by PSP Volume 25 – No.

1/ 2016, pages 325-336 Fresenius Environmental Bulletin

 

IRRIGATION OPTIMIZATION BY THE USE OF FUZZY LOGIC

TECHNOLOGY
1
Hussein Bizimana, 2Muhammad Yaqub, 3Osman Sönmez, 4Beytullah Eren, 5Fatma Demir
1, 4
Sakarya University, Environmental Engineering, 54187 Serdivan/Sakarya, Turkey
2, 3, 5
Sakarya University, Civil Engineering, 54187 Serdivan/Sakarya, Turkey

ABSTRACT Irrigation has seen its roots in the history of

mankind since earliest beginning. Irrigation helps to
The aim of this study was to develop a Fuzzy reduce the uncertainties, particularly the climatic
Inference System (FIS) to optimize the water uncertainties in agriculture practices [2].
requirements as per standards for irrigation
purposes.There are three major inputs in FIS For example equatorial part of the world
including water quality, field capacity and water which mostly was used to have high rainfall
quantity while one output named as optimized intensities has a mean rainfall of 900 mm a year and
irrigation. Firstly water quality FIS was developed according to projections rainfall would be of 630
by using three important such as electrical mm a year which may speed up the expansion of
conductivity (EC), pH and sodium absorption ratio dry area. In this regard, many plants must adapt
(SAR) with excellent, good, permissible and themselves to new climate conditions and improve
harmful membership functions for inputs and same their management strategies of available water
for output. To create fuzzy inference system for resource; highly water optimizing irrigation
water quantity, five membership functions were methods are required [2].
developed as very low, low, middle, high and very
high considering the discharge of water available in Therefore, it is necessary to implement
the Sakarya River, Turkey and low, medium and efficient irrigation systems by using strategies of
high as fuzzy membership functions for field water resource to assure food safety in regions and
capacity were also developed. To complete the FIS environmental sustainability [3]. With other
of the study, very low, low, medium, high and very strategies, approximations to modeling studies
high fuzzy membership functions were developed happened to required and require nowadays new
for a fuzzy optimized irrigation as the output of the abstractions allowing improving our nature
system and the main finding of the study. In this predictive skills and consequently striving for a
study a practical solution based on FIS is presented sustainable development [4]. The application of
for Sakarya River Basin in Turkey to meet system known as intelligent systems in Agricultural
irrigation objectives locally. The findings clearly Sciences studies is relatively recent in comparison
indicate that fuzzy logic method may successfully to other knowledge areas like electronic
interpret complex conditionsto classify the engineering, electric engineering, mechatronics and
optimized irrigation with respect to water quality, telecommunications.
water quantity discharge and soil filed capacity.
The developed fuzzy model was validated by using Huang [5] insured that intelligent systems are
data collected from Pamukova, Doğançay and an integration of biological structures with
Adatepe three monitoring stations of Sakarya River computing techniques. In agricultural sciences
during 1993-2013. worldwide fuzzy logic is used for modeling and
prediction, control, classification, fuzzy clustering,
KEYWORDS: rule based inference, multisensory data fusion,
Electrical conductivity (EC), Fuzzy logic, Field capacity, optimization [6]. Finally, this work aimed to
Irrigation, pH (Hydrogen proton concentration), SAR implement fuzzy logic system as a tool to evaluate
(Sodium Absorption Ratio), Optimization. the optimization of irrigation systems from a set of
indicators for efficient use of water.

INTRODUCTION
MATERIAL AND METHODS
Irrigation is defined as replenishment of soil
water storage in plant root zone using methods This research was carried out by students from
other than natural methods such as precipitation [1]. two departments, civil and environmental

325

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

engineering departments. To be able to implement Water quantity discharge. Without a certain

the projects, some important indexes in irrigation source of water irrigation can’t be performed and if
such as: Water Quality, Field capacity, Water the source is with seasonal regime (means available
Quantity Discharge were studied and used for the only for some time) also irrigation performance
simulated project. through a year is very critical [10].It means the
source of water for irrigation is very important for
Water quality. Both irrigation water quality plants growth, their production and sustainable use
and appropriate irrigation management are of water.
significant factors to maximize crops production.
The quality of irrigation water may influence soil Design of fuzzy inference system (fis). In
physical conditions as well as crops yield, although case of imprecise, vague, incomplete, ambiguous or
all other parameters and practices are optimal. noisy input information of water sources where by
According to American Public Health Association, global climate change nowadays it is very hard to
irrigation water characteristics differ with the predict with exactitude about rainfall, hence FIS is a
source of the water [7]. Therefore, testing of tool that allows arriving at a conclusion in a simple
irrigation water is a critical and important parameter way, trying to simulate expert knowledge of a
that must be consider before the selection of crop human being. In this work three variables such as
type and soil site [8]. The quality of some water water quality, field capacity and water quantity
sources may change significantly with time or were considered to develop FIS. Firstly water
during certain periods such as in dry or rainy quality was analyzed by an environmental
seasons [9] so it is recommended to take more than engineering team, keeping pH, Electrical
one sample and must be analyzed, in different time Conductivity (EC) and Sodium Absorption Ratio
periods. The parameters which determine the (SAR) as inputs.
irrigation water quality are divided in to three After the simulation and set of membership
categories: chemical, physical and biological. In function for water quality, it was used as input
this study, chemical properties of the irrigation along with field capacity and water quantity for the
water are discussed. The chemical characteristics of final simulation to optimize irrigation system.
irrigation water refer to the content of salts in the Membership functions of different inputs were
water as well as to parameters derived from the fixed in FIS according to literature review and
composition of those salts in the water parameters interview with field experts and their experiences in
such as EC/TDS (Electrical Conductivity/ Total civil engineering based project. For this purpose,
Dissolved Solids), SAR (Sodium Adsorption Ratio) Fuzzy Logic toolbox in Matlab was used to
alkalinity i.e pH and hardness. construct FIS, wheremembership’s ranges, active
rules functions are chosen; likewise, response
Field capacity. Field Capacity (FC) is the functions are analyzed and the system is verified.
amount of water in the soil layer that can be easily
used by the plant for its growth different from Design steps. A Mamdani-type fuzzy logic
gravitational water and wilting point[3]. FC for system was designed based on the tool Fuzzy Logic
irrigation is an important factor providing an idea Toolbox of Matlab to evaluate the optimization of
about water requirement of soil in order to avoid the irrigation system according to the indexes of
over irrigation which may cause leaching and outputs as Water Quality (WQ), Water Quantity
lodging of soil layers [5]. (WaQ) discharge and Field Capacity (FC) and the
final output of the system as optimized irrigation.

326

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

FIGURE 1
General view of the final simulation

FUZZY SIMULATION OF SYS. INPUTS Control Regulation (WPCR) Turkey provided the
standards for irrigation water quality, in which
Water Quality (WQ) simulation with fuzzy permissible limits for SAR, EC, pH and other
logic. All the studies pertaining to assess the quality elements are specified, and from those acceptable
of water for irrigation consider two parameters, limits it helps to create the membership functions
namely Sodium Absorption Ratio (SAR) and for our simulated Water Quality system in fuzzy
Electrical conductivity (EC). In this project we logic and at the end we were able to use specified
considered pH as a third variable to optimize the membership functions in the final fuzzy simulation
irrigation more accurately. The Water Pollution which will provide us optimized irrigation.

FIGURE 2
General view of a simulation for Water Quality

The membership functions are created on the different parameters that are important to control
basis of already done research work in this area and the irrigation water quality and vary from place to
by consultation with our teachers. There are place so, it is difficult to develop a standard model

327

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

that will be suitable for all water sources. In such main problem related to irrigation water quality
circumstances it is the need of hour to develop the which refers to the total amount of dissolved salts
various models by keeping all the factors in view but does not indicate which salts are present. High
and fuzzy logic provide a good solution in such level of salts in the irrigation water reduces water
situations. availability to the crops and causes yield reduction.
Above a certain threshold, reduction in crop yield is
Electrical conductivity (EC). Salinity is a proportional to the increase in salinity level.
measure of the Total amount of salts in the water is
known as salinity and if high than permissible range The salinity effects are generally evidenced by
causes to reduce water availability to the crop and reducing transpiration rates and proportionally
affect the yield. Electrical conductivity (EC) or related growth, and producing smaller plants. The
Total Dissolved Solids (TDS) tests are two means salinity also alters the soil ability to retain nutrients,
of measuring salinity and EC is a helpful and thus suppressing plant growth [12].Different crops
reliable tool for the measurement of water salinity vary in their tolerance to salinity and therefore have
or TDS. The total amount of TDS should be used different thresholds and yield reduction rates [13].
together with SAR.The properties of the soil, the The most common parameters used for determining
ground water and the landscape interact with the the irrigation water quality, in relation with its
salinity of the irrigation water to either increase or salinity, are EC and TDS. In this study conductivity
decrease the salinity hazard [11].The salinity is a ranges from 479- 1208 µmho/cm.

FIGURE 3
Membership functions selected for Electrical Conductivity (EC) simulation.

pH. It is an index of hydrogen ions dissolved form and make them more readily
concentration (H+) in water and defined as –Log available. In irrigation process normal pH range is
(H+) and ranges from 0 to 14 where, lower than 7 is from 6.5 to 8.4. The pH ranges of obtained data
considered to be acidic and higher than 7 is from Sakarya River are 7.61-8.63. Unusually low
considered to be basic while pH of 7.0 is considered pH may cause acceleration of corrosion in irrigation
to be neutral. In irrigation water quality could be systems while pH above 8.5, causes to form
basic that mainly affects the availability of nutrients insoluble salts of calcium and magnesium, leaving
while acidic water is also important to know how sodium as leading ion in the solution. Alkaline
quickly irrigation is required, to keep the metals in water intensifies solidity of soil [14].

328

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

FIGURE 4
Membership functions selected for pH simulation.

Alkalinity is the measure of a water’s capacity

to neutralize acids (consume H+ ions), is the sum of Direct effect of salts on plant growth may be
the amounts of bicarbonates (HCO3-), carbonates classified into three categories: (i) a decrease in
(CO32-) and hydroxide (OH-) in water and is osmotic potential of the soil solution causes to
expressed as mg/l or meq/l CaCO3. Alkalinity reduces available water to plants, (ii) a deterioration
buffers the water against sudden changes in pH and in physical structure of soil happens due to which
if the alkalinity is too low, then any addition of water permeability and soil aeration are decreased,
acidic fertilizers will immediately lower the pH. and (iii) an increase in concentration of certain ions
The main characteristic of irrigation water system is that have an inhibitory effect on plant metabolism
alkalinity as it affects the pH of root medium not [18]. Salinity is predicted from a recurrent
the water. measurement of total dissolved solids (TDS). TDS
is a measure of salts present in water, counting
Sodium Absorption Ratio (SAR). Sodium is organic and very small suspended particles [19].
adsorbed and becomes attached to soil particles
causes to harden and compact soil structure and Irrigation water whether of good quality or
increasingly impervious to water penetration. The not, can have effect on plant growth, for example
soil salinity is a severe factor that causes various poor irrigation water quality with excess salt can
abnormal, physio-morphological and biochemical damage plants in various ways.Water with high
changes, delay germination, high seedling amount of salts can hinder the conversion of
mortality, poor crop stand, stunted growth [15] as ammonium salts to nitrate by nitrifying micro-
literature stated crops production is seriously organisms in soil when used for
affected by soil salinity [16].The breakage of soil irrigation.Furthermore, most of tomato plants are
structure is commonly caused by continuous more sensitive to salt during seed germination,
application of high SAR water in irrigation because seedling growth and when flowering or fruiting.
salts effecting osmotic relationship between the root The seed and seedling stages are vulnerable not
and the soil moisture [17].The salinity level is the only because the plant structures are immature and
primary water quality parameter mostly applicable delicate, but also because tiny roots system draw
in irrigation systems because the soil structure and moisture and nutrients from the soil surface where
crop yield is affected by the salts quantity. salts tend to concentrate [20].

329

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

FIGURE 5
Membership functions selected for Sodium Absorption Ration (SAR) simulation.

The sodium problem is reduced if the amount SAR was calculated from the data obtained from
of calcium plus magnesium is high as compare to Sakarya River and found it ranges 2.35-14.50 in
sodium. This relation is called the sodium this study.
adsorption ratio (SAR) and it is a calculated value The combination of EC,pH and SAR as inputs in
from the formula: Mamdani fuzzy inference system basis on literature
�� reviews, international and local standards, expert’s
� = advices and was found important parameters to
��+��
√ assess the output water quantity with accuracy.
2

330

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

FIGURE 6
Water Quality Fuzzy system rules.

Field Capacity (FC) simulation with fuzzy to over irrigate which may cause leaching and
logic. Field Capacity (FC) is the amount of water in lodging of soil layers [21]. It is measured in height
the soil layer that can be easily used by the plant for inch per ha, where height inch times hectare that is
its growth different from gravitational water and 100 m3. Hence in the simulation in order to keep the
wilting point which are the amounts of water in the use of SI units a range from 0 to 300 cubic meters
soil layer but which can’t be used by the plant for was used as FC is ranging from 0 to 300 cubic
its growth [21]. The FC for irrigation is very meters in crisp sets and correspond to low, middle
important because it is the one giving the idea of and high in fuzzy sets.
how much water to be added to the soil in order not

331

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

FIGURE 7
Membership functions selected for FC


Water quantity (discharge) simulation with be with same membership functions too .With river
fuzzy logic. This index gives the information on the engineering discharge quantifyingmethods; water
amount of water available from the source that will which is not flooding the upland of the channel
be used to irrigate. If the amount is very low, low, have to be between 0-500 m³/s.
medium, high or very high then the irrigation will

FIGURE 8
Fuzzy simulated Water Quantity

RESULTS AND DISCUSSION showed precise and reliable resultsand those results
will be helpful in decision making. The output of
MATLAB Fuzzy Inference System (FIS) was the system was the optimized irrigation, the
applied in present study to optimize irrigation by membership functions were defined by the help of
using three inputs and one output variable. literature review and experts advices in the field. A
Different membership functions were given to FIS set of 53 system rules and surface view were
for evaluation of results.FIS Tool of MATLAB presented. The fuzzy simulated irrigation water

332

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

(Fig.9), FIS rules (Fig.10 and Fig.11) and the FIS rules surface review (Fig.12) are shown below.

FIGURE 9
Fuzzy simulated irrigation water.

FIGURE 10
FIS rules.

333

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

FIGURE 11
FIS and rules and output minimums

FIGURE 12
FIS rules on a surface view

DISCUSSION In the present study, a Fuzzy Inference System

(FIS) has been developed to classify the water for
An optimized irrigation system is proposed, irrigation purposes. The membership functions
taking into consideration indicators of optimization were developed for three significant parameters
use of water resource and tools of intelligent such as SAR, EC and pH and rules were fired to
systems like fuzzy logic. The results of this study arrive at a Fuzzy Inference System for assessing
showed that fuzzy set theory is more reliable and irrigation water quality. In last studies initially two
flexible option in irrigation optimization as parameters were used to develop the fuzzy logic but
compare to CROPWAT 8.0 and SAPWAT 8.1 in this work it was further improved by
because it deals with different hydrological incorporating one more significant parameter
uncertainties. It is helpful in decision making namely pH. Thus, the developed fuzzy model
regarding irrigation optimization in vague showed more reliable results for irrigation water
conditions without relying on Evapotranspiration classification and avoided the vagueness in the
equations. decision making.

334

by PSP Volume 25 – No. 1/ 2016, pages 325-336 Fresenius Environmental Bulletin
 

The set of indicators proposed is a basis to compare [9] Islam, M. S. and S.Z.K.M.
the optimization and fuzzy simulation of the Shamsad,(2009).Assessment of irrigation
irrigation systems which can help in the water qualityof Bogra District in
identification of infrastructure which can be Bangladesh.Bangladesh J. Agril.Res.34(4)
replaced by the use of fuzzy controllers in irrigation 597-608.
and drainage control, detection of needs for [10] Murray, F. W., “On the computation of
technology transfer in research and the saturation vapor pressure”, J. Appl. Meteor. 6:
improvement of the system operation. 203-204, 1967.
This system of fuzzy logic was implemented in [11] Peterson, H.G., (1999). Water Quality and
order to estimate the optimization of an irrigation Micro-irrigation for Horticulture
system based on the index of output per available http://www.safewater.org/PDFS/resourceswat
Water Quality, Field Capacity and available water erqualityinfo/Resource_Water_Qual_Facts.pdf
supply. This type of tool implemented in the [12] Guodie, A. S., (1992).the human impact:man’s
irrigation sector and specifically in the evaluation of Role in environmental change –black well
optimization of irrigation systems are innovative publishing press,Oxford, England
and could be helpful in decision making for [13] Bauder, T. A., R. M. Waskom and J. G. Davis
agricultural professionals. (2007), Irrigation water quality criteria.
Colorado State University, Us Department of
Agriculture. Research Report, 7/03.
REFERENCES [14] Mass., (1990).Crop salt tolerance. In:
Agricultural Assessment and Management
[1] Allen, R.G., Pereira, L.S., Raes, D., and Manual. K.K. Tanji (ed.). ASCE, New York.
Smith, M, “Crop evapotranspiration: pp. 262-304. pH and Alkalinity
Guidelines for computing crop requirements”, [15] Ahmad, M.S.A., Ashrafm, M., Ali, Q., 2010.
Food and Agriculture Organization (FAO), Soil salinity as a selection pressure is a key
Rome, Italy, 1998. determinant for the evolution of salt tolerance
[2] Ruiz, J. Gutiérrez, and J. Fernández, “A fuzzy in blue panicgrass (Panicumantidotale Retz.).
controller with an optimized defuzzification Flora 205, 37–45.
algorithm,” IEEE Micro, pp. 1–10, Dec. 1995 [16] Munns, R., 2002. Comparative physiology of
[3] Burman, R. and L.O. Pochop, 2004. salt and water stress. Plant Cell Environ. 25,
Evaporation evapotranspiration and climatic 239–250.
data. Elsevier, Amsterdam. [17] Malash, N., T.J Flowers and R. Ragab.,
[4] Ioslovich, I. P. Gutman and I. Seginer, 2006. (2005). The Effect of irrigation systems and
A non linear optimal greenhouse control water management practies using saline and
problem with heating and ventilation. Optimal non-saline water on tomato production. Agric.
Control Applications and Methods, 17: 157- Water. Manage, 78:25-38. Department of
169. Agriculture Engineering, University of Cape
[5] P. Javadi Kia, A. Tabatabaee Far, M. Omid, R. Coast, Ghana.
Alimardani and L. Naderloo, “Intelligent [18] Yeo, A.R.: Molecular biology of salt tolerance
Control Based Fuzzy Logic for Automation of in the context of whole plant physiology. J.
Greenhouse Irrigation System and Evaluation Exp. Bot., 49, 915-929 (1998).
in Relation to Conventional Systems” World [19] Grattan, S.R. and C.M. Grieve: Salinity-
Applied Sciences Journal 6 (1): pp. 16-23, mineral nutrient relations in horticultural
2009. crops. Sci. Hort., 78, 127-157 (1999).
[6] Harrison, L. P., “Fundamental‟s concepts and [20] Podmore, C., 2009. Irrigation salinity – causes
definitions relating to humidity”, In Wexler, A and impacts. PRIMEFACT for profitable,
(Editor) Humidity and moisture, Vol. 3, adaptive and sustainable primary industries-
Reinhold Publishing Co., New York, 1963. 937.October.
[7] APHA., (2005).Standard methods for [21] Jensen, M.E., R.D. Burman and R.G. Allen.
examination of water and wastewater. 21st ed. 1990. Evaporation and Irrigation Water
American Public Health Association, Requirements. ASCE Practice No. 70. ASCE,
Washington DC,USA NY, NY.
[8] Shahimasi, E and V.Kashuta.,
(2008).Irrigation water quality and its effects
upon soil.Tirana Agricultural
University,Tirana,Albania BALWOIS 2008- Received: 01.07.2015
ohrid,Republic of Macedonia Accepted: 22.12.2015

335
