Skip to main navigation Skip to main content
  • E-Submission

JKSPE : Journal of the Korean Society for Precision Engineering

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS
REGULAR

CNN을 이용한 방전 표면에 따른 방전 가공 조건 예측

Prediction of Machining Conditions from EDMed Surface Using CNN

Journal of the Korean Society for Precision Engineering 2024;41(11):865-873.
Published online: November 1, 2024

1 숭실대학교 대학원 기계공학과

2 숭실대학교 기계공학부

1 Department of Precision Engineering, Graduate School, Soongsil University

2 School of Mechanical Engineering, Soongsil University

#E-mail: bhkim@ssu.ac.kr, TEL: +82-2-820-0653
• Received: June 28, 2024   • Accepted: July 28, 2024

Copyright © The Korean Society for Precision Engineering

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 72 Views
  • 1 Download
  • 1 Crossref
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • Experimental Study on the Formation of Discharge Crater Morphology in Micro EDM
    Jae Yeon Kim, Ui Seok Lee, Hee Jin Kong, Bo Hyun Kim
    Journal of the Korean Society for Precision Engineering.2026; 43(1): 61.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

Prediction of Machining Conditions from EDMed Surface Using CNN
J. Korean Soc. Precis. Eng.. 2024;41(11):865-873.   Published online November 1, 2024
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
Prediction of Machining Conditions from EDMed Surface Using CNN
J. Korean Soc. Precis. Eng.. 2024;41(11):865-873.   Published online November 1, 2024
Close

Figure

  • 0
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
  • 11
  • 12
Prediction of Machining Conditions from EDMed Surface Using CNN
Image Image Image Image Image Image Image Image Image Image Image Image Image
Fig. 1 (a) Micro EDM system and (b) EDM circuit
Fig. 2 SEM image of a micro tool electrode
Fig. 3 SEM image of single discharge craters
Fig. 4 Single crater width according to capacitance (workpiece: brass, voltage: 100 V)
Fig. 5 Overlapping discharge crater SEM images (a) brass, (b) stainless steel, and (c) WC-Co
Fig. 6 Dataset images (a) SEM image and (b) digital image
Fig. 7 CNN architecture
Fig. 8 Confusion matrix of CNN model (a) digital image and (b) SEM image
Fig. 9 Grad cam and prediction class of (a) correct predictions and (b) inaccurate predictions
Fig. 10 Grad cam visualization of input data according to (a) brass, (b) stainless steel, and (c) WC-Co
Fig. 11 Model structure to enhance prediction accuracy
Fig. 12 Prediction results for images that were not used for model training
Fig. 13 Prediction results for EDM surface machined with 2 nF
Prediction of Machining Conditions from EDMed Surface Using CNN
Material Brass Stainless steel WC-Co
Thermal conductivity [W/m·K] 99 16.8 110
Specific heat [J/kg × K] 920 490 280
Melting temperature [K] 1,228 1,700 3,140
Latent heat of melting [kJ/kg] 168 285 330
Density [kg/m³] 8,700 8,000 15,800
Thermal diffusivity [mm²/s] 12.4 4.28 24.9
Material Capacitance [nF] Ra [μm]
Brass 20 1.5
200 2.7
2,000 4.3
Stainless steel 20 1.1
200 2.1
2,000 3.2
WC-Co 20 1.7
200 2.3
2,000 3.3
Table 1 Properties of workpiece materials
Table 2 Surface roughness of EDMed surface