Comprehensive study and scientific process to increase the accuracy in estimating the thermal conductivity of nanofluids containing SWCNTs and CuO nanoparticles using an artificial neural network

Micro and Nano Systems Letters

Table 3 Characteristics of top 14 ANN structures

Case num	Num. of hidden neurons	Transfer function		Regression
Case num	Num. of hidden neurons	Function1	Function2	R	Train R	Val R	Test R
1	[1 1]	Tansig	Tansig	0.980636	0.990243	0.9986	0.970701
2	[1 1]	Tansig	Logsig	0.982659	0.983369	0.925103	0.989236
3	[1 2]	Logsig	Tansig	0.985406	0.981134	0.993419	0.984336
4	[1 3]	Tansig	Logsig	0.986581	0.990757	0.984235	0.98257
5	[2 1]	Tansig	Tansig	0.995615	0.997098	0.992525	0.998934
6	[2 1]	Logsig	Tansig	0.996408	0.997992	0.996166	0.992992
7	[2 1]	Tansig	Logsig	0.997169	0.997745	0.999489	0.996758
8	[2 2]	Tansig	Tansig	0.997928	0.99867	0.998473	0.990627
9	[2 5]	Tansig	Tansig	0.998127	0.999235	0.999203	0.994305
10	[2 7]	Tansig	Tansig	0.998351	0.998545	0.997819	0.998492
11	[2 9]	Tansig	Logsig	0.998588	0.999069	0.997443	0.999077
12	[2 10]	Logsig	Tansig	0.998746	0.999733	0.999319	0.984129
13	[3 6]	Tansig	Tansig	0.999819	0.999979	0.999508	0.995985
14	[5 7]	Tansig	Logsig	0.999903	0.999999	0.999716	0.999643