input chlorine at given times

[Vladyman]

I would like to simulate one case where the chlorine in one strage tank is regulated every day. Every day at a given time, chlorine is regulated at given concentration (e.g. 1 mg/l).

I specify the initial concentration. Then, when I simulate the concentration decays (figure). How can I input additional chlorine in time, so that every certain time (eg 24 h) the chlorine increases to a given value?

[Patrick Moore]

Vladimir,

Re-dosing stations in EPANET have always been a bit quirky in my opinion which makes using them oftentimes confusing to the modeler. I went ahead and ran several tests in InfoWater and EPANET to see if I could get one of the Quality source methods (how EPANET does re-dosing stations) and looked up what I could find in the helpfiles.

EPANET (and thus InfoWater) allows for the following types of Quality Source options for Node elements

From EPANET Help (pages 78-79)

Field Description

Source Type Select either:

- Concentration

- Mass Booster

- Flow Paced Booster

- Setpoint Booster

Source Quality

Baseline or average concentration (or mass flow rate per minute) of source – leave blank to remove the source

Quality Pattern

ID label of time pattern used to make source quality vary with time – leave blank if not applicable

A water quality source can be designated as a concentration or booster source.

· A concentration source fixes the concentration of any external

inflow entering the network, such as flow from a reservoir or from a

negative demand placed at a junction.

· A mass booster source adds a fixed mass flow to that entering the

node from other points in the network.

· A flow paced booster source adds a fixed concentration to that

resulting from the mixing of all inflow to the node from other points

in the network.

· A setpoint booster source fixes the concentration of any flow

leaving the node (as long as the concentration resulting from all inflow to the node is below the setpoint).
The concentration-type source is best used for nodes that represent source water
supplies or treatment works (e.g., reservoirs or nodes assigned a negative demand).
The booster-type source is best used to model direct injection of a tracer or additional
disinfectant into the network or to model a contaminant intrusion.

My Notes:
From what I have tested the Quality source options work ideally on "reservoir" elements and reasonably well on junctions for the most part, but do not appear to work on "Tank" elements. I have tried Concentration type, Mass Booster, and Setpoint Booster types on "tank" elements and both in InfoWater and EPANET these appear to essentially be ignored as I have not seen EPANET essentially add any constituent to the system using these methods.

Unfortunately this means that it is likely not possible to "exactly" do what you want by using a Quality Source control on a Tank to due to this apparent Error in the EPANET code which would be reflected in also in InfoWater which uses EPANET methods.

However I did seem to find a workaround that should be able to do the trick.

I created a small pump loop connected to the tank that had a timer set on the pump to run only for a few hours a day. I then put a Setpoint Booster on the suction node of the pump set to the concentration I wanted to dose the tank. In the model this will add mass to all water leaving that junction so that it exactly matches the concentration specified. In my example I used a concentration of 2 for Chlorine as I wanted to get the tank back up to around 1.

The tank was set as a complete mix tank.

When the tank was serving demands and the pump was off the tank concentration for Chlorine dropped from 1 to about 0.2 in Chlorine. Then the pump loop would kick on and would essentially Re-Dose the tank with chlorine and bring the concentration back up to about 1 which was my goal.

Here is what this looked like:
(click if need larger image)
WQ - Tank Re-dosing workaround.jpg

So by adjusting your Setpoint Control Concentration and the amount of time the pump runs each day you can effectively "re-dose" the tank once a day.

I imagine any red-dosing at a tank would use a similar type dosing pump to do this, such that it kind of makes sense why EPANET would ignore a tank setpoint control.

Try doing this and you should be able to get it set exactly like you want and even have a sense of the flows and concentrations you would need to add in the real world to make this work.

Patrick Moore

[Vladyman]

Thank you very much. Sorry, one more question.
I added the dummy loop with booster and pump. Results show 2 phases (attached image). A first phase in which chlorine tends to decrease. Then, it reaches an equilibrium
What could be the reason for having two phases with different trends (I use the same daily water demand pattern that repeats every 20 h)

[Patrick Moore]

Vladimir,
I would say there appears to be only one equilibrium here and it is not achieved in this tank until somewhere around 140-180 hours. The early part of the curve just represents conditions that have not fully stabilized or reached a "steady state" where the chlorine concentration reaches a nice repeating pattern.

With any WQ work, you generally need to run the model "long enough" for this repeating pattern or "steady state' to be achieved. This is because your model likely has no initial WQ in the system and so takes time for water to travel through it and "stabilize". The amount of time needed for this depends on several factors including the demand in the model, the amount of storage, as well as the mixing method used for the tanks. This is a key step that any Engineer/modeler should do when performing a WQ analysis to make sure the results are reasonable as presented.

The best way to check if your system has reached this is to run the model for an extended period and verify the water in each tank has reached a steady state repeating pattern of WQ values. The tanks are usually the key because so much volume is associated with them that usually when the tanks are stable, the system itself is also stable. Not a guarantee, but generally a good rule of thumb, because if the tanks are not stable the system is guaranteed not to be.

People often then use an "average" chlorine of the last 24-48 hrs or more (once the model is known to be stable to represent the WQ at a given location, but there are no clear "guidelines" typically of how to present WQ results. But since WQ results have repeating patterns an average of a period generally makes more sense of how to present results.

Incidentally once you can identify how "long" of a run you need to reach a steady state in your system, you can then rerun the analysis and adjust the "report Start" time so that the output of the model is only going to be for the last 24-48 hours. The benefit of this is that the WQ values "range" values of Avg, Min, and Max WQ parameters will include only the time period reported. This can be an easy way to quickly calculate the "average" WQ values (or max and min values) to use in presentations of results.

Hope this helps.

Patrick Moore