Civil Engineering

You are tasked with designing the pipe network included in this file. In EPANET,

You are tasked with designing the pipe network included in this file. In EPANET, you need to import this file as a Network. You can view this file in Excel if you want. The file contains a reservoir (modeled as endless supply of water at constant elevation) and 10 nodes, which each have an associated maximum demand (in cfs), elevation, and coordinates (dimensions in ft). This maximum demand then gets multiplied by a factor during the 24 hours of the day to account for daily use patterns as shown below (and included in this file)
When you click Project→Run Analysis, it will analyze the flow through the network over the 24 hour period. To find the major loss, it will use the Hazen-Williams equation, which is an empirical equation that only works for water at standard temperature. The advantage of this equation is that it is simple and that it uses ?
C (independent of Reynolds number) rather than ?f like the Darcy-Weisbach equation (dependent on Reynolds number, often requiring iteration):
where ?Q is the volumetric flow rate (gal/min), ?C is the roughness coefficient, ?D is the hydraulic diameter (inches), and ?S is the slope of the energy line (head loss per length of pipe). For this analysis, we are using ?=100C=100, which is conservative. Newer and smoother pipes have higher values of ?C, so this is likely a value for a pipe with considerable build-up of scale.
Once you’ve run the analysis, you will want to use the Map tab in the Browser window to analyze the Nodes for Pressure. This will bring up a scale, and if you press the play button in the Browser, you can watch how the pressure changes with time at each node according to color. The goal is to keep the average pressure around 50 psi. To maintain adequate service, the system should not dip below 35 psi at any node. While the system can probably sustain pressures as high as 80 psi or more, water hammer can lead to larger pressure fluctuations. To be safe, avoid any pressures above 65 psi.
The .inp file has 6 inch pipes by default, and you will find after the analysis that some nodes will have too low pressure when flow rates are high. This means that there is too much head loss leading up to them, and you can simply size up the pipe to lower the head loss. Pipe sizes are in inches—remember to use standard pipe sizes (e.g., 12, 14, 16, 18, 20, 24 inch). You will also find that some nodes have high pressure. This will occur when flow rates are low. The fix for this may not be as easy. You may be able to mitigate it by using smaller pipes upstream of the junction, but at the lowest flow rates, head loss is negligible, and the high pressures are a function of elevation difference (generally when there is a difference of about 80 ft or more). What is typically done is to separate these points from the rest of the system by connecting them to a water tank that is connected to the main system, forming what is called a “pressure zone”. When creating pressure zones, you will need to add a tank with an elevation approximately at the ground level of the point where you place it. It will need to have a lower maximum elevation than the reservoir to provide a lower pressure when flows are low. You will need to play with its min, max, and starting elevations. Here is an introductory video on adding tanks.
When you have finished modifying the network, please write a one-to-two-page memo (see this linkand this link) describing the challenges you faced and justifying the engineering decisions you made. Please attach tables showing the diameters of your pipes, the dimensions of your storage tank, and the maximum and minimum pressures at the times of concern, as well as your project (.net) file and an image of your new pipe layout.
You are allowed to work in groups up to two on this project. If you do work with someone, please be sure to include both of your names on the project. If working as a group, you must submit an alternative design (e.g., that includes a pump for the low pressure issue) and explain which you believe to be the better design.
For a more detailed work-through of pipe network design in EPANET, I encourage you to consult this great resource put together by Robert Pitt at the University of Alabama.