycle

Knowledge-Based Design of Artificial Neural Network Topology for Additive Manufacturing Process Modeling: A New Approach and Case Study for Fused Deposition Modeling

[+] Author and Article Information
Hari P. N. Nagarajan

Mechanical Engineering and Industrial Systems (MEI),
Tampere University of Technology,
P.O. Box 589,
Tampere 33101, Finland
e-mail: hari.nagarajan@tut.fi

Hossein Mokhtarian

Mem. ASME
Mechanical Engineering and Industrial Systems (MEI),
Tampere University of Technology,
P.O. Box 589,
Tampere 33101, Finland;
G-SCOP Laboratory,
CNRS,
University Grenoble Alpes,
Grenoble 38000, France
e-mail: Hossein.mokhtarian@tut.fi

Hesam Jafarian

Mechanical Engineering and Industrial Systems (MEI),
Tampere University of Technology,
P.O. Box 589,
Tampere 33101, Finland
e-mail: hesam.jafarian@tut.fi

Saoussen Dimassi

Mechanical Engineering and Industrial Systems (MEI),
Tampere University of Technology,
P.O. Box 589,
Tampere 33101, Finland
e-mail: swndimassi@gmail.com

Shahriar Bakrani-Balani

LGP-ENIT-INPT & Institut Clément Ader,
CNRS UMR 5312,
University of Toulouse,
47th Avenue d´Azereix,
Tarbes Cedex BP1629-65016, France
e-mail: sbakrani@enit.fr

Azarakhsh Hamedi

Mechanical Engineering and Industrial Systems (MEI),
Tampere University of Technology,
P.O. Box 589,
Tampere 33101, Finland
e-mail: Azarakhsh.hamedi@tut.fi

Eric Coatanéa

Mem. ASME
Mechanical Engineering and Industrial Systems (MEI),
Tampere University of Technology,
P.O. Box 589,
Tampere 33101, Finland
e-mail: eric.coatanea@tut.fi

G. Gary Wang

Mem. ASME
School of Mechatronics Systems Engineering,
Simon Fraser University,
250-13450 102 Avenue,
Surrey, BC V3A0A3, Canada
e-mail: gary_wang@sfu.ca

Karl R. Haapala

Mem. ASME
School of Mechanical, Industrial and
Manufacturing Engineering (MIME),
Oregon State University,
204 Rogers Hall,
Corvallis, OR 97331
e-mail: karl.haapala@oregonstate.edu

1Corresponding author.

Contributed by the Design for Manufacturing Committee of ASME for publication in the JOURNAL OF MECHANICAL DESIGN. Manuscript received July 7, 2018; final manuscript received November 5, 2018; published online December 20, 2018. Assoc. Editor: Mian Li.

J. Mech. Des 141(2), 021705 (Dec 20, 2018) (12 pages) Paper No: MD-18-1545; doi: 10.1115/1.4042084 History: Received July 07, 2018; Revised November 05, 2018

Additive manufacturing (AM) continues to rise in popularity due to its various advantages over traditional manufacturing processes. AM interests industry, but achieving repeatable production quality remains problematic for many AM technologies. Thus, modeling different process variables in AM using machine learning can be highly beneficial in creating useful knowledge of the process. Such developed artificial neural network (ANN) models would aid designers and manufacturers to make informed decisions about their products and processes. However, it is challenging to define an appropriate ANN topology that captures the AM system behavior. Toward that goal, an approach combining dimensional analysis conceptual modeling (DACM) and classical ANNs is proposed to create a new type of knowledge-based ANN (KB-ANN). This approach integrates existing literature and expert knowledge of the AM process to define a topology for the KB-ANN model. The proposed KB-ANN is a hybrid learning network that encompasses topological zones derived from knowledge of the process and other zones where missing knowledge is modeled using classical ANNs. The usefulness of the method is demonstrated using a case study to model wall thickness, part height, and total part mass in a fused deposition modeling (FDM) process. The KB-ANN-based model for FDM has the same performance with better generalization capabilities using fewer weights trained, when compared to a classical ANN.

FIGURES IN THIS ARTICLE
<>
Copyright © 2019 by ASME
Your Session has timed out. Please sign back in to continue.

References

Bourell, D. L. , Leu, M. C. , and Rosen, D. W. , 2009, “ Roadmap for Additive Manufacturing: Identifying the Future of Freeform Processing,” University of Texas at Austin Laboratory of Freeform Fabrication Advance Manufacturing Center, Austin, TX.
Gibson, I. , Rosen, D. , and Stucker, B. , 2014, Additive Manufacturing Technologies: 3D Printing, Rapid Prototyping, and Direct Digital Manufacturing, Springer, New York.
Wang, X. , Jiang, M. , Zhou, Z. , Gou, J. , and Hui, D. , 2017, “ 3D Printing of Polymer Matrix Composites: A Review and Prospective,” Compos. Part B Eng., 110, pp. 442–458. [CrossRef]
Rabinovich, J. E. , 2002, “ Rapid Manufacturing System for Metal, Metal Matrix Composite Materials and Ceramics,” U.S. Patent No. 6,459,069. https://patents.google.com/patent/US6459069
Batchelder, J. S. , Curtis, H. W. , Goodman, D. S. , Gracer, F. , Jackson, R. R. , Koppelman, G. M. , and Mackay, J. D. , 1994, “ Model Generation System Having Closed-Loop Extrusion Nozzle Positioning,” U.S. Patent No. 5,303,141. https://patents.google.com/patent/US5303141A/en
Hilmas, G. E. , Lombardi, J. L. , Hoffman, R. A. , and Stuffle, K. , 1996, “ Recent Developments in Extrusion Freeform Fabrication (EFF) Utilizing Non-Aqueous Gel Casting Formulations,” Solid Freeform Fabrication Symposium (SFF), Austin, TX, Aug. 12–14, pp. 443–450. https://repositories.lib.utexas.edu/handle/2152/70267
Bellehumeur, C. , Li, L. , Sun, Q. , and Gu, P. , 2004, “ Modeling of Bond Formation Between Polymer Filaments in the Fused Deposition Modeling Process,” J. Manuf. Process., 6(2), pp. 170–178. [CrossRef]
Atif Yardimci, M. , and Güçeri, S. , 1996, “ Conceptual Framework for the Thermal Process Modelling of Fused Deposition,” Rapid Prototyp. J., 2(2), pp. 26–31. [CrossRef]
Yardimci, M. A. , Guceri, S. I. , Agarwala, M. , and Danforth, S. C. , 1996, “ Part Quality Prediction Tools for Fused Deposition Processing,” International Solid Freeform Fabrication Symposium (SFF), Austin, TX, Aug. 12–14, pp. 539–548. http://sffsymposium.engr.utexas.edu/Manuscripts/1996/1996-62-Yardimci.pdf
Witherell, P. , Narayanan, A. , and Lee, J. , 2011, “ Using Metamodels to Improve Product Models and Facilitate Inferencing,” Fifth IEEE International Conference on Semantic Computing (ICSC), Palo Alto, CA, pp. 506–513.
Papadrakakis, M. , Lagaros, N. D. , and Tsompanakis, Y. , 1998, “ Structural Optimization Using Evolution Strategies and Neural Networks,” Comput. Methods Appl. Mech. Eng., 156(1–4), pp. 309–333. [CrossRef]
Varadarajan, S. , Chen, W. , and Pelka, C. J. , 2000, “ Robust Concept Exploration of Propulsion Systems With Enhanced Model Approximation Capabilities,” Eng. Optim., 32(3), pp. 309–334. [CrossRef]
Atashkari, K. , Nariman-Zadeh, N. , Gölcü, M. , Khalkhali, A. , and Jamali, A. , 2007, “ Modelling and Multi-Objective Optimization of a Variable Valve-Timing Spark-Ignition Engine Using Polynomial Neural Networks and Evolutionary Algorithms,” Energy Convers. Manage., 48(3), pp. 1029–1041. [CrossRef]
Magnier, L. , and Haghighat, F. , 2010, “ Multiobjective Optimization of Building Design Using TRNSYS Simulations, Genetic Algorithm, and Artificial Neural Network,” Build. Environ., 45(3), pp. 739–746. [CrossRef]
Yegnanarayana, B. , 2009, Artificial Neural Networks, PHI Learning, New Delhi, India.
Wang, G. G. , and Shan, S. , 2007, “ Review of Metamodeling Techniques in Support of Engineering Design Optimization,” ASME J. Mech. Des., 129(4), pp. 370–380. [CrossRef]
Schmidhuber, J. , 2015, “ Deep Learning in Neural Networks: An Overview,” Neural Networks., 61, pp. 85–117. [CrossRef]
Bengio, Y. , Goodfellow, I. J. , and Courville, A. , 2015, “ Deep Learning,” Nature, 521, pp. 436–444. https://www.nature.com/articles/nature14539 [PubMed]
Coatanéa, E. , Roca, R. , Mokhtarian, H. , Mokammel, F. , and Ikkala, K. , 2016, “ A Conceptual Modeling and Simulation Framework for System Design,” Comput. Sci. Eng., 18(4), pp. 42–52. [CrossRef]
Montgomery, D. C. , 2017, Design and Analysis of Experiments, Wiley, Hoboken, NJ.
Dowdy, S. , Wearden, S. , and Chilko, D. , 2011, Statistics for Research, Wiley, Hoboken, NJ.
Levy, P. S. , and Lemeshow, S. , 2013, Sampling of Populations: Methods and Applications, Wiley, Hoboken, NJ.
Schillewaert, N. , Langerak, F. , and Duharnel, T. , 1998, “ Non-Probability Sampling for WWW Surveys: A Comparison of Methods,” Mark. Res. Soc. J., 40(4), pp. 1–13. [CrossRef]
Marshall, M. N. , 1996, “ Sampling for Qualitative Research,” Fam. Pract., 13(6), pp. 522–526. [CrossRef] [PubMed]
Robbins, H. , 1985, “ Some Aspects of the Sequential Design of Experiments,” Herbert Robbins Selected Papers, Springer, Berlin, pp. 169–177.
Plackett, R. L. , and Burman, J. P. , 1946, “ The Design of Optimum Multifactorial Experiments,” Biometrika, 33(4), pp. 305–325. [CrossRef]
Roy, R. K. , 2001, Design of Experiments Using the Taguchi Approach: 16 Steps to Product and Process Improvement, Wiley, New York.
Maghsoodloo, S. , Ozdemir, G. , Jordan, V. , and Huang, C.-H. , 2004, “ Strengths and Limitations of Taguchi's Contributions to Quality, Manufacturing, and Process Engineering,” J. Manuf. Syst., 23(2), pp. 73–126. [CrossRef]
Efthymiou, K. , Sipsas, K. , Mourtzis, D. , and Chryssolouris, G. , 2015, “ On Knowledge Reuse for Manufacturing Systems Design and Planning: A Semantic Technology Approach,” CIRP J. Manuf. Sci. Technol., 8, pp. 1–11. [CrossRef]
Witherell, P. , Feng, S. , Simpson, T. W. , Saint John, D. B. , Michaleris, P. , Liu, Z.-K. , Chen, L.-Q. , and Martukanitz, R. , 2014, “ Toward Metamodels for Composable and Reusable Additive Manufacturing Process Models,” ASME J. Manuf. Sci. Eng., 136(6), p. 061025. [CrossRef]
Hirtz, J. , Stone, R. B. , McAdams, D. A. , Szykman, S. , and Wood, K. L. , 2002, “ A Functional Basis for Engineering Design: Reconciling and Evolving Previous Efforts,” Res. Eng. Des., 13(2), pp. 65–82. [CrossRef]
Karnopp, D. C. , Margolis, D. L. , and Rosemberg, R. C. , 2012, System Dynamics: Modeling, Simulation, and Control of Mechatronic Systems, Wiley, Hoboken, NJ.
Shim, T. , 2002, “ Introduction to Physical System Modelling Using Bond Graphs,” University of Michigan-Dearborn, Dearborn, MI.
Coatanéa, E. , 2005, “ Conceptual Modelling of Life Cycle Design: A Modelling and Evaluation Method Based on Analogies and Dimensionless Numbers,” Ph.D. dissertation, Helsinki University of Technology, Espoo, Finland.
Mokhtarian, H. , Coatanéa, E. , Paris, H. , Ritola, T. , Ellman, A. , Vihinen, J. , Koskinen, K. , and Ikkala, K. , 2016, “ A Network Based Modelling Approach Using the Dimensional Analysis Conceptual Modeling (DACM) Framework for Additive Manufacturing Technologies,” ASME Paper No. DETC2016-60473.
Shavlik, J. W. , Mooney, R. J. , and Towell, G. G. , 1991, “ Symbolic and Neural Learning Algorithms: An Experimental Comparison,” Mach. Learn., 6(2), pp. 111–143.
Atlas, L. , Cole, R. , Muthusamy, Y. , Lippman, A. , Connor, J. , Park, D. , El-Sharkawai, M. , and Marks, R. J. , 1990, “ A Performance Comparison of Trained Multilayer Perceptrons and Trained Classification Trees,” Proc. IEEE, 78(10), pp. 1614–1619. [CrossRef]
Ahmad, S. , and Tesauro, G. , 1989, “ Scaling and Generalization in Neural Networks: A Case Study,” Advances in Neural Information Processing Systems (NIPS), Denver, CO, Nov. 27–30, pp. 160–168. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.1003.3492&rep=rep1&type=pdf
Hochreiter, S. , Bengio, Y. , Frasconi, P. , and Schmidhuber, J. , 2001, “ Gradient Flow in Recurrent Nets: The Difficulty of Learning Long-Term Dependencies,” A field Guide to Dynamical Recurrent Neural Networks, IEEE Press, New York.
Panchal, G. , Ganatra, A. , Kosta, Y. P. , and Panchal, D. , 2011, “ Review on Methods of Selecting Number of Hidden Nodes in Artificial Neural Network,” Int. J. Comput. Theory Eng., 3(2), pp. 332–337. [CrossRef]
Dawson, C. , and Wilby, R. , 2001, “ Hydrological Modelling Using Artificial Neural Networks,” Prog. Phys. Geogr., 25(1), pp. 80–108. [CrossRef]
Tu, J. V. , 1996, “ Advantages and Disadvantages of Using Artificial Neural Networks Versus Logistic Regression for Predicting Medical Outcomes,” J. Clin. Epidemiol., 49(11), pp. 1225–1231. [CrossRef] [PubMed]
Ingrassia, S. , and Morlini, I. , 2005, “ Neural Network Modeling for Small Datasets,” Technometrics, 47(3), pp. 297–311. [CrossRef]
Towell, G. G. , and Shavlik, J. W. , 1994, “ Knowledge-Based Artificial Neural Networks,” Artif. Intell., 70(1–2), pp. 119–165. [CrossRef]
Psichogios, D. C. , and Ungar, L. H. , 1992, “ A Hybrid Neural Network‐First Principles Approach to Process Modeling,” AIChE J., 38(10), pp. 1499–1511. [CrossRef]
Mokhtarian, H. , Coatanéa, E. , Paris, H. , Mbow, M. M. , Pourroy, F. , Marin, P. R. , Vihinen, J. , and Ellman, A. , 2018, “ A Conceptual Design and Modeling Framework for Integrated Additive Manufacturing,” ASME J. Mech. Des., 140(8), p. 081101. [CrossRef]
Moré, J. J. , 1978, “ The Levenberg-Marquardt Algorithm: Implementation and Theory,” Numerical Analysis, Springer, Berlin, pp. 105–116.
Anderson, J. D. , and Wendt, J. , 1995, Computational Fluid Dynamics, Springer, Berlin.
Bakrani Balani, S. , Chabert, F. , Valerie, N. , Cantarel, A. , and Christian, G. , 2017, “ Toward Improvement of the Properties of Parts Manufactured by FFF (Fused Filament Fabrication) Through Understanding the Influence of Temperature and Rheological Behaviour on the Coalescence Phenomenon,” AIP Conf. Proc., 1896(1), p. 040008.
Bridgman, P. , 1922, “ Dimensional Analysis,” Philos. Mag., 2(12), pp. 1263–1266. [CrossRef]
Szirtes, T. , 2007, Applied Dimensional Analysis and Modeling, Butterworth-Heinemann, Oxford, UK.
Mokhtarian, H. , Coatanéa, E. , and Paris, H. , 2017, “ Function Modeling Combined With Physics-Based Reasoning for Assessing Design Options and Supporting Innovative Ideation,” AI Edam, 31(4), pp. 476–500.

Figures

Grahic Jump Location
Fig. 1

Visual representation of the DACM framework

Grahic Jump Location
Fig. 2

The causal relationship extraction algorithm

Grahic Jump Location
Fig. 3

Knowledgeable zone in the causal graph with precomputed weights

Grahic Jump Location
Fig. 4

Traditional KB-ANN development algorithm [44]

Grahic Jump Location
Fig. 5

Solid model (left) and dimensions (right) for the sample part [46]

Grahic Jump Location
Fig. 6

Concurrent experimental and modeling process

Grahic Jump Location
Fig. 7

A classical ANN model for FDM

Grahic Jump Location
Fig. 8

Performance curves for classical fully connected ANN to model: (a) part wall thickness, (b) part height, and (c) part mass

Grahic Jump Location
Fig. 9

Performance curve for best-fit scenario (standard function z = sin(x).cos(y))

Grahic Jump Location
Fig. 10

Performance curves for: (a) modular ANN 1 (viscosity) in the KB-ANN, (b) modular ANN 2 (thickness) in the KB-ANN, (c) modular ANN 3 (height) in the KB-ANN, and (d) modular ANN 4 (mass) in the KB-ANN

Grahic Jump Location
Fig. 11

Comparison of validation error for the fully connected classical ANN and the developed KB-ANN

Tables

Errata

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In