Tóm tắt Luận án The research on determining optimal size and layout of detetion pond for the mixed urban and agricultural basins

 
reached to a certain rate. 
Requirement: Determing the size and layout form of detention pond in order to 
minimize the construction investment cost and ensure no waterlogging 
corresponding the design storm. 
2.3. Method 
2.3.1. Proposing the sequence to solve the problem 
a. Traditional method 
It is based on experience, by calculating and direct comparision, the designers 
give out the best appropriate method. The steps of solving will be shown in the 
figure 2.9. The advantage of this method is easy to implement and the 
disadvantage is that the calculation volume is too large and the result depends 
subjectively on the proposing person. 
7 
b. New method 
The author proposes a new method to determine the size and form of rational 
layout for detention pond (Figure 2.8) including building objective function, 
solving the optimization problem by regression method, proposing a reasonable 
solution. 
 Figure 2.8. New method Figure 2.9 Traditional method 
(Diagram of determining steps of the reasonable layout of the detention pond) 
The basic advantage of this proposed method is to increase the considered cases 
(more than 500 cases), regardless of the experience of proposer. Thus, the 
rationality of the proposed method is very high. 
2.3.2. Facility to determine the size and layout form of detention pond 
The construction area should take advantage of the natural ponds and farmland 
and are not coincided to the technical infrastructure already planned. 
Proposing some different 
solutions: (canal, 
Establishing methods (hydraulic, 
hydrological calculating – 
determining designing flow rate and 
method scale by modelling) 
Establishing regression 
function 
Studing basin characteristic: 
Topography, geology, land 
using 
Solving multi-objective 
optimization problem (using 
empirical planningmethods to 
solve) 
Proposing selected method 
Proposing some solutions: 
(channel, detention pond) 
Determining the designing flow 
rate of each method 
 Economy, technical calculating of 
each method 
Studing basin characteristic: 
Topography, geology, land using 
Comparing methods, determining 
the reasonable method 
Proposing selected method 
8 
The calculating scenarios are created from the combination of size and layout 
form of the detention ponds (concentrated or dispersed) in the determined region 
under the plane containing thescale and layout form axis (Figure 2.10). 
nnnnnn KB
KB
KB
VTNmVTNiVTNVTNVTKC
VTNmVTNiVTNVTNVTKC
VTNmVTNiVTNVTNVTKC
mBranchiBranchBranchBranchcannalMain
.
.
.
......0201
...........
...........
...........
......0201
......0201
_..._...02_01_
2
1
22222
11111
Figure 2.10. The determining method of layout form scenario 
2.3.3. Determining the size of the drainage system in scenarios 
After proposing the scenarios on size and layout of detention ponds, it is period 
to determine the construction scale in system by following steps: 
1st Step: Determining the preliminary dimensions of channels, pumping stations 
according to the current regulations, namely follow to the formula in TCVN 
7957: 2008, TCVN 4118-85. 
2nd Step: Using Storm Water Management Model 5.0 (SWMM 5.0, EPA USA), 
to simulate and to check the system determined from the 1st Step. This problem 
suggests for the urban drainage system and appling “on-farm detention pond” 
method for the agricultural area. The calculating result of on-farm detention 
pond method is outflows and they are entered into the SWMM 5.0 model as 
inflows at Nodes. 
2.3.4. Establishing the objective function and the constraint conditions 
a. General objective function 
General objective function for reasonable design problem is only minimum the 
construction investment cost. The general objective function form: 
MinCCCCCCC
j
T
1
hdh
m
1
thkj
n
1
đmi
j
hdh
j
htk
j
đm
j
ht 
  (2.25) 
9 
With: Cjht: The total construction investment cost of drainage system for the 
scenario j, Cjđm: The total construction investment cost for drainage headworks 
for the scenario j, Cjhtk: The total construction investment cost for conveying line 
(conveying cannal) for the scenario j, Cjhdh: The total construction investment 
cost for the detention ponds for the scenario j, J: Number of scenario J = 1, 2, 
3, ..., n: Number of cannals in system, T: Number of detention pond in system. 
b. Constraint condition 
The constraint conditions in the simulation: 1) Fngập = 0; no flooding in the case 
of calculating rain. 2) Fhồ < (a%* Flưu vực); detention pond area is smaller than a% 
total drainage area. 3) Construction the detention ponds are workable. 4) 
Assuming the studing field has defined boundaries, planned ground of canal and 
pipepline, identified the slope of each channel segment, and fixed the structure 
form and the detention pond work in the gravity form. 5) The scenarios are 
siminar in work structure, unit price and calculating period. 
c. Establishing the objective function for studying area 
The identification of the objective function for the studying area is done on the 
basis of determining the regression function’s components. Following as 02 
methods below: 
Method 1: Calculating directly the construction cost, establishing the regression 
function between construction cost and discharge. 
Method 2: Construction costs are taken from the project in the region. 
Establishing of regression function between the construction costs and discharge 
or area is to determine the objective function’s components. 
In this study, both of two ways has chosen to build the specific objective 
fundtion to studying areas. 
The order of elaboration of the regression function for each item is implemented 
as follows: 
* Drainage headwork is pumping station. 
The headwork of system is pumping station. The regression function 
corresponding to each pumps forming: 
10 
 Cđm = f1(Qđm); (2.30) 
* Drainage channel system. 
Drainage channel system including multi-channel grade, each channel considers 
a representative length segment L (m). 
 Chtk = ΣCkênh cấp 1 + ΣCkênh cấp 2 + ΣCkênh cấp 3 (2.31) 
* Grade 1 and 2 channels: Assuming the grade 1 and 2 channels have trapezoidal 
cross-section with earth channel and earth channel with embankment. The 
regression functions of grade 1 and 2 channels: 
Ck1,2 = f2(Qk1,2); (2.32) 
* Grade 3 channel: Most grade 3 channels’s common structure is prefabricated 
reinforced concrete culverts or box reinforced concrete culverts and buried 
underground. 
The regression functions of grade 3 channel: 
Ck3 = f3 (Qk3); (2.33a) 
* Detention pond: In this studying, the detention pond has function of regulating 
rainfall and reducing flooding. The regession function of detention pond: 
Chdh = f4(Fh); (2.33b) 
* The specific objective function: 
Regarding to (2.25) and (2.30, 2.31, 2.32, 2.33a, 2.33b) established, the specific 
objective function has following formula: 
C = f1(Qđm) + f2(Qk1,2) + f3(Qkc3) + f4(Fh); (2.34) 
2.3.5. Applying the objective function to choose an optimal scenario 
a. The relationship between discharge and area of the detention pond 
The relationship between total peak discharge on the drainage headwork, total 
peak discharge on channel system and pond area is the inverse relationship. 
b. The relationship between total cost and pond area 
11 
According to the above relationships, the figure 2.26 has summersized the 
relationship between construction investment cost and pond area or pond’s area 
ratio. 
In where: Cđm Construction investment cost in headwork, Chtk Construction 
investment cost for channel system, Chdh Construction investment cost for 
detention pond, Cht Construction investment cost for whole system. 
Figure 2.26. The relationship form between construction investment cost and 
pond area 
2.4. Conclusions of chapter 2 
Using experimental planning method to solve the problem to get the results of 
high accuracy. 
The extreme points of construction investment cost that is minimum (reasonable 
method) depends on the construction unit prices and unit costs for site clearance 
compensation. When compensation unit price is higher, the extreme points will 
close to the origin and when construction cost is higher, the extreme points tend 
to back away from the origin (Figure 2.26). 
The methodology also presents a problem description and resolution procedures. 
With the method described, this methodology can be applied to all the basins 
where have the composite area. 
The Storm Water Management Model 5.0 is selected to simulate the hydrology, 
hydraulics for urban area, on the other hand, the on-farm pond is selected to 
12 
resolve to the agricultural area. Methods and tools described above are applied to 
specific research areas in the west of Hanoi. 
CHAPTER 3. RESEARCHING AND DETERMINING THE OPTIMAL 
SIZE AND LAYOUT OF DETENTION POND FOR THE WESTERN 
HANOI BASIN 
3.1. Selecting and describing the research area 
3.1.1. Selecting the research area 
The west of Hanoi is selected to be a case study for the problem of determining 
the size and form of detention pond. The spindle is Nhue river from the Lien 
Mac culvert to the Ha Dong culvert. The Dam river, Cau Nga river and La Khe 
river also are studied in the research. The incharging area of drainage is 17,965 
ha including 13,917 ha of urban planning, 2.090ha of greenbelts planning along 
Nhue river, 1,958 ha of agricultural land [68]. The drainage system satisfied the 
data and boundary conditions, so it satisfy the requirements of the proposed 
problem in Chapter 2. 
Picture 3.1. The study area in the West of Hanoi 
13 
3.1.2. Features of ponds and lakes in the study area 
The study area has flat topography; natural ponds are few in number and small 
in size. The natural pond is not involved in regulating rainfall. 
3.2. Validation of the Storm Water Management Model 5.0 
3.2.1. Purpose 
To ensure the reliability of the model, consistent with the research problem, 
using this parameter is to simulate the scenario of the research problem. 
3.2.2. Validation results 
Table 3.1: The evaluation table of the error of the calculating and 
measuring process 
Measuring 
station Daily rain 
Max discharge (m3/s) Total discharge (m3) Error 
S/ Measuring 
Calcu
lating 
Error 
(%) Measuring 
Calculati
ng 
Error 
(%) 
Hà Đông 24/8 - 
28/8/2010 
81,58 82,78 1,48 8.052,9 8.076,72 0,70 0,15 
Đồng Quan 105,66 107,16 1,42 10.365,7 10.739,8 3,61 0,18 
From table 3.1, the indicators on the maximum discharge, the total discharge 
and the line form are achieved. So it is possible to apply these simulations to test 
the model. 
Table 3.2: The evaluation table on the error of calculating and measuring 
process (02 rains) 
Measuring 
Station Daily rain 
Max discharge (m3/s) Total discharge (m3) Error 
S/ Measuring 
Calcul
ating 
Error 
(%) Measuring 
Calculatin
g 
Error 
(%) 
Hà Đông 22/5 - 
26/5/2012 
89,60 93,92 4,82 8.819,67 8.773,56 0,52 0,08 
Đồng Quan 101,19 101,36 0,17 11.671,77 11.549,44 1,05 0,32 
Hà Đông 17/8 - 
19/8/2012 
65,11 66,73 2,68 7.192,45 7.340,79 2,06 0,10 
Đồng Quan 100,61 99,08 -0,25 11.091,29 11.022,57 0,62 0,08 
From table 3.2, the error of line form is from 0.08 to 0.32, the error of maximum 
discharge is from 0.17% to 4.82%, the error of total discharge is from 0.52% to 
2.06%. Thurs, the errors were smaller than allowance. 
14 
Therefore, the calculating line and measuring line is well reasionable with each 
other. The SWMM can be appied to calculate for this thesis. 
3.2.3. The model parameters were chosen after testing 
a. The parameters of padded surface 
- Coefficient of surface roughness with hard coating: n = 0.025 – 0.15 
- The parameters on the characteristics of soil permeability: Maximum 
permeable coefficient kmax = 8 mm/h; kmin =0.5 mm/h, the saturated seapage 
time is 5 days, average permeability is 25mm. 
b. The bed roughness coefficient: River and earth channel: n = 0.025 – 0.03; 
river and earth channel with embankment stone roof: n = 0.017. 
c. Calculating time: t = 60 s 
3.3. Establishing the specific objective function for the West basin of Hanoi 
3.3.1. The general objective function 
The Chapter 2 have determined the general objective function as following 
formula: 
C = Cđm + Chtk + Chdh =  
T
hdh
m
thkj
n
đmi CCC
111
 => Min 
3.3.2.Determining the components of objective function 
The construction costs and the clearance price at the fourth quarter period in 
2013 of the west of Hanoi to determine the components of the objective function. 
By considering the impact of costs for site clearance to the selection of 
appropriate layout scripts, author has selected 03 cases of calculations, 
including: TH1 (100% of agricultural land), TH2 (85% of agricultural land and 
15% residential land), TH3 (70% and 30% of agricultural land for residential 
land). 
a. The drainage system headwork is pumping stations 
To determine the parameters in the objective function of Cđm ~ Qđm, this studying 
used the data of 21 pumping station in the west of Hanoi and vicinity area and 
updated to the cost estimates with unit price at the fourth quarter period in 2013. 
15 
The regression function: 
TH1: Cdm = 1,6578*(Qdm)1,4532; (3.4) 
TH2: Cdm = 1,7064*(Qdm)1,4491; (3.5) 
TH3: Cdm = 1,7566*(Qdm)1,4451; (3.6) 
With, Cdm is construction 
investment cost in billion Dong, 
Qdm is the upstream discharge in 
m3/s. 
Figure 3.18. Diagram showing the 
relationship of Cđm ~ Qđm 
b. Drainage channel system 
* The regression relationship between discharge and construction investment 
cost of the channel grade 3: 
The channel grade 3 usually structured as reinforce concrete round sewer or box 
sewer buried underground. The regression function has formula as the following 
table: 
TH The regression function at the channel grade 3 with two structural types 
TH1 
TH2 
TH3 
 Rectangular sewer: 
Ckc3 = 0,6417*ln(Qkc3) + 0,6246; (3.11) 
Ckc3 = 0,6764*ln(Qkc3) + 0,7114; (3.12) 
Ckc3 = 0,7010*ln(Qkc3) + 0,7983; (3.13) 
 Box sewer: 
 Ckc3 = 0,4731*ln(Qkc3) + 0,487; (3.8) 
 Ckc3 = 0,5179*ln(Qkc3) + 0,5736; (3.9) 
 Ckc3 = 0,5628*ln(Qkc3) + 0,6603; (3.10) 
Where, Ckc3 is the construction investment cost in billion Dong per 100m, Qkc3 is 
discharge of the channel grade 3 in m3/s. 
* The regression relationship between discharge and construction investment 
cost of channel grade 1, 2: 
The channel grade 1 and 2 in researching system is the trapezoidal channel. The 
author used the data of Nhue River to calculate for channel grade 1 and 2. 
The regression function for the dredged and reinforced roof channel: 
 TH1: Cc1-2 = 0,000003*(Qc1-2)2 + 0,0032*Qc1-2 + 2,393; (3.14) 
 TH2: Cc1-2 = 0,000003*(Qc1-2)2 + 0,0041*Qc1-2 + 2,3192; (3.15) 
16 
 TH3: Cc1-2 = 0,000002*(Qc1-2)2 + 0,0051*Qc1-2 + 2,2455; (3.16) 
The regression equation for the dredged channel: 
 TH1: Cc1-2 = 0,000003*(Qc1-2)2 + 0,0032*Qc1-2 – 0,0558; (3.17) 
 TH2: Cc1-2 = 0,000003*(Qc1-2)2 + 0,0041*Qc1-2 - 0,1296; (3.18) 
 TH3: Cc1-2 = 0,000002*(Qc1-2)2 + 0,0051*Qc1-2 - 0,2034; (3.19) 
Where: Cc1-2 Investment cost for channel grade 1 and 2 (billion/100m); Qc1-2 
Designing discharge for channel grade 1 and 2 (m3/s); 
c. The detention pond 
Assuming the detention pond was built and reinforced roof by stone 
The regression function: 
 TH1: Chdh = 4,8997*(Fho) + 9,7424; (3.20) 
 TH2: Chdh = 6,2804*(Fho) + 9,7424; (3.21) 
 TH3: Chdh = 7,6611*(Fho) + 9,7424; (3.22) 
Where, Chdh is construction investion and site clearance cost in billion, Fho is 
detention pond area in ha. The Software Eview 6.0 is applied to test the 
regression function on the probability of error and stability. 
3.3.3. The specific regression function for the West basin of Hanoi 
From the general regression function and established regression function, the 
specific regression function has formula as following: 
 C = f1(Qđm) + f2(Qc1-2) + f3(Qkc3) + f4(Fh) => min 
3.4. Establishing the scenarios of layout of the detention ponds. 
Based on the literature on drainage plan for the west of Hanoi, the drainage basin 
is divided into 03 drainage headwork. The detention pond area selected by 
percentage of drainage area ranged from 0% to 6%. About detention pond layout, 
this study considered 03 scenario groups: (1) pond concentrated in the drainage 
headwork, (2) pond distributed along the main channel (PT1), (3) pond 
distributed along the main channel and branch channels (PT2). 
a. The scenarios of the detention pond concentrated at the drainage headwork 
(TT) 
17 
The author proposes 12 scenarios of the detention pond that concentrated at the 
drainage headwork and had the same area ratio in the 03 drainage headworks. 
The area ratio of detention pond ranged from 0% (no pond) to 6%. Otherwise, 
the authorlso proposes 24 scenarios of the different detention ponds in every 
drainage headworks. Totally, 36 scenarios were considered. 
b. The scenarios of the distributed detention pond 
* The scenario of detention pond distributed along the main channel (PT1) 
including 11 scenarios, each scenario has 11 locations of detention pond.. 
* The scenario of detention pond dstributed along the main channel and branch 
channels (PT2) including 11 scenarios, each scenario has 38 locations of 
detention pond. 
3.5. The result 
3.5.1. The results of simulation of the flow corresponding to each scenario 
The discharge in calculating or modeling is the average value in hour. According 
to designing standard, this value will be used in designation. 
3.5.1.1. The scenarios of the detention pond concentrated at the drainage 
headwork (TT) 
Figure 3.10. The relationship between 
upstream discharge and the detention pond 
area 
Figure 3.11. At Lien Mac inflow when the 
detention pond area changing 
At the upstream, discharge decrease when the pond area increases. The 
reduction is smaller and smaller (Figure 3.10 and 3.11). The peak discharge 
summary of the headworks reaches to minimum at TT666 (pond ratio 6%). 
When the pond area ratio is different at the headworks, the headwork discharge 
18 
summary reaches to minimum corresponding to pond ratio 5.3% (as scenario 
TT664). 
3.5.1.2. The scenarios of the distributed detention pond 
The calculation results for the scenario group (PT1) and (PT2) present the total 
peak discharge at upstream and total peak discharge at channels are minimum 
when the pond ratio is 4.35% (as scenario PT1-6) and 4.58% (as scenario PT2-6) 
corresponding to the scenario groups PT1 and PT2. 
3.5.1.3. Comparision between scenarios of concentrated and distributed 
detention pond 
The comparision is shown in the following diagram: 
Figure 3.15. The diagram of relationship 
between upstream discharge and the 
detention pond area 
Figure 3.16. The diagram of 
relationship between the total of peak 
discharge and the detention pond area 
The headwork peak discharge reduces in all 3 scenario groups (TT, PT1 và PT2). 
In which, the scenario of distributed detention pond along the main chaneel and 
branch channel (PT2) gave out the minimum value (figure 3.15). 
The diagram 3.16 has shown that the peak discharge of the channel system 
corresponding to scenario (TT) unchanged (horizontal line). On the other hand, 
both of the scenarios (PT1) and (PT2) decreased strongly. Especially, the 
scenario of pond distributed along the main channel and branch channels have 
strongest reduction (the pink line) 
Summary: Regulating effect of the pond depends on not only the pond scale, but 
also the form layout (concentrated or distributed). Considering overview of the 
system, if the distribution of detention pond is more and more large, the effect of 
peak di