Abstract

EPANET and its commercial derivatives are the most widely-used software packages for modeling free chlorine and its by-products in drinking water distribution systems. Yet, they are not sufficiently accurate, general, or efficient for deriving optimal chlorine dosing strategies at different seasonal temperatures. To overcome EPANET's limitations, an integrated set of rigorously validated multispecies process models are proposed for application within the EPANET-MSX environment. An executable (command-line) version of these MSX models is supplied for use either within the MSX environment or embedded in commercial versions of MSX. A new general method was devised to obtain output of any intermediate coefficient or variable involved in the simulation. This overcomes MSX's limited output options. When the debugged models were applied to a real distribution system, the optimal chlorine dose for summer required almost double the chlorine dose needed in winter. A lower initial dose combined with a downstream booster dose required less chlorine in total. Formal optimization techniques are needed to efficiently obtain similar strategies in more complex systems. PRACTITIONER POINTS: EPANET water quality models are not accurate or general enough for deriving optimal chlorine dosing strategies in distribution systems. Integrated EPNET-MSX models of chlorine reactions in bulk water and at pipe walls, and associated by-product formation, overcome EPANET's limitations. To verify model authenticity, a general technique was developed to obtain values of coefficients and variables within an EPANET-MSX simulation. EPANET-MSX command lines implementing these integrated EPANET-MSX models are presented with verified results for optimal initial and booster dosing strategies. Optimal summer dosing in a real system of rough pipes was almost double that required in winter.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call