Table 3 Different models vs. EPANET error comparisons in Network B

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

Models

Network B

MAE (mg/L)

RMSE (mg/L)

MAPE %

R2%

ST-GCN

0.0174

0.0210

5.98

81.91

LSTM

0.0381

0.0542

7.05

71.54

BP

0.0758

0.0993

16.21

62.17

This study

0.0080

0.0098

3.11

98.91

Back to article page