Table 1 ST-GPINN vs. EPANET error comparisons in Network A

From: ST-GPINN: a spatio-temporal graph physics-informed neural network for enhanced water quality prediction in water distribution systems

Node Index

MAE (mg/L)

RMSE (mg/L)

MAPE %

R2%

1

0.0018

0.0288

0.35

3.63

2

0.0080

0.0100

0.90

99.44

3

0.0082

0.0103

0.98

99.53

4

0.0081

0.0102

1.05

99.63

5

0.0083

0.0105

0.95

99.52

6

0.0074

0.0095

1.05

99.63

7

0.0078

0.0095

1.20

99.64

8

0.0078

0.0097

1.06

99.66

9

0.0082

0.0102

1.17

99.60

Average

0.0073

0.0121

0.97

88.91

Back to article page