Revealing Hydrodynamic Dependence Of Water Quality Based On Real Time Demand Pattern In Distribution Modeling

Distribution network modeling aims to investigate and manage water quality variations in a distribution network.As automatic water meter reading(AMR)and the supervisory control and data acquisition(SCADA)technologies are widely used,processes such as stochastic,pulse water demand and reaction kinetics on solute transport characteristics can be investigated.First of all,an all-pipe all-demand(APAD)model is simulated by using real time water demand measurements and network operation data.The modeling performance on system hydraulic simulation is assessed by using a comparative network simulation,an hourly demand variation curves(HDVC)demand model.The APAD model results show the prevalence of intermittent and pulse water demand particularly in network perimeters and dead-end branches.The results also highlight different node hydraulic properties such as Re,water age,and flow oscillation when water demand in APAD model is replaced by HDVC-based time-continuous generalized demand patterns.In addition,with the circulating water distribution simulators at U.S.EPA Testing and Evaluation Facility,we examine chemical reaction variability in the water quality changes under various hydrodynamic conditions in drinking water distribution.The variations of kinetic constant for overall chlorine decay(kE)and disinfection byproducts(DBP)formation are determined under stagnant to turbulent flows using three devices of different wall demand and two types of natural organic matters(NOM)in water.The chlorine decay in DI pipe is faster than PVC pipe because of its large wall demand.The results from the comparative experiments and modeling analyses show the relative importance of wall demand(kw),DBP-forming chlorine decay(kD),and other bulk demand(kb')for pipe flows of Re=0-52500.It is found that chlorine reactivity of virgin NOM is the overriding factor.Secondly,for tap water NOM of lower reactivity,pipe flow properties(Re or u)can significantly affect kE,the TTHM yield(T),formation potential(Y),and the time to reach the maximum TTHM concentration(tmax)through their influence on kinetic ratio kD/kb'+kw.These observations,corroborating with turbidity variations during experiments,cannot be explained alone by chlorine dispersion to and from the pipe wall.Mass exchanges through deposition and scale detachment,most likely being flow-dependent,may have contributed to the overall chlorine decay and DBP formation rates.After experimental testing and modeling analysis,here we conclude a water quality model for chlorine decay and DBP formation including the hydrodynamic effects.Last but not least,we propose an Eulerian-Lagrangian numerical approach for simulating chlorine residuals in distribution system,when considering into the hydrodynamic effects on stochastic demand,water dispersion and chemical reaction kinetics.The simulated concentration profiles show significant dependence of chlorine variation on demand allocation and water dipersion,while chlorine values on reaction kinetics.Moreover,the new modeling results show better agreement with field-measured concentrations of free chlorine disinfectant than other models.It is a reliable way for chlorine estimation.