Application of Support Vector Machine for River flow Estimation

International journal of Engineering, Business and Management (IJEBM) [Vol-4, Issue-2, Mar-Apr, 2020]
7781-ISSN: 2456https://dx.doi.org/10.22161/ijebm.4.2.1
29|Pagewww.aipublications.com
Application of Support Vector Machine for
River flow Estimation
Reza Dehghani*1
, Hassan Torabi Poudeh2
1
Phd student, Department of Water Engineering, , university of lorestan,Iran
2
Associate Professor of Water Engineering, University of Lorestan, Khorramabad, Iran
*Corresponding author
Abstract— In recent years application of intelligent methods has been considered in forecasting hydrologic
processes. In this research, month river discharge of kakareza, a river located in lorestan province at the
west of Iran, was forecasted using Support vector machine and as genetic programming Inference System
methods in dehno stations. In this regard, some different combinations in the period (1979-2015) as input
data for estimation of discharge in the month index were evaluated. Criteria of correlation coefficient, root
mean square error and Nash Sutcliff coefficient to evaluate and compare the performance of methods were
used. It showed that combined structure by using surveyed inelegant methods, resulted to an acceptable
estimation of discharge to the kakareza river. In addition comparison between models shows that Support
vector machine has a better performance than other models in inflow estimation. In terms of accuracy,
Support vector machine with correlation coefficients ( 0.970 ) has more propriety than root mean square
error (0.08m3
/s ) and Nash Sutcliff ( 0.94 ) . To sum up, it is mentioned that Support vector machine method
has a better capability to estimate the minimum, maximum and other flow values.
Keyword— Genetic Programming, Estimate, Kakareza River, Support Vector Machine
I. INTRODUCTION 1
Nowadays one of the most important issues for managing 2
flood and preventing the economic and physical damage 3
caused by it, are correctly prediction the river flows. 4
Accurate estimates of inflow to reservoirs could play an 5
important role in the planning and management of water 6
resources. But factors and various effects that have an 7
influence on this phenomenon that analysis makes difficult. 8
The statistical Models and the regression models are the 9
most commonly analytical techniques that frequently 10
according to a linear resolution of these phenomena 11
presented results along with error and cannot model with 12
acceptable accuracy temporal changes the phenomenon. So 13
choose a model that could using affective factors, estimates 14
acceptable the input current seems imperative. Recently 15
artificial intelligent (AI) techniques have been applied to 16
estimate/predict the discharge(Kisi and Cobaner 2009). 17
These AI techniques are simple, robust and can handle 18
complex non-linear processes with ease. From the 19
literature, it is seen that the AI techniques such as gene 20
expression programming (GEP), support vector machines 21
(SVM), etc. were used to predict the discharge(Wang et al. 22
2008). As they are fully non-parametric, AI techniques 23
have a major advantage that they do not require a priori 24
concept of the relations between the input variables and 25
output data (Bhagwat and Maity 2012). A classical feature 26
of AI is that the models that are able to analyze the 27
stochasticity, dynamicity, patterns and attributes in the 28
input variables used to simulate the evaporation data, and 29
so, are considered more feasible over the other methods of 30
the estimating of discharge data (e.g. experimental 31
approaches and physically-based models). 32
Examples using the SVM capability include: Stage– 33
discharge modeling (Barzegar et al 2019;Sahoo et al 2019; 34
Elkiran et al 2019; Rezaei et al 2019; Adnan et al 2019; 35
Fathian et al 2019; Yassen et al 2018; Imani et al 2018; 36
Tongal et al 2018; Ghorbani et al 2016;Londhe and 37
Gavraskar 2018;Ghazvinei et al 2017; Karahan et al2014; 38
He et al 2014). 39
In a research, Presented appropriate method for 40
seasonal flow discharge and horary used by SVM, in the 41
research using the amount of snow equivalent water and 42
the volume of the previous periods, forecasted amount 43
volume flow for the six-month time scales and 24-hour 44
than the result showed satisfactory model (Asefa et 45
al.2005). Using by genetic programming were modeled the 46
process rainfall-runoff with daily data in two fairly big 47
China basin that results of GP showed good agreement 48

with real data (Jayawardenaet al.2005). In this paper, the 1
support vector machine (SVM) is presented as a promising 2
method for hydrological prediction. Through the 3
comparison of its performance with those of the ARMA 4
and ANN models, it is demonstrated that SVM is a very 5
potential candidate for the prediction of long-term 6
discharges(Lin et al,2006). Also in order to forecasts daily 7
discharge flow Shevell river in America used of genetic 8
programming and artificial neural network and showed 9
both methods had acceptable results but GP has relatively 10
higher precision than artificial neural network 11
(Guven.2009). Support Vector Machine (SVM) is used to 12
forecast daily river flow and the results of these models are 13
compared with observed daily values. The results showed a 14
good performance in network support vector machine is 15
estimating the daily discharge(Moharrampour et al.2012). 16
In total, according to the researches done and the fact that 17
the river Kakareza is one of the most important rivers in 18
Lorestan province and the most important source of water 19
supply to different parts of its neighboring areas, which 20
over the past decades has reduced the flow rate of the river 21
in the basin, which can be explained by lower river basin 22
fluxes and surface flows. Therefore, the importance of 23
river discharge modeling and management measures to 24
improve its water quality is more than necessary. 25
Therefore, the aim of this study was to estimate the 26
discharge of Kakareza River using a support vector 27
machine based on the use of the principle of inductive 28
minimization of structural error. In simulation, the learning 29
method with monitoring in radial base functions makes 30
estimating the parameter of high speed and error Less than 31
other kernel functions.) Vapnik,1995;Vapnik,1998). 32
33
II. MATERIALS AND METHODS 34
Case study and used data 35
Study area is kakareza river in the province of Lorestan, 36
Iran. this river is one of permanent rivers in the province 37
and is originated from southeastern mountains of aleshtar 38
and biranshahr (dehno). When this river passes through 39
aleshtar suburbs it is known as kakareza. The river is 40
between "15 ° 48 ° 49 ° longitude to the" 22 ° 32 to "52 ° 41
33 degrees latitude and it flows across the east of 42
Khorramabad (capital city of Lorestan Province). This 43
river is one of initial branches of karkhe river in zagros 44
mountains and have the average altitude of 1550 meters 45
above sea level. kakareza river basin area is about 1148 46
square kilometers and its river has a length of 85 km. 47
kakareza river joins Kashkan, Cimmeria, and Karkhe rivers 48
in its way and eventually pours into the Persian Gulf. The 49
geographical location of the study area is shown in Figure 50
1. In this study, available runoff data at monthly scale of 51
horod station (kakareza) from 1979 to 2015 in Lorestan 52
Regional Water was used. Table 1, the statistical properties 53
of kakareza river is shown during the mentioned period. 54
55
56
Fig 1. Geographical location kakareza river 57
58
59
60

Table1. Statistical properties discharge parameter daily discharge (1979-2015) 1
Parameter
Training Testing
Minimum Mean Maximum Minimum Mean Maximum
Q 0.01 2.718464 25.15 0.05 1.701161 21.69
2
One of the most important steps in modeling, is select the right combination of input variables. Also shown in Table 3
2.The structure of input combinations. 4
Table2.The structure of input combinations 5
OutputInputStructure
Q(t)Q(t-1)1
Q(t)Q(t-1)Q(t-2)2
Q(t)Q(t-1)Q(t-2)Q(t-3)3
Q(t)Q(t-1)Q(t-2)Q(t-3)Q(t-4)4
6
In this Table Q(t-4), Q(t-3), Q(t-2), and Q(t-1) are 7
respectively discharge in t-4, t-3, t-2, and t-1 time as input 8
and Q(t) is discharge in t time as output being considered. 9
Due to the significant cross-correlation between input and 10
output data, in order to achieve an optimal model to 11
estimate the inflow to kakareza river use of different 12
combinations of input parameters that showed them in 13
Table3. To estimate input discharge kakareza river using 14
by Gene Expression Programming and Support Vector 15
Machine with have catchment hydrometric data from 432 16
registered records during the period (1979-2015), count in 17
345 records to training and 87 remaining records to 18
verification. 19
Table 3.Correlation between input and output parameters 20
Q(t-1) Q(t-2) Q(t-3) Q(t-4)
Q(t) 0.980 0.964 0.928 0.784
21
Gene Expression Programming 22
Gene Expression Programming method presented with 23
Ferreira in 1999 (Ferreira.2001). This method is a 24
combination of genetic algorithms (GA) and genetic 25
programming (GP) method than in this, simple linear 26
chromosomes of fixed length are similar to what is used in 27
genetic algorithm and branched structures with different 28
sizes and shapes aresimilar to the decomposition of trees in 29
genetic programming.Since this method all branch 30
structures of different shapes and size are encoded in linear 31
chromosome with fixed length, this is equivalent than 32
Phenotype and Genotype are separated from each other and 33
system could use all evolutionary advantagesbecause of 34
their. Now,however the Phenotype in GEP included branch 35
structures used in GP, but the branch structures be 36
inferences by GEP (than also calledtreestatement) are 37
explainer all independent genomes. In short can say 38
improvements happened in linear structure then is 39
expressed similar with tree structure and this causes only 40
the modified genomemoved to the Next Generation and 41
don't need with heavy structure to reproduce and mutation 42
(Ferreira.2001). In this method different phenomena are 43
modeling by collection of functions and terminals. 44
Collection of functions generally include the main 45
functions of arithmetic {+, -, ×, /}, the trigonometric 46
functions or any other mathematical function {√, x2, sin, 47
cos, log, exp, …} or defined functions by author whom 48
believed they are appropriate for interpreting model. 49
Collection of terminals consist problem's constants values 50
and independent variables (2001). For applying gene 51
expression programming method is used GenXproTools 52
4.0 Software. In order to obtain more information can 53
recourse to (Ghorbaniet al.2012). 54
Support Vector Machine 55
Support Vector Machine is anefficient learning system 56
based on optimization theory that used the principle of 57
induction minimization Structural error and results an 58
overall optimal solution(Vapnik,1998). In regression 59
model SVM is estimated function associated with the 60
dependent variable Y as if is afunction of several 61
independent variables X(Xu et al.2007).Like other 62

regression problems is assumed the relationship between 1
the dependent and independent variables to be determined 2
with algebraic function similar f(x) plus some allowable 3
error (𝜀). 4
f(x)=W T
.∅(x)+b (1) 5
(2) 6
If W is coefficients vector, b is constant characteristic 7
of regression function, and also ∅ is kernel function, then 8
goal is to find a functional form for f(x). It is realized with 9
SVM model training by collection of samples (train 10
collection). To calculate w and b require to be optimized 11
error function in 𝜀-SVM with considering the conditions 12
embodied in Equation 4(Shin et al.2005). 13
WT
. ∅(Xi)+b-yi
≤ ε+ εi
*
,
1
2
WT
. W + C ∑ εi
N
i=1 + 14
C ∑ εi
∗N
i=1 (3) 15
yi − WT
. ∅ (Xi) − b ≤ ε + εi , εi , εi
∗
≥ 0 , i = 16
1,2, … , N (4) 17
In the above equations, C is integer and positive, that 18
it’s factor of penalty determinant when an error occurs. ∅ is 19
kernel function, N is number of samples and two 20
characteristics εi and εi
∗
are shortage variables. Finally can 21
rewrite SVM function as follow(Shin et al,2005): 22
f(x)= ∑ α̅i
N
i=1 ∅(xi)T
. ∅(x)+b (5) 23
Average Lagrange Coefficients α̅i in characterized 24
space is ∅(x).Maybe calculation be very complex. To 25
solve this problem, the usual process of SVM model is 26
choose a kernel function as follow relation. 27
K(XJ ,X)=∅(Xi)T√ b2
-4ac (6) 28
Can be used of different kernel functions to create 29
different types of 𝜀-SVM. Various kernel functions used in 30
SVM regression models are: Polynomial with three 31
Characteristics of the target, Radial Basis Functions (RBF) 32
with one Characteristics of the target, and Linear 33
respectively, are calculated as follows 34
relation(Vapnik.1998). 35
k(xi,xj)=(xi.xj)
d
(7) 36
K(x,xi)=exp (-
‖x-xi‖
2
2σ2 ) (8) 37
k(xi,xj)=xi.xj (9) 38
Evaluation Criteria 39
In this research to evaluate the accuracy and efficiency 40
of the models was used indices Correlation Coefficient 41
(CC), Root Mean Square Error (RMSE), Nash–Sutcliffe 42
coefficient (NS), and Bias according to the following 43
relations.Best values for these four criterions are 44
respectively 1, 0, 1, and 0. 45
CC=
∑ (xi-x̅)(yi-y̅)N
i=1
√∑ (xi-x̅)
2N
i=1 ∑ (yi-y̅)
2N
i=1
-1≤ R ≤1 (10) 46
RMSE=√
1
N
∑ (xi-yi
)
2N
i=1 (11) 47
NS=1-
∑ (xi-yi)
2N
i=1
∑ (xi-y̅)
2N
i=1
-∞≤ NS ≤1 (12) 48
In the above relations xi and yi are respectively 49
observed and calculated values in time step i, N is number 50
of time steps, x̅ and y̅ are respectively mean observed and 51
calculated values. 52
53
III. RESULTS AND DISCUSSION 54
The general purpose of intelligent models is to express the 55
relation between variables that find their complexity 56
difficult in the nature of work with high uncertainty. Daily 57
stream flow is one of the important hydrological 58
parameters that is of great importance in future steps. In 59
order to reduce the error and also to estimate the daily flow 60
rate parameter with high accuracy using the lowest input 61
parameters, this method has been used which will provide 62
a better performance compared to approximate methods. 63
The aim of this study is to obtain this natural complexity 64
between hydrological parameters and provide a model for 65
prediction in the future, because daily discharge is more 66
important than other parameters, so this parameter is 67
selected as the target variable. 68
The results of Gene Expression Programming 69
Using gene expression programmingdue to the 70
selection of variables in the model and remove variables 71
with less impact and also ability to provide a clear 72
relationship were considered to estimating inflow to the 73
kakareza river. Since ever four input areincorporated to 74
determining the significant variables and more reviews in 75
addition four of the original operator (F1) and the states 76
based on arithmetic operators default (F2). The reason for 77
choice this type of operator has been based on studies 78
(Ghorbaniet al.2012) and (Khatibi et al.2012). 79
F1:{+, −,∗,/, √, Exp, Ln,2
,3
, ∛, Sin, Cos, Atan} (13) 80
F2:{+, −,∗,/}(14) 81
Results of gene expression programming model for 82
both operator in Table4 show that F2 operator in both 83
stages training and verification with maximum correlation 84
coefficient R=0.88, root mean square error RMSE=0.15 85
and NS=0.76 has high accurate than other operators. 86
Therefore gene expression programming with F2 operator 87

include four the main mathematical operators with a simple 1
mathematical relationship has the most accurate to 2
estimating inflow to the kakareza river. The scatter plots of 3
gene expression programming related to the verification 4
stage in Fig(2-b) showthe fit line of computational values 5
with four mathematical operators to the best fit line y=x.As 6
is clear from this Fig, all of the estimated and observation 7
values are in the fit line except few points that are not 8
inbisector line which it isdenoted the estimated and 9
observed values of equality on the line (y=x). The 10
operation of gene expression programming is acceptable to 11
estimating inflow, it should be noted this model worked 12
fine, meanwhile these values estimate equal to actual 13
values. 14
These results are consistent with Kisi and Shiri (2012) 15
research. And it can be stated that the equation obtained 16
from gene expression planning is obtained from the 17
random combination of the sum of the terminals and 18
functions. Therefore, if the relationship between inputs and 19
outputs is linear, but the operators sin, cos, etc. are selected 20
in the set of functions, the gene expression planning uses 21
the selective operators to extract the relationship, which 22
reduces the accuracy of the model. In this study, to 23
increase the precision of the model of the operators’ sin, 24
cos, and so on, and with accuracy and simplicity, the 25
model derived from four basic mathematical operations 26
was proposed to estimate sediment load. 27
Table 4. The results of the planning model of gene expression programming using two sets of selected mathematical 28
operator 29
Number Model
Training Testing
R
RMSE
(m3
/s) NS R
RMSE
(m3
/s) NS
1
F1 0.70 0.31 0.63 0.76 0.25 0.64
F2 0.73 0.32 0.64 0.78 0.23 0.66
2
F1 0.75 0.38 0.68 0.80 0.22 0.68
F2 0.76 0.34 0.69 0.80 0.21 0.71
3
F1 0.79 0.26 0.71 0.82 0.19 0.72
F2 0.80 0.21 0.73 0.84 0.19 0.73
4
F1 0.80 0.19 0.73 0.87 0.15 0.76
F2 0.82 0.15 0.75 0.88 0.15 0.76
30
31
32
33
34
35
36
37
38
39
40
41
42
43
Fig 2. The resulting chart of optimal values of gene expression programming model to the data step verification. a) 44
Computational and observational values of time. b) The scatter plot between estimated and observed values 45
b-F2
y = 0.7226x + 0.7688
R² = 0.878a

The results support vector machine 1
In order to estimate the inflow to the kakareza river by 2
SVM model can examine types of kernel function, than 3
was selected linear kernel, polynomial and radial basis 4
functions that are common types used in hydrology. The 5
results of study models is given in Table5. According to 6
this table combined model number4 with radial basis 7
functions kernel has the highest correlation coefficient 8
R=0.97, lowest root mean square error RMSE=0.08 m3
/s 9
and NS=0.94 in verification stage that has optimal solution 10
than other models. In Fig3 shown the best model for 11
verification of data. 12
As shown in Fig(3-b) is clear computational values 13
discharge of the support vector machine model verification 14
corresponded with observed values. In this Fig can be seen 15
insignificant difference some of values with the best fit line 16
y=x. According to the diagram (3-a) can be seen high 17
capability of the model. Also, according to Table 5, a high 18
performance support vector machine has been shown in the 19
Kakareza River discharge estimation, even if only one 20
input parameter is used, which leads to the presence of 21
statistical deficiencies in this network with Having the 22
minimum input parameters, such as flow rate, one day 23
before, would have acceptable performance in flow rate 24
forecasting. In Fig. 3, changes in computational and 25
observational values of time are shown, it is seen that this 26
model was in the estimation of most of the values of 27
acceptable accuracy in such a way that these estimates are 28
close to their actual value. The results are consistent with 29
the research by Buyukyildiz and Kumcu (2017) and 30
Nourani et al (2015). This can be explained by the fact that 31
the backup machine is based on the use of the principle of 32
inductive minimization of structural error. Therefore, in 33
simulation, using a learning method with monitoring in 34
radial base functions, the prediction of the parameter has a 35
higher velocity and less error than other kernel functions, 36
and this is a privilege of radial base functions. 37
Table 5. Results of the three kernel methods used in Support Vector Machine for training and verification data 38
Number Kernel
Training Testing
R
RMSE
(m3
/s) NS R
RMSE
(m3
/s) NS
1
RBF 0.87 0.13 0.76 0.90 0.16 0.88
Poly 0.74 0.19 0.67 0.79 0.17 0.80
Line 0.64 0.24 0.54 0.71 0.29 0.69
2
RBF 0.89 0.12 0.80 0.93 0.11 0.90
Poly 0.76 0.17 0.69 0.81 0.16 0.82
Line 0.67 0.22 0.58 0.75 0.27 0.71
3
RBF 0.90 0.11 0.81 0.94 0.10 0.92
Poly 0.79 0.15 0.75 0.81 0.14 0.84
Line 0.69 0.18 0.62 0.79 0.27 0.72
4
RBF 0.91 0.09 0.82 0.97 0.08 0.94
Poly 0.81 0.14 0.75 0.84 0.13 0.87
Line 0.80 0.18 0.66 0.80 0.26 0.73
39
40
41
42
43
44
45
46
47

1
2
3
4
5
6
Fig 3. The resulting chart of optimal values ofsupport vector machine model to the data step verification. a) Computational 7
and observational values of time. b) The scatter plot between estimated and observed values 8
Comparison Performance of models 9
Choosing the optimal solution for each of the models and compare together was defined all three methods can with good 10
accurate simulate inflow to the kakareza river. As can be seen in Table6throughtheused models, support vector machine 11
model have highest accurate R=0.97, lowest root mean square error RMSE=0.08m3
/s and highest Nash-Sutcliffe NS=0.94 in 12
verification stage. Comparison of gene expression programming model and support vector machine model shown proximity 13
the results of these two models. In Fig4shown the results of all three models to the observed value during the time that all 14
two models good function, whereas support vector machine model is well covered minimum, maximum, and middle values. 15
Table 6. The final results of the training and verification gene expression programming and support vector machine 16
Model
Training Testing
R
RMSE
(m3/s) NS R
RMSE
(m3/s) NS
S.V.M 0.91 0.09 0.82 0.97 0.08 0.94
GEP 0.82 0.15 0.75 0.88 0.15 0.76
17
Fig 4. The scatter plot between estimated and observed values gene expression programming and support vector machine 18
models for recorded data in verification stage 19
0
5
10
15
20
25
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85
Observation
GEP
SVM
Q(m3/s)
time (month)
b-RBF
y = 0.8857x + 0.2758
R² = 0.970
a

1
Fig 5. All two models graph optimization error as a percentage of the mean observed value 2
Finally difference between the observed inflow values 3
and optimal computational models calculated as a 4
percentage of the mean observed values (error value) and 5
was drawn this diagram in comparison with the data 6
recorded (Fig5). As seen in this Fig, more errors to ever 7
three models has been ±5 band the highest error rate gene 8
expression programming and support vector machine 9
models are respectively 6.61 and3.10 percent of the mean 10
observed values. Among these models (GEP and SVM) 11
svm model has lowest error value. Totally due to the high 12
estimation accuracy and reliability gene expression 13
programming and support vector machine models the 14
correlation between the observed values and the computed 15
values are respectively 0.970 and 0.880. Also the results of 16
was significant estimated and observed values in the 17
probability levels %5 and %10 shown, SVM model has 18
significant correlation in both probability levels. 19
20
IV. CONCLUSIONS 21
In this research, we tried to evaluated performance 22
some models to simulating discharge to the kakareza river 23
In the province lorestan using by discharge month data in 24
kakareza river. Used models include gene expression 25
programming and support vector machine models. 26
Observed inflow values compared with estimated inflow in 27
these models (GEP and SVM). The results summarized as 28
follows: 29
A: SVM model has high accurate and a little error to 30
estimate minimum, maximum, middle values and peak 31
discharge, and high correlation with the observed value. B: 32
Gene expression programming model with the four basic 33
arithmetic operations has high ability to estimating 34
minimum, maximum, and middle values and peak values, 35
also support vector machine with radial basis functions 36
kernel has high ability estimating minimum and middle 37
values but to estimating maximum values doesn't have 38
enough operation. C: Increasing the number of parameters 39
in the various models to simulating inflow cause to 40
improve operation to estimating inflow. D: Estimating 41
inflow using by combined models have lower error and 42
high correlation than other models to estimated inflow in 43
reservoirs dam. 44
Totally the results of this research showed support 45
vector machine method has highest accurate than other 46
models. As research results (Ghorbaniet 47
al.2016),(Moharrampour et al.2012) and (Asefa et al.2005) 48
has been proven its. Also this research shown using of 49
gene expression programming and support vector machine 50
models could use to estimating inflow to the river. 51
52
V. ACKNOWLEDGMENTS 53
The authors are very grateful of the Regional Water 54
Company, Lorestan Province, Iran, because they 55
participate in gathering the data required to perform the 56
work. 57
58
FUNDING 59
The University of Lorestan, Khorramabad, Iran , supported 60
our research work (Grant No. 1). 61
Compliance with ethical standards 62
63
CONFLICT OF INTEREST 64
-6
-4
-2
0
2
4
6
8
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85
GEP
SVM
Error
Time (month)

The authors declare that they have no conflict of interest. 1
2
REFERENCES 3
[1] Adnan M, Liang Z, Heddam S, Zounemat Kermani M, Kisi 4
O, Li B (2019) Least square support vector machine and 5
multivariate adaptive regression splines for streamflow 6
prediction in mountainous basin using hydro-meteorological 7
data as inputs. Journal of hydrology, 580: 983-993 8
[2] Aggarwal SK, Goel A, Singh VP (2012) Stage and discharge 9
forecasting by SVM and ANN techniques. Water 10
ResourManag 26:3705–3724 11
[3] Asefa T, Kemblowski M, McKee M, Khalil A (2005) Multi- 12
time scale stream flow predictions: The support vector 13
machines approach. J. of Hydrology, 318 (1-4): 7-16. 14
[4] Barzegar R, Ghasri M, Qi Z, Quilty J, Adamowski J (2019) 15
Using bootstrap ELM and LSSVM models to estimate 16
river ice thickness in the Mackenzie River Basin in the 17
Northwest Territories, Canada. Journal of Hydrology, 18
577:880-903 19
[5] Bhagwat PP, Maity R (2012) Multistep-ahead river flow 20
prediction using LS-SVR at daily scale. J Water Resource 21
Prot 4:528–539 22
[6] Buyukyildiz M, Kumcu SY (2017) An Estimation of the 23
Suspended Sediment Load Using Adaptive Network Based 24
Fuzzy Inference System, Support Vector Machine and 25
Artificial Neural Network Models. Water Resources 26
Management 31(2):1-17 27
[7] Dibike YB, Velickov S, Solomatine D, Abbott MB (2001) 28
Model induction with support vector machines: introduction 29
and applications. J Comput Civil Eng 15:208–216 30
[8] Duan QY, Sorooshian S, Gupta VK (1994) Optimal use of 31
the SCEUA global optimization method for calibrating 32
watershed models. J Hydrol 158:265–284 33
[9] Elkiran G, Nourani V, Abba SL (2019) Multi-step ahead 34
modelling of river water quality parameters using ensemble 35
artificial intelligence-based approach. Journal of Hydrology, 36
577:962-974 37
[10] Fathian F, Mehdizadeh S, KozekalaniSales A, Mir J, Safari 38
S (2019) Hybrid models to improve the monthly river flow 39
prediction: Integrating artificial intelligence and non-linear 40
time series models. Journal of Hydrology, 575:1200-1213 41
[11] Ferreira C (2001) Gene expression programming: a new 42
adaptive algorithm for solving problems. Complex Systems, 43
Vol.13(2): 87–129. 44
[12] Ghazvinei T, ShamshirbandSh, MotamediSh, Hassanpour 45
H, Salwana E (2017) Performance investigation of the dam 46
intake physical hydraulic model using Support Vector 47
Machine with a discrete wavelet transform algorithm. 48
Computers and Electronics in Agriculture, 140:48-57 49
[13] Ghorbani MA, Khatibi R, Asadi H, Yousefi P (2012) Inter- 50
Comparison of an Evolutionary Programming Model of 51
Suspended Sediment Time-series whit other Local Model. 52
INTECH.doi. org/10.5772/47801,. 255-282 53
[14] Ghorbani MA, KhatibiR,Goel A, FazeliFard MH, Azani A 54
(2016) Modeling river discharge time series using support 55
vector machine and artificial neural networks. Environ Earth 56
Sci. DOI: 10.1007/s12665-016-5435-6 57
[15] Goel A, Pal M (2012) Stage–discharge modeling using 58
support vector machines.IJE Trans A Basics. 59
doi:10.5829/idosi.ije.2012.25. 01a.01 60
[16] Guven A (2009) Linear genetic programming for time-series 61
modelling of daily flow rate. Journal Earth System Science, 62
vol 118, 157-173. 63
[17] He Z, Wen X, Liu H, Du J (2014) A comparative study of 64
artificial neural network, adaptive neuro fuzzy inference 65
system and support vector machine for forecasting river flow 66
in the semiarid mountain region. Journal of Hydrology, 67
509:379-386 68
[18] Imani M, Chao H, Lan W, Kuo C (2018) Daily sea level 69
prediction at Chiayi coast, Taiwan using extreme learning 70
machine and relevance vector machine. Global and Planetary 71
Change, 161:211-221 72
[19] Jayawardena AW, Muttil N, Fernando TMKG (2005) 73
Rainfall-Runoff ModellingusingGenetic Programming. 74
International Congress on Modelling and Simulation Society 75
of Australia and New Zealand December: 1841-1847.. 76
[20] Karahan H, Iplikci S, Yasar M, Gurarslan G (2014) River 77
Flow Estimation from Upstream Flow Records Using 78
Support Vector Machines. Journal of Applied 79
Mathematics.14(7):7-14 80
[21] Khatibi R, Naghipour L, Ghorbani MA, Aalami MT (2012) 81
Predictability of relative humidity by two artificial 82
intelligence techniques using noisy data from two 83
Californian gauging stations. Neural computing and 84
application: 643-941. 85
[22] Kisi O, Shiri J (2012) River suspended sediment estimation 86
by climatic variables implication: Comparative study among 87
soft computing techniques. Computers & 88
Geosciences,43(4):73-82 89
[23] Kisi O, Cobaner M (2009) Modeling river stage–discharge 90
relationships using different neural network computing 91
techniques. Clean Soil Air Water 37:160–169 92
[24] Lin J, Cheng C, Chau K (2006) Using support vector 93
machines for long-term discharge prediction. Hydrological 94
Sciences Journal.51(4):599-610. 95
[25] Liong SY, Sivapragasam C (2002) Flood stage forecasting 96
with support vector machines. J Am Water Resour 38:173– 97
186 98
[26] Londhe S, Gavraskar S (2018) Stream Flow Forecasting 99
using Least Square Support Vector Regression. Journal 100
of Soft Computing in Civil Engineering , 2(4):56-88 101
[27] Misra D, Oommen T, Agarwal A, Mishra SK, Thompson 102
AM (2009) Application and analysis of support vector 103
machine based simulation for runoff and sediment yield. 104
BiosystEng 103:527–535 105
[28] Moharrampour M, Mehrabi A, Katouzi M (2012) Daily 106
Discharge Forecasting Using Support Vector Machine. 107
International Journal of Information and Electronics 108
Engineering.2(5):769-772 109
[29] Nourani V, Alizadeh A, Roushangar K (2016) Evaluation of 110
a Two-Stage SVM and Spatial Statistics Methods for 111

Modeling Monthly River Suspended Sediment Load. Water 1
Resources Management,30(1):393-407 2
[30] Rezaei M, Kim B, Fallah H, Alghmand S (2019) Daily river 3
flow forecasting using ensemble empirical mode 4
decomposition based heuristic regression models: 5
Application on the perennial rivers in Iran and South Korea. 6
Journal of Hydrology,572:470-485 7
[31] Sahoo BB, Jha R, Singh A, Kumar D (2019) Application of 8
Support Vector Regression for Modeling Low Flow Time 9
Series. Water Resources and Hydrologic Engineering, 10
23:923-934 11
[32] Shin S, Kyung D, Lee S, Taik& Kim J, Hyun J (2005) An 12
application of support vector machines in bankruptcy 13
prediction model. Expert Systems with Applications, 28: 14
127-135. 15
[33] Sivapragasam C, Muttil N (2005) Discharge rating curve 16
extension a new approach. Water ResourManag 19:505–520 17
[34] Tongal H, Booij M (2018) Simulation and forecasting of 18
streamflows using machine learning models coupled with 19
base flow separation. Journal of Hydrology, 564:266-282 20
[35] Vapnik VN (1998) Statistical Learning Theory. Wiley, New 21
York. 22
[36] Wang W, Men C, Lu W (2008) Online prediction model 23
based on support vector machine. Neurocomputing 71:550– 24
558 25
[37] Xu L, Wang J, Guan J, Huang F (2007) A Support Vector 26
Machine Model for Mapping of Lake Water Quality from 27
Remote-Sensed Images. IC-MED. Vol. 1(1), pp. 57-66. 28
[38] Yassen ZM, Awadh SM, Sharafati A, Shahid S (2018) 29
Complementary data-intelligence model for river flow 30
simulation. Journal of Hydrology, 567:190-190 31
[39] Yu PS, Chen ST, Chang IF (2006) Support vector regression 32
for realtime flood stage forecasting. J Hydrol 328:704–716 33

Application of Support Vector Machine for River flow Estimation

More Related Content

What's hot (19)

Similar to Application of Support Vector Machine for River flow Estimation (20)

More from AI Publications (20)

Recently uploaded (20)

Application of Support Vector Machine for River flow Estimation