Page 169 - Data Science Algorithms in a Week

The Estimation of Cutting Forces in the Turning of Inconel 718 Assisted … 153

Table 2. Input parameters and experimental results

Machining parameters Cutting forces
No. Dn s p vc f Fc Ff Fp
[mm] [mm] [MPa] [m/min] [mm/rev] [N] [N] [N]
Training data set
1 0.25 0 50 46 0.2 1280 615 475
2 0.25 0 90 57 0.224 1295 545 450
3 0.25 1.5 50 46 0.25 1508 645 530
4 0.25 1.5 90 57 0.2 1150 540 425
5 0.25 3 90 57 0.25 1350 660 520
6 0.25 3 130 74 0.2 1150 545 420
7 0.3 0 50 57 0.2 1245 520 400
8 0.3 0 90 74 0.224 1265 505 410
9 0.3 1.5 50 57 0.25 1460 560 485
10 0.3 1.5 130 46 0.224 1145 565 470
11 0.3 3 90 74 0.25 1385 505 405
12 0.3 3 130 46 0.2 1055 565 435
13 0.4 0 50 74 0.2 1187 505 410
14 0.4 0 130 57 0.25 1305 520 440
15 0.4 1.5 90 46 0.2 1160 560 435
16 0.4 1.5 130 57 0.224 1275 530 465
17 0.4 3 90 46 0.25 1375 560 470
18 0.4 3 130 57 0.2 1250 545 430
Testing data set
1 0.25 0 130 74 0.25 1370 570 470
2 0.25 1.5 130 74 0.224 1235 520 440
3 0.25 3 50 46 0.224 1400 630 510
4 0.3 0 130 46 0.25 1390 565 485
5 0.3 1.5 90 74 0.2 1190 475 415
6 0.3 3 50 57 0.224 1320 555 465
7 0.4 0 90 46 0.224 1450 620 475
8 0.4 1.5 50 74 0.25 1465 565 478
9 0.4 3 50 74 0.224 1320 590 460

processing elements (neurons) organized in several layers. These neurons are connected
to each other by weighted links denoted by synapses which establish the relationship
between input data and output data. There are many ANN models and multilayer
perceptions, which only feed forward and multilayered networks, were considered in this
paper. The structure of these ANN has three types of layers: input layer, hidden layer and
out
inp
output layer. The biases in the neurons of the hidden and output layers, Oi and Oj ,
respectively, are controlled during data processing. The biases in the neurons of the