Environmental Engineering Reference
In-Depth Information
often applied. In this case, if depth of water observation parameters is more
accurate than discharge values, then the depth of water as observation values are
usually included in the object function and discharges will be considered as the
penalty function. If the calculated discharge is a certain value of the observed
discharge the penalty function is zero, but if the difference between observed and
calculated discharge is too large, a penalty will be considered to the object
function''. The related equation of object function (OF) with penalty function (PF)
is given by [ 54 ] as follows:
OF ¼ X
Stations
X
j
h calculated h observed
j þ PF
ð 3 : 18 Þ
time
(
\0 : 05
Q calculated Q QH
Q QH
0
if
PF ¼
ð 3 : 19 Þ
10 ; 000
else
where:
h is depth of water, Q is discharge, and Q QH is the discharge obtained from a
discharge-stage.
In this research, calibration analysis was limited to the observed depth of water
found in the literature and field reports on the 2008 flood. The adjustment was
performed using different friction values in the channel. Manning friction coeffi-
cients were assigned based on different land use types to generate a friction surface
[ 11 ]. The calibration of the model was based on the surface roughness parameters
mainly within the channel. But for floodplain and overland flow modules the value
was generated to produce real-world situations based on Chow [ 49 ].
3.9 Hydrodynamic Modeling
3.9.1 Basic Theory of the Models SOBEK and HEC-RAS
In this study, the SOBEK (1D-2D) and HEC-RAS (1D) modeling approaches were
selected as hydrodynamic modeling because these two models were freely avail-
able for this research. In the HEC-RAS (1D) and SOBEK (1D-2D) models, the
hydraulic equations of mass conservation and full momentum are solved so as to
simulate the flood. In HEC-RAS (1D) the total discharge is distributed to the
channel and the floodplain. In SOBEK (1D-2D), it is assumed that networks are
two independent map layers, with the 2D network map layer overlaying the 1D
network. The computational code determines the connection between the 1D and
2D layers, based on the map coordinates for the center of the 2D grid cell and the
1D connection node/calculation node. The flood simulation and model results
provide the authorities with relevant information on how the flood will behave at
the location where people live and how the flood will affect them [ 56 ].
Search WWH ::




Custom Search