Table 7 Performance metrics of the analytical models from training data (best results are highlighted).

From: Salt rock creep deformation forecasting using deep neural networks and analytical models for subsurface energy storage applications

Sr. no.

Data (MPa)

Analytical models

R2

RMSE

MAE

MAPE (%)

SMAPE (%)

1

5

Power law

0.98

0.33

0.28

1.45

1.46

Burger

0.97

0.52

0.42

3.49

3.55

Kelvin

0.95

0.92

0.91

6.05

5.95

Spring Dashpot

0.93

0.93

0.91

6.70

6.51

Logarithmic

0.92

1.01

0.97

6.91

6.72

2

15

Power law

0.97

0.36

0.34

2.46

2.42

Burger

0.97

0.52

0.42

3.49

3.55

Kelvin

0.94

0.92

0.91

6.05

5.95

Spring Dashpot

0.93

0.93

0.91

6.70

6.51

Logarithmic

0.92

1.01

0.97

6.91

6.72

3

25

Power law

0.98

0.36

0.29

1.49

1.46

Burger

0.96

0.52

0.42

3.49

3.55

Kelvin

0.95

0.92

0.91

6.05

5.95

Spring Dashpot

0.92

0.93

0.91

6.70

6.51

Logarithmic

0.91

1.01

0.97

6.91

6.72

4

35

Power law

0.97

0.42

0.38

2.53

2.48

Burger

0.94

0.52

0.42

3.49

3.55

Kelvin

0.93

0.92

0.91

6.05

5.95

Spring Dashpot

0.93

0.93

0.91

6.70

6.51

Logarithmic

0.90

1.01

0.97

6.91

6.72

Back to article page