118
Simulating of Changes in Water Distribution Uniformity Coefficient in Classic
Stationary Sprinkler Irrigation Using Data-Mining Models
Fariborz Ahmadzadeh Kaleybar
1*
| Shahram Shahmohammadi Kalalagh
2
| Sina Fard Moradinia
3
1
Department of Water Sciences and Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran - Sustainable
Development Management Research Center of Urmia Lake and Aras River Basin, Tabriz Branch, Islamic Azad
University, Tabriz, Iran
2
Department of Water Sciences and Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran
3
Department of Civil Engineering, Tabriz Branch, Islamic Azad University, Tabriz, Iran
Article Info ABSTRACT
Article type:
Research Article
Article history:
Received 04 Agust
2024
Received in revised
form 31 October 2024
Accepted 10 Novamber
2024
Published online 29
January 2024
Keywords:
Gene expression
programming
Malekan plain
Sensitivity analysis
Support vector machine
Performanceevaluation
Objective: The lack of water in the world and its heavy consumption in the crop irrigation sector
makes it necessary to use different sciences to increase water efficiency. The coefficient of
uniform water distribution in sprinkler irrigation systems is one of the important effective
indicators in evaluating their performance. Only high values can justify implementing these
systems. The purpose of this research is to use support vector machine (SVM) and gene
expression programming (GEP) models to simulate the coefficient of uniform water distribution
in the farm conditions of Malekan Plain in the northwest of Iran, placed in the catchment area of
the Urmia lake is experiencing severe water stress.
Methods: Field experiments were carried out on seven farms equipped with a classic stationary
sprinkler irrigation system with a movable sprinkler (Komet 162, 163) with variables of sprinkler
intervals on laterals and manifolds, operating pressure, and wind speed. Then, uniform
distribution coefficient data were obtained. Two models (SVM) and (GEP) were used to simulate
the value of the uniformity coefficient. The sensitivity analysis showed that all three variables
should be selected as model inputs. 70% and 30% of data were considered for the share of
training and test processes, respectively. Using these data, the adjustment parameters of each
model were calculated to reach the most optimal output. The evaluation of the performance of
the models was done with four indicators: RMSE (sum of square mean error), MAE (mean
absolute error), R2 (explanation coefficient), and DDR (developed difference ratio).
Results: The first rank of simulation accuracy was assigned to the GEP model. The values of
the indicators (RMSE, MAE, R
2
) were obtained in the training and test steps, respectively
(3.5087, 2.6827, 0.8634) and (1.1787, 0.9494, 0.9833) for GEP. The values of the evaluation
indices (RMSE, MAE, R
2
) for the most optimal SVM model in the test and training steps were
obtained (4.8917, 4.2704, 0.7884) and (2.6790, 2.4113, 0.9185) respectively. In the training step,
the value of CU(DDR(max)) for GEP and SVM models was calculated as 7.0540 and 5.2925
respectively. The value of this index in the test step for these two models was 20.8355 and 9.2863
respectively. The comparison of the value of this index also showed that the GEP model was
more accurate than the SVM model. In general, both models can simulate the amount of water
distribution uniformity in sprinkler irrigation in field conditions. However, using the GEP model
will lead to better results.
*
Corresponding author, Email: f.ahmadzadeh@iaut.ac.ir
Cite this article: Ahmadzadeh Kaleybar, F., Shahmohammadi Kalalagh, Sh., Fard Moradinia, S (2024). Simulating of
Changes in Water Distribution Uniformity Coefficient in Classic Stationary Sprinkler Irrigation Using Data-
Mining Models Journal of New Approaches in Water Engineering and Environment ,
© The Author(s). Publisher: Gonbad Kavous University.
DOI: http//doi.org/10.22034/nawee.2024.471274.1100