Research Article
Optimizing Cutting Conditions and Prediction of Surface Roughness in Face Milling of AZ61 Using Regression Analysis and Artificial Neural Network
Table 3
Comparison between regression model and ANN predictions.
| Machining parameters | Measured Ra | Regression model | ANN | SS | DOC | Fr | Predicted Ra | Residual | Predicted Ra | Residual |
| 500 | 0.5 | 150 | 0.332 | 0.301 | 0.031 | 0.316 | 0.016 | 500 | 1 | 200 | 0.37 | 0.327 | 0.043 | 0.356 | 0.014 | 500 | 2 | 50 | 0.164 | 0.147 | 0.017 | 0.149 | 0.015 | 1000 | 0.5 | 100 | 0.184 | 0.193 | −0.009 | 0.158 | 0.026 | 1000 | 1 | 150 | 0.207 | 0.224 | −0.017 | 0.208 | −0.001 | 1000 | 1.5 | 200 | 0.258 | 0.255 | 0.003 | 0.272 | −0.014 | 1500 | 0.5 | 50 | 0.115 | 0.127 | −0.012 | 0.142 | −0.027 | 1500 | 1 | 100 | 0.181 | 0.164 | 0.017 | 0.162 | 0.019 | 1500 | 1.5 | 150 | 0.217 | 0.200 | 0.017 | 0.199 | 0.018 | 1500 | 2 | 200 | 0.235 | 0.237 | −0.002 | 0.218 | 0.017 | 2000 | 1 | 50 | 0.153 | 0.146 | 0.007 | 0.144 | 0.009 | 2000 | 1.5 | 100 | 0.155 | 0.188 | −0.033 | 0.169 | −0.014 | 2000 | 2 | 150 | 0.255 | 0.230 | 0.025 | 0.266 | −0.011 |
|
|