FLOOD01647

• County: Statewide
• Title: Optimal Sizing of Urban Flood Control Systems
• Date: 3/1/1974
• Prepared For: US
• Prepared By: COE
• Floodplain - Doc Type: Educational/Technical/Reference Information

a unique optimum are less than with an objective function that varies regularly
(termed well-behaved). Results of applications to date suggest that the objective
function is reasonably well-behaved and that unique solutions do in fact come
out of the procedure, However, note that this particular methodology (univariate
gradient procedure) does not guarantee that the true optimum (global optimum)
is achieved. However, the derived system will be very near optimum for the
component sizes in the general order of magnitude of the initial component
sizes, A study methodology that considers that local optimums may occur; e.g.,
testing a few starting values would be appropriate,

"

APPlICATION TO URBAN fLOOD-CoNTROL PROJECT

Ii

The technique was developed for the United States Army Engineer District,
St. Louis, Mo" for use in plan formulation studies for the Harding Ditch unit
of the East St. Louis and Vicinity, Interior Flood Control Project. The District
desired a technique that would enable automatically determining the scales of
flood-control system components comprising three to four reservoirs, a diversion,
and one to two pumping plants. The development work had proceeded well
so that when it became necessary for the District to perform additional analysis
of a unit of the project that had previously been studied, an application of
the technique was undertaken to assist the studies and provide for testing,
The area studied was the Blue Waters Ditch unit of the project that encompasses
approx 9,000 acres of the American Bottoms area, The area consists of a number
. of smaller and a few major communities. A few drainage canals and levee
segments exist and the lower (outlet) end of the area is protected by major
levees of the Mississippi River system necessitating that most flood flows be
pumped from the basin, Fig, 5 is a schematic of the system.
Previous studies had defined two detention storage sites and a pumping facility
as potential system components. The technique was applied to determine the
best size of the pumping facility and detention storage areas for a range of
storage site characteristics, project discount rates, assumed economic conditions,
and performance standards. A major objective of the study was to determine
the sensitivity of the component scales to assumed flood-plain land-use controls.
This was accomplished by optimizing the sizes of the components for: (I) No
target degree of protection and economic flow-damage functions prepared for
damage potential as it existed in 1973; (2) economic flow-damage functions
reflecting uncontrolled future growth; and (3) for a reasonably controlled future
growth compatible with the flood-control system, Optimization of the component
sizes was then repeated for the same sets of data for a target degree of protection
of IOO-yr exceedence interval. The sensitivity of the system to detention site
characteristics was examined by altering the reservoir elevation-storage and
reservoir storage-cost functions and optimizing. The sensitivity to the project
discount rate was examined by optimizing the component sizes for one of the
previously studied conditions for three discount rates.
The results of the studies are preliminary and should be considered as a
test application of the methodology rather than the final results of the formulation
studies for Blue Waters. However, the studies were a real component of the
plan formulation and evaluation strategy and the results presented in Table
1 are not a selected case study. The solutions were sufficiently promising that