Artificial Neural Network on Tool Condition Monitoring in Hard Turning of AISI4140 Steel Using Carbide Tool

Table 5

Optimal results for different algorithms.

Learning algorithm	Network structure	Training data			Testing data
		MSE	R^2	MEP	MSE	R^2	MEP
SCG	5–10–1	0.000152959	0.990022	0.607591	0.004894	0.959475	2.930778
LM	5–10–1	0.000133152	0.996602	0.575407	0.004443	0.969437	2.977617
BFGS	5–10–1	0.000153264	0.990021	0.623932	0.005043	0.960051	3.776345
RP	5–10–1	0.000266462	0.989964	0.757306	0.003076	0.960214	3.166478
CGP	5–10–1	0.000417316	0.989886	1.056837	0.004681	0.960468	3.682791

Among all the learning algorithm given in the table 5, LM algorithm provides the least mean square error for both the training data and the testing data.