Civil Engineering Reference
In-Depth Information
friction loss, and a change in the velocity or direction of the flow creates additional
losses. Such losses are classified as minor, as they are not significant compared to the
pipeline friction losses. A pipeline with a length in excess of 500 diameters is usually
classified as long.
13
Minor losses become important when small-diameter pipes and
high velocities are considered in the total piping and appurtenances relating to a pump-
ing station. Minor losses are produced by flow-through bends, fittings, valves, and
openings such as fire hydrants, and in and out of reservoirs.
In large-scale network simulations, the minor losses may become insignificant and
can be ignored. Deciding when to include minor losses is a matter of judgment; when
minor losses are included, they can easily be introduced as equivalent lengths of pipe,
although some programs provide for inclusion of these losses separately. Where C
values (coefficient of pipe roughness) are determined from field measurements, they
invariably include a component due to the various minor losses encountered.
Pipe Friction Factor
One of the input requirements for most water system simulation models is the pipe
friction factor appropriate to the pipe flow equations upon which the model is based.
The interior of a pipeline changes with time, and these changes affect the pipe
friction factor. Typically, over time, the friction head losses in a section of pipe will
increase with the pipe's age, because of various physical and chemical characteristics
of the water that change the finish or roughness of the inside of the pipe. The inside
diameter of the pipe also may be reduced. Actions that affect the line capacity include
sedimentation, scaling, organic growth, tuberculation, and corrosion. The effect of
these actions may be partially reversed by pipeline cleaning.
When performing water system simulation, it is important to have relatively accu-
rate information on the friction loss characteristics of the distribution piping. The
hydraulic capacity of pipelines may be determined by conducting flow tests on rep-
resentative sections of the pipelines. Procedures for such tests are based on a measured
flow through a straight section of pipeline (as long as possible) with the pressure drop
recorded by gauges installed at both ends of the section.
Model Input and Output Data
The water distribution system is defined to the model as a set of ''nodes'' and ''lines.''
Nodes are connected in pairs by network elements (lines) such as pumps, pipes, one-
way valves, pressure-reducing valves, and so forth. The nodes are specified as having
either a fixed head or a fixed flow. An outflow is considered to be a demand on the
system; an inflow is considered to be a supply (well, treatment plant, or reservoir) to
the system. Nodes may have flow values of 0—thus no inflow or outflow. These nodes
are used solely for the purpose of connecting different distribution system facilities.
This schematic representation is then translated onto a computer coding form for input
into the computer.
Computer programs simulate the conditions within a water distribution system
based on data provided by the user. For best results, it is important that the input data
be accurate, and that all assumptions regarding input data be carefully made. It is also
necessary that the model being used describe the behavior of the existing system. This
can be checked by ''calibrating'' the model, which requires running the model under
a set of known conditions. Typically, a number of recording pressure gauges will be
in operation on an existing water system while demands are carefully monitored. It is
