• Laser & Optoelectronics Progress
  • Vol. 59, Issue 7, 0714011 (2022)
Weihao Mu, Xuehui Chen*, Yu Zhang, Lei Huang, Darong Zhu, and Bichun Dong
Author Affiliations
  • School of Mechanical and Electrical Engineering, Anhui Jianzhu University, Hefei , Anhui 230601, China
  • show less
    DOI: 10.3788/LOP202259.0714011 Cite this Article Set citation alerts
    Weihao Mu, Xuehui Chen, Yu Zhang, Lei Huang, Darong Zhu, Bichun Dong. Surface Morphology Analysis and Roughness Prediction of 316L Stainless Steel by Selective Laser Melting[J]. Laser & Optoelectronics Progress, 2022, 59(7): 0714011 Copy Citation Text show less
    Surface topography of 316L stainless steel powders
    Fig. 1. Surface topography of 316L stainless steel powders
    Forming strategy of the SLM
    Fig. 2. Forming strategy of the SLM
    OM image of samples surface fabricated by SLM. (a) X-Y direction; (b) Y-Z direction
    Fig. 3. OM image of samples surface fabricated by SLM. (a) X-Y direction; (b) Y-Z direction
    SEM image of 316L stainless steel microstructure fabricated by SLM
    Fig. 4. SEM image of 316L stainless steel microstructure fabricated by SLM
    SEM images of the surface of SLM-formed samples under different LED. (a) 270 J/m; (b) 240 J/m; (c) 210 J/m; (d) 180 J/m; (e) 150 J/m
    Fig. 5. SEM images of the surface of SLM-formed samples under different LED. (a) 270 J/m; (b) 240 J/m; (c) 210 J/m; (d) 180 J/m; (e) 150 J/m
    SEM images of the surface of SLM-formed samples under different powers. (a) 140 W; (b) 160 W;
    Fig. 6. SEM images of the surface of SLM-formed samples under different powers. (a) 140 W; (b) 160 W;
    Relationship between LED and surface roughness of samples under different laser power
    Fig. 7. Relationship between LED and surface roughness of samples under different laser power
    Structure of the BP neural network
    Fig. 8. Structure of the BP neural network
    Flow chart of the GA-BP neural network
    Fig. 9. Flow chart of the GA-BP neural network
    Prediction error of GA-BP and BP neural networks
    Fig. 10. Prediction error of GA-BP and BP neural networks
    ComponentFeCrNiMoMnSiPCSON
    Mass fractionbal17.9411.922.460.0510.56<0.010.00940.020.0150.0086
    Table 1. Chemical components of 316L stainless steel powder
    ParameterValue
    LED /(J·m-1150/180/210/240/270
    Laser power P /W140/160/180/200/220
    Scanning speed v /(mm·s-1500‒1500
    thickness of powder layer d /mm0.3
    Spot diameters D /µm35
    Scanning strategyadjacent layers rotate 67°
    Table 2. Process parameters of the SLM
    LED /(J·m-1Laser power /W
    140160180200220
    15017.65717.60217.96515.71024.198
    18014.08316.05816.25319.38915.877
    21013.34714.84014.62819.00017.764
    24016.48716.89718.30616.92320.066
    27016.39424.68621.49724.20029.270
    Table 3. Surface roughness of samples with different laser power and LED
    P /Wv/(m·s-1Experimental value /μmBP neural networkGA-BP neural network
    Predicted value /μmMAPE /%Predicted value /μmMAPE /%
    1400.5216.39425.581156.017.83968.8
    1600.6716.89725.562251.317.28422.3
    1800.8614.62825.394073.616.378412.0
    2001.1119.38920.68816.717.76618.4
    2201.4724.1987.558668.824.26100.3
    Table 4. Experimental and predicted results of surface roughness
    Weihao Mu, Xuehui Chen, Yu Zhang, Lei Huang, Darong Zhu, Bichun Dong. Surface Morphology Analysis and Roughness Prediction of 316L Stainless Steel by Selective Laser Melting[J]. Laser & Optoelectronics Progress, 2022, 59(7): 0714011
    Download Citation