Skip to main content

Shrinkage during solidification of complex structure castings based on convolutional neural network deformation prediction research

Abstract

In the investment casting process of castings with complex structures, due to shrinkage deformation, the casting accuracy and pass rate are low, which largely limits the application scope. In this study, a convolutional neural network (CNN) was used to predict and analyze the shrinkage deformation of complex castings in a investment casting process, which provides a new method to improve the precision of investment casting and cavity design. Through the establishment of a hollow thin-walled structure model, simulation calculations, mathematical modeling, and nonlinear fitting analyses of the shrinkage rate were performed, and a mapping model between the shrinkage rate and the structural parameters was constructed. The CNN model was constructed based on the structural characteristics of the model. Finally, experimental verification was carried out to analyze actual hollow turbine blades. The average error for the training set was 0.04%, and the average error for the test set was 0.03%. The experimental results showed that the CNN model proposed in this paper could accurately predict the shrinkage deformation and trends of complex castings in the investment casting process, which has important practical significance for improving the accuracy of castings and mold cavity design.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12

References

  1. 1.

    Sabau AS, Viswanathan S (2003) Material properties for predicting wax pattern dimensionsin investment casting. Mater Sci Eng A 362:125–134

    Article  Google Scholar 

  2. 2.

    Wang D, He B, Liu S, Liu C, Fei L (2016) Dimensional shrinkage prediction based ondisplacement field in investment casting. Int J Adv Manufact Technol 85:201–208

    Article  Google Scholar 

  3. 3.

    Tian GL, Bu K, Zhao D-Q, Zhang Y-L, Qiu F, Zhang X-D, Ren S-J (2018) A shrinkageprediction method of investment casting based on geometric parameters. Int J Adv Manufact Technol 96:1035–1044

    Article  Google Scholar 

  4. 4.

    Pao W, Ransing RS, Lewis RW, Lin C (2004) A medial-axes-based interpolation methodfor solidification simulation. Finite Elemen Anal Design 40:577–593

    Article  Google Scholar 

  5. 5.

    Silva JN, Moutinho DJ, Moreira AL, Ferreira IL, Rocha OL (2011) Determination of heat transfer coefficients at metalˆamold interface during horizontal unsteady-statedirectional solidification of snˆapb alloys. Mater Chem Phys 130:179–185

    Article  Google Scholar 

  6. 6.

    Sabau AS (2006) Alloy shrinkage factors for the investment casting process. Metallurgicaland Mater Trans B 37:131–140

    Article  Google Scholar 

  7. 7.

    Liu C, Jin S, Lai X, He B, Li F (2015) Influence of complex structure on the shrinkage of partin investment casting process. Int J Adv Manufact Technol 77:1191–1203

    Article  Google Scholar 

  8. 8.

    Serin G, Sener B, Ozbayoglu A, Unver H (2020) Review of tool condition monitoring inmachining and opportunities for deep learning. Int J Adv Manufact Technol 1–22

  9. 9.

    Song GW, Tama BA, Park J, Hwang JY, Bang J, Park SJ, Lee S (2019) Temperaturecontrol optimization in a steel-making continuous casting process using a multimodaldeep learning approach. Steel Res Int 90:1900321

    Article  Google Scholar 

  10. 10.

    Ravi B, Srinivasan M (1990) Decision criteria for computer-aided parting surface design. Comput Aided Des 22:11–18

    Article  Google Scholar 

  11. 11.

    Wang Y, Ning D, Feng S (2020) A novel capsule network based on wide convolution and multi-scale convolution for fault diagnosis. Appl Sci 10:3659

    Article  Google Scholar 

  12. 12.

    Stoffel M, Bamer F, Markert B (2018) Artificial neural networks and intelligent finite elements in non-linear structural mechanics. Thin-Walled Struct 131:102–106

    Article  Google Scholar 

  13. 13.

    Kuo C. -C., Qiu W. -K., Liu H. -A., Chang C. -M. (2019) Filling mechanism for prototype partsproduced by vacuum differential pressure casting technology. Int J Adv Manufact Technol 105:1469–1481

    Article  Google Scholar 

  14. 14.

    Yu J, Wang D, Li D, Tang D, Hao X, Tan S, Shu D, Peng Y, Sun B (2020) Enginee-ring computing and data-driven for gating system design in investment casting. Int J Adv Manufact Technol 111:829–837

    Article  Google Scholar 

  15. 15.

    Dong Y, Yan W, Wu Z, Zhang S, Liao T, You Y (2020) Modeling of shrinkage charac-teristics during investment casting for typical structures of hollow turbine blades. Int J Adv Manufact Technol 110:1249–1260

    Article  Google Scholar 

  16. 16.

    Galantucci L, Tricarico L (1998) A computer-aided approach for the simulation of thedirectional-solidification process for gas turbine blades. J Mater Process Technol 77:160–165

    Article  Google Scholar 

  17. 17.

    Tian Y, Yang D, Jiang M, He B (2021) Accurate simulation of complex temperature fieldin counter-pressure casting process using a356 aluminum alloy. Int J Metalcasting 15:259–270

    Article  Google Scholar 

  18. 18.

    Eder MA, Bitsche R (2015) A qualitative analytical investigation of geometrically nonlineareffects in wind turbine blade cross sections. Thin-Walled Struct 93:1–9

    Article  Google Scholar 

  19. 19.

    Shieh J-I, Wu H-H (2020) A framework of importance-performance analysis based on themultiple determination coefficient. IAENG Int J Comput Sci 47:463–467

    Google Scholar 

  20. 20.

    Zheng S, Feng R, Huang A (2020) A modified moving least-squares suitable for scattered datafitting with outliers. J Comput Appl Math 370:112655

    MathSciNet  Article  Google Scholar 

  21. 21.

    Liu J, Fu C (2021) Mnist data set recognition research based on tensorflow framework. Int Core J Eng 7:410–414

    Google Scholar 

  22. 22.

    Liu Z, Feng R, Li X, Wang W, Wu X (2021) Gradient-sensitive optimization for convolu-tional neural networks. Comput Intell Neurosci 2021

Download references

Funding

This research was partially sponsored by the National Natural, Science Foundation of China (Grant Number 51705440), the Natural Science Foundation of Fujian Province, China (Grant Number 2019J01044), and the National Science and Technology Major Project of China (Nos. J2019-III-0008, J2019-VII-0013-0153).

Author information

Affiliations

Authors

Corresponding author

Correspondence to Yiwei Dong.

Ethics declarations

Ethics approval

The authors declare compliance with ethical standards.

Conflict of interest

The authors declare no competing interests.

Additional information

Author contribution

Xiang Guo: methodology, software, validation, writing—original draft. Yiwei Dong: conceptualization, writing—review and editing. Qianwen Ye: software, numerical simulation. Weiguo Yan: experimental design.

Availability of data and materials

The authors confirm that the data that support the findings of this study are available from the corresponding author upon reasonable request.

Code availability

The authors confirm that the code supporting the findings of this study are available from the corresponding author upon reasonable request.

Consent to participate and publish

The authors have read and agreed to the published version of the manuscript.

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

Verify currency and authenticity via CrossMark

Cite this article

Dong, Y., Guo, X., Ye, Q. et al. Shrinkage during solidification of complex structure castings based on convolutional neural network deformation prediction research. Int J Adv Manuf Technol (2021). https://doi.org/10.1007/s00170-021-08137-5

Download citation

Keywords

  • Investment casting
  • Shrinkage deformation
  • Convolutional neural network
  • Predictive analysis