Page 280 - Fair, Geyer, and Okun's Water and wastewater engineering : water supply and wastewater removal

JWCL344_ch07_230-264.qxd 8/2/10 8:44 PM Page 240

240 Chapter 7 Water Distribution Systems: Modeling and Computer Applications
roughness value is adjusted to make the model results match the measured or expected
values in the real system. However, many other parameters could influence the modeled re-
sults. For example, the water demand at junctions and the status of pipes and valves in the
system could also be adjusted when calibrating a model.
Calibration of a model relies on accurate field measurement data. Field measurements
of pressures in the system, pipe flow rates, water levels in tanks, valve status, and pump op-
erating status and speed are all used to calibrate models. Critical to all of these measure-
ments is the time for which the measurements are made. The times of these measurements
must all be synchronized to the time frame of the model. In addition, because the condi-
tions within a real system change throughout the day or year, field data should be collected
for many different conditions and times. The calibration process is used to adjust the
model to simulate multiple demand loadings and operational boundary conditions. Only
then can the modeler be confident that the model is valid for many different conditions.
WaterGEMS has a module called Darwin Calibrator that it uses to assist in optimizing the
model to match field measurement data. Darwin Calibrator allows the modeler to input field
data, then request the software to determine the optimal solutions for pipe roughness values,
junction demands or status (on/off). Pipes that have the same hydraulic characteristics where
one roughness value is assigned to all pipes can be grouped together. Junctions can also be
grouped based on the demand pattern and location. Caution should be used when grouping
pipes and junctions because this could greatly affect the model’s calibration accuracy.

7.7.2 System Design
The goal of water distribution system design is to maximize the benefits of the system while min-
imizing the cost. The optimal solution is a design that meets all the needs of the system at mini-
mal cost. Some planning is needed to account for additional future needs of the system including
potential growth of the system in terms of demand and its location. The modeler must work with
the system owner and planning groups to account for both the current and future needs.
Another module in WaterGEMS, called Darwin Designer, assists engineers with the
planning and design of water distribution networks. Darwin Designer can be used to size
new pipe and/or rehabilitate old pipes to minimize cost, maximize benefit, or create a sce-
nario for trading off costs and benefits. The least cost optimization is used to determine the
pipe material and size needed to satisfy the design requirements. The maximum benefit op-
timization is used to determine the most beneficial solution based on a known budget.
Darwin Designer will generate a number of solutions that meet the design requirements at
minimal cost or maximum benefit. In either case, the best solution for new pipe or rehabil-
itation of old pipe will be based on the following input hydraulic criteria:
• Minimum and maximum allowable pressures
• Minimum and maximum allowable pipe flow velocity
• Additional demand requirements
• Pipe, pump, tank, valve, etc., status change requirements.
Critical to creating an accurately designed system is time and peak demand requirements.
The peak demand and fire flow conditions are used to size pipes since the pipe network must
work for all conditions. Using average demand values to size pipe without accurately account-
ing for peaking factors can create networks that are either undersized and will not deliver the
required water needs, or oversized and much more expensive than need be. The daily and sea-
sonal variations can also greatly affect the final design. Demand variations need to be synchro-
nized in the model to accurately reflect what could happen in the real system.