Fig. 5: Chlorine concentrations with ST-GPINN calculated, compared with EPANET in junction 2 in Network A. | npj Clean Water

Fig. 5: Chlorine concentrations with ST-GPINN calculated, compared with EPANET in junction 2 in Network A.

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

Fig. 5

Compares the ST-GPINN model’s chlorine concentration predictions with EPANET results at junction 2 over 24 h in network A, showing high accuracy and robustness in tracking temporal water quality dynamics using spatial discretization (100 ft) and a 300 s time step.

Back to article page