Table 5 Performances of ANNs with different structures regarding the prediction of viscosity.

From: Modeling and optimization of thermal conductivity and viscosity of MnFe2O4 nanofluid under magnetic field using an ANN

Number of hidden layers

Number of neurons in each layer

R2

MSE

Test data

Training data

Test data

Training data

1

2

0.986

0.990

4.20E-06

2.40E-06

1

4

0.992

0.994

1.70E-06

1.03E-06

1

6

0.992

0.995

1.48E-06

7.04E-07

1

8

0.992

0.995

1.65E-06

6.50E-07

1

10

0.994

0.995

1.61E-06

6.46E-07

1

12

0.992

0.995

1.68E-06

8.28E-07

2

2

0.990

0.993

2.44E-06

1.38E-06

2

4

0.992

0.995

1.71E-06

6.62E-07

2

6

0.992

0.996

1.44E-06

4.66E-07

2

8

0.990

0.996

1.08E-06

4.32E-07

2

10

0.994

0.996

8.56E-07

4.14E-08

2

12

0.993

0.996

1.26E-06

1.31E-06

Back to article page