Skip to main content
SpringerLink
Log in

Tool path optimization of selective laser sintering processes using deep learning

  • Original Paper
  • Published:
Computational Mechanics Aims and scope Submit manuscript

Abstract

Advancements in additive manufacturing (3D printing) have enabled researchers to create complex structures, offering a new class of materials that can surpass their individual constituent properties. Selective laser sintering (SLS) is one of the most popular additive manufacturing techniques and uses laser power to bond powdered material into intricate structures. It is one of the fastest additive manufacturing processes for printing functional, durable prototypes, or end-user parts. It is also widely used in many industries, due to its ability to easily make complex geometries with little to no additional manufacturing effort. In the SLS process, tool path selection is important because it is directly related to the integrity of a 3D printed structure. In this research, we focus on how to obtain an optimal tool path for the SLS process from a numerical simulation. Also, we apply a deep learning technique to accelerate the simulation of the SLS processes, while obtaining accurate numerical results.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19
Fig. 20
Fig. 21
Fig. 22

Similar content being viewed by others

Availability of data and material

All the relevant data and the material is available at https://github.com/kdhoon2/SLS_DHKIM.

Notes

  1. PA12 is a widely used thermoplastic material for selective laser sintering which is suitable for many applications [1].

  2. Selective laser sintering (SLS) processes heat the powdered material below its melting point so that the powder can fuse together at a particle level. In selective laser melting (SLM), however, the powdered material is not merely fused together but is fully melted. While the SLM normally works with metals, the SLS generally works with plastics and ceramics.

  3. One could use a micro-scale particle-based model to include the expression for the estimated extinction coefficient.

  4. It is hard to say that the given configuration of the hyper-parameters is the optimal choice for the deep learning model. One could try optimizing hyper-parameters by using optimization techniques such as Bayesian optimization, evolutionary algorithms, or gradient-based optimization.

References

  1. Aldahash SA (2018) Optimum manufacturing parameters in selective laser sintering of PA12 with white cement additives. Int J Adv Manuf Technol 96(1–4):257–270

    Article  Google Scholar 

  2. Steen WM, Mazumder J (2010) Laser material processing. Springer, Berlin

    Book  Google Scholar 

  3. Zeng K, Pal D, Stucker B (2012) A review of thermal analysis methods in laser sintering and selective laser melting. In: Proceedings of solid freeform fabrication symposium Austin, TX, vol 60, pp 796–814

  4. Dong L et al (2009) Three-dimensional transient finite element analysis of the selective laser sintering process. J Mater Process Technol 209(2):700–706

    Article  Google Scholar 

  5. Kolossov S et al (2004) 3D FE simulation for temperature evolution in the selective laser sintering process. Int J Mach Tools Manuf 44(2-3):117–123

    Article  Google Scholar 

  6. Matsumoto M et al (2002) Finite element analysis of single layer forming on metallic powder bed in rapid prototyping by selective laser processing. Int J Mach Tools Manuf 42(1):61–67

    Article  Google Scholar 

  7. Simchi A (2006) Direct laser sintering of metal powders: mechanism, kinetics and microstructural features. Mater Sci Eng A 428(1–2):148–158

    Article  Google Scholar 

  8. Simchi A, Pohl H (2003) Effects of laser sintering processing parameters on the microstructure and densification of iron powder. Mater Sci Eng A 359(1–2):119–128

    Article  Google Scholar 

  9. Gusarov AV, Kruth J-P (2005) Modelling of radiation transfer in metallic powders at laser treatment. Int J Heat Mass Transf 48(16):3423–3434

    Article  MATH  Google Scholar 

  10. Gusarov AV et al (2009) Model of radiation and heat transfer in laser-powder interaction zone at selective laser melting. J Heat Transf 131(7):072101

    Article  Google Scholar 

  11. Ganeriwala R, Zohdi TI (2014) Multiphysics modeling and simulation of selective laser sintering manufacturing processes. Procedia CIRP 14:299–304

    Article  Google Scholar 

  12. Zohdi TI (2013) Rapid simulation of laser processing of discrete particulate materials. Arch Comput Methods Eng 20(4):309–325

    Article  Google Scholar 

  13. Zohdi TI (2014) Additive particle deposition and selective laser processing-a computational manufacturing framework. Comput Mech 54(1):171–191

    Article  Google Scholar 

  14. Zohdi TI (2014) A direct particle-based computational framework for electrically enhanced thermo-mechanical sintering of powdered materials. Math Mech Solids 19(1):93–113

    Article  MATH  Google Scholar 

  15. Zohdi TI (2017) Modeling and simulation of laser processing of particulate-functionalized materials. Arch Comput Methods Eng 24(1):89–113

    Article  MathSciNet  MATH  Google Scholar 

  16. Ganeriwala R, Zohdi TI (2016) A coupled discrete element-finite difference model of selective laser sintering. Granular Matter 18(2):21

    Article  Google Scholar 

  17. Zohdi TI (2018) Dynamic thermomechanical modeling and simulation of the design of rapid free-form 3D printing processes with evolutionary machine learning. Comput Methods Appl Mech Eng 331:343–362

    Article  MathSciNet  MATH  Google Scholar 

  18. Weinan E, Han J, Jentzen A (2017) Deep learning-based numerical methods for high-dimensional parabolic partial differential equations and backward stochastic differential equations. Commun Math Stat 5(4):349–380

    Article  MathSciNet  MATH  Google Scholar 

  19. Cai Z et al (2020) Deep least-squares methods: An unsupervised learning- based numerical method for solving elliptic PDEs. J Comput Phys 420:109707

    Article  MathSciNet  Google Scholar 

  20. Weinan E, Yu B (2018) The deep Ritz method: a deep learning-based numerical algorithm for solving variational problems. Commun Math Stat 6(1):1–12

    MathSciNet  MATH  Google Scholar 

  21. Mohan AT, Gaitonde DV (2018) A deep learning based approach to reduced order modeling for turbulent ow control using LSTM neural net- works. arXiv preprint arXiv:1804.09269

  22. Wang Z et al (2018) Model identification of reduced order UID dynamics sys- tems using deep learning. Int J Numer Meth Fluids 86(4):255–268

    Article  Google Scholar 

  23. Bertsekas DP et al (1995) Dynamic programming and optimal control. Athena Scientific, Belmont

    MATH  Google Scholar 

  24. Bellman R (1966) Dynamic programming. Science 153(3731):34–37

    Article  MATH  Google Scholar 

  25. Howard RA (1960) Dynamic programming and Markov processes

  26. Bellman RE, Dreyfus SE (2015) Applied dynamic programming. Princeton University Press, Princeton

    MATH  Google Scholar 

  27. Hecht J (2018) Understanding lasers: an entry-level guide. Wiley, Hoboken

    Book  Google Scholar 

  28. Rombouts M et al (2005) Light extinction in metallic powder beds: correlation with powder structure. J Appl Phys 98(1):013533

    Article  Google Scholar 

  29. Valueva MV et al (2020) Application of the residue number system to reduce hardware costs of the convolutional neural network implementation. In: Mathematics and computers in simulation

  30. LeCun Y et al (1989) Backpropagation applied to handwritten zip code recognition. Neural Comput 1(4):541–551

    Article  Google Scholar 

  31. Y LeCun, Y Bengio et al (1995) Convolutional networks for images, speech, and time series. In: The handbook of brain theory and neural networks, vol 3361, no 10

  32. LeCun Y et al (1998) Gradient-based learning applied to document recognition. Proc IEEE 86(11):2278–2324

    Article  Google Scholar 

  33. Goodfellow I, Bengio Y, Courville A (2016) Deep learning. MIT Press, Cambridge

    MATH  Google Scholar 

  34. Mitchell RS, Michalski JG, Carbonell TM (2013) An artificial intelligence approach. Springer, Berlin

    Google Scholar 

  35. Murphy KP (2012) Machine learning: a probabilistic perspective. MIT Press, Cambridge

    MATH  Google Scholar 

  36. Langley P et al (1994) Selection of relevant features in machine learning. In: Proceedings of the AAAI fall symposium on relevance, vol 184, pp 245–271

  37. Kononenko I, Kukar M (2007) Machine learning and data mining. Horwood Publishing, Cambridge

    Book  MATH  Google Scholar 

  38. Gu GX, Chen CT, Buehler MJ (2018) De novo composite design based on machine learning algorithm. Extreme Mech Lett 18:19–28

    Article  Google Scholar 

  39. Chen C-T, Gu GX (2019) Machine learning for composite materials. MRS Commun 9(2):556–566

    Article  Google Scholar 

  40. Y Kim et al. (2020) Designing adhesive pillar shape with deep learning-based optimization. ACS Appl Mater Interfaces

  41. K He et al. (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp 770–778

  42. Guzella TS, Caminhas WM (2009) A review of machine learning approaches to spam filtering. Expert Syst Appl 36(7):10206–10222

    Article  Google Scholar 

  43. Santos I et al. (2009) Machine-learning-based mechanical properties prediction in foundry production. In: 2009 ICCAS-SICE. IEEE, 4536–4541

  44. Mannodi-Kanakkithodi A et al (2016) Machine learning strategy for accelerated design of polymer dielectrics. Sci Rep 6:20952

    Article  Google Scholar 

  45. Pilania G et al (2020) Data-based methods for materials design and discovery: basic ideas and general methods. Synth Lect Mater Opt 1(1):1–188

    Google Scholar 

  46. Oishi A, Yagawa G (2017) Computational mechanics enhanced by deep learning. Comput Methods Appl Mech Eng 327:327–351

    Article  MathSciNet  MATH  Google Scholar 

  47. Paszke A et al (2019) Pytorch: an imperative style, high-performance deep learning library. In: Advances in neural information processing systems, pp 8026–8037

  48. Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: Proceedings of the thirteenth international conference on artificial intelligence and statistics, pp 249–256

  49. Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv preprint arXiv:1412.6980

  50. Wilt JK, Yang C, Gu GX (2020) Accelerating auxetic metamaterial design with deep learning. Adv Eng Mater 22:1901266

    Article  Google Scholar 

  51. Kim DH, Zohdi TI, Singh RP (2020) Modeling, simulation and machine learning for rapid process control of multiphase owing foods. Comput Methods Appl Mech Eng 371:113286

    Article  MATH  Google Scholar 

Download references

Acknowledgements

This research was supported by Samsung Scholarship. Also, the first author, Dong Hoon Kim, thanks Euihyun Choi for providing valuable thoughts on tool path generation.

Funding

This research was supported by Samsung Scholarship.

Author information

Authors and Affiliations

  1. Department of Mechanical Engineering, 6195 Etcheverry Hall, University of California, Berkeley, CA, 94720-1740, USA

    D. H. Kim & T. I. Zohdi

Authors
  1. D. H. Kim
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. T. I. Zohdi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to D. H. Kim.

Ethics declarations

Conflicts of interest/Competing interests

Not applicable

Code availability

The simulation code is available at https://github.com/kdhoon2/SLS_DHKIM.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Kim, D.H., Zohdi, T.I. Tool path optimization of selective laser sintering processes using deep learning. Comput Mech 69, 383–401 (2022). https://doi.org/10.1007/s00466-021-02079-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00466-021-02079-1

Keywords

Search

Navigation