1.

ABSTRACT
This paper proposes a new irrigation system using fuzzy logic technique by
mapping the knowledge and experience of a traditional farmer. Fuzzy logic control,
which is similar to the human way of thinking, has emerged as the most active tool in
automatic control. The purpose of fuzzy logic controller is to automatically achieve and
maintain some desired state of a system and process by monitoring system variables as
well as taking appropriate control action.
The aim of this work is to develop an intelligent control using fuzzy logic
approach for irrigation of agricultural field, which simulates or emulates the human
beings intelligence. The status of any agricultural field, in terms of evapotranspiration
and error may be assumed as input parameters and the decision is made to determine the
amount of water required for the area to be irrigated, well in advance. This leads to use
effective utilization of various resources like water and electricity and hence becomes a
cost effective system for the expected yield.

2. INTRODUCTION
From the past, agriculture has been playing an important role in human societies
to suffice the growing and dynamic demands. Irrigation is an essential component of crop
production in many areas of the world. Precision Agriculture(PA) is an integrated system
designed to increase long-term, field-specific, and farm production efficiencies,
productivity, and profitability in the field of agriculture. The PA is very essential for the
countries like India whose agriculture completely depends upon the rains and climatic
conditions. Precision farming ensures quicker response times, better quality control for
the yield with a minimum labor effort. There is a requirement for use of sensing
technologies in the field of PA to monitor the crop parameters and control the utilization
of resources towards the societal benefits
In the past few years, there has been an increasing interest in the application of the
fuzzy set theory to many control problems. For many complex control systems, the
construction of an ordinary model is difficult due to nonlinear and time varying nature of
the system. Fuzzy Control has been applied in traditional control systems, which yields
promising results, It is applied for the processes, which yields promising results, it is
applied for the processes, which are too complex to be analyzed by conventional
techniques or where the available information is uncertain. In fact, fuzzy logic controller
(FLC) is easier to prototype, simple to describe and verify, can be maintained and also
extended with grater accuracy in less time. These advantages make fuzzy logic
technology to be used for irrigation system also.

3. LITERATURE REVIEW
The most important finding in the literature is the unanimity on the need to have a
site specific focus to irrigation schemes and to ensure that the community is brought into
process from the start, with their priorities, in order to equip them and their elected
committee to manage the scheme once the department, or agency or donor withdraws
form the process. In word there must be revitalization which implies a move away from
pure infrastructure rehabilitation to a comprehensive programme to structure, train and
capacitate the smallholder farmers to run their scheme profitably and sustainably (de
Lange, 2004). The literature supports rehabilitation in the strongest terms warning of
failure if capacity building of the community is left out: The experience from the review
is explicitly clear that infrastructure development alone or as a dominant part of the
intervention is destined to failure. Farmers in smallholder schemes need support systems
that go far beyond just the irrigation system if they are to improve their livelihoods
significantly. Irrigation is a highly complex mix of social, agriculture, market and
technical parameters, which are in a state of on-going flux and interconnectedness.
Irrigation planners and advisors must, as a critical priority, embrace the multiple sectoral
interests and dynamics in planning thinking. Narrow isolated, engineering and
infrastructure driven programs are destined to fail in their objectives.

The growth processes of the crops are often effected by a lot of environmental
factors, including water deficit, temperature anomaly, disease insect damage and
disadvantageous soil condition etc..Effect in water deficit among them was most serious,
and exceeded the sum of the other environment affects.At the same time, water resource
saving status have already been the important index appraising a country or regional
economies sustained development. Study on water resource saving has been paid
attention to by the home and abroad scholars. It was shown in reference that agriculture
irrigation using water occupied 70% of the whole world fresh water, and or so 40% was
wasted owing to evaporation, deep sorption of soil etc..Therefore, precision irrigation
must be vigorously developed and promoted. Preliminary research results on fuzzy
control model of precision irrigation based on water stress monitoring for the corps were
designed in the paper. Five sensors were introduced and respectively monitored AE, the
temperature, humidity, illumination and the CO2 density. Self-learning fuzzy model on
precision irrigation was layouted. Present given volume on water was by five inputs. It
was shown that five inputs and signal output of double fuzzy control model on precision
irrigation system could effectively fulfill the tasks of normal irrigation and precision
irrigation, timely, suitably and scientifically irrigate under water required information for
the corps growth, so as to save water and expand productivity.
To a lesser extent, fuzzy logic applied to control is another discipline we explored.
First introduced by Zadeh in the early 1960s, this discipline has been widely used for
different applications. Our work extended the load-matching training procedure designed
for neural-network controllers to fuzzy-logic control. Therefore, the concept of
backpropgation is used here as well. Jang produced an important contribution related to
self-adapting, fuzzy-logic control systems. He developed the concept of adaptive
network-based fuzzy inference system, also know as ANFIS. Fuzzy-logic system
identification was part of his approach. The fuzzy-logic defuzzification used by ANFIS is
based on a zero-order Sugeno fuzzy model (or FIS, Fuzzy Inference System) . Along with
ANFIS, Jang introduced the concept of universal approximator and using the StoneWeierstrass Theorem he proved that when the number of rules is not restricted, a zeroorder Sugeno model can match any arbitrary nonlinear function. He also related the
Sugeno model with the Tsukamoto model. An important issue that relates the neural-

network world with fuzzy-logic models is the connection between FISs and Radial Basis
Function Networks (RBFN) . The latter are single layer neural networks with Gaussian
functions as squashing functions, with non-weighted input data. Kosko also describes the
basis of the universal approximator used in Sugeno models. The same issue is addressed
by Mendel .

4. CONCEPT
4.1 IRRIGATION CONTROLLER TYPES
Irrigation is the manual application of water for growing crops. In crop production
it is mainly used in dry areas and during shortage of rainfall to protect crops. There are
different types of irrigation such as Surface irrigation, Localized irrigation, Drip irrigation
and Sprinkler irrigation. In the field of agriculture, use of proper method of irrigation and
its control is important. Two general types of controllers are used to control irrigation
systems: Open control loop systems and Closed control loop systems. Open control loop
systems apply a preset action, such as is done with simple irrigation timers. Closed
control loops receive feedback from sensors, make decisions and apply the results of
these decisions to the irrigation system.
I. OPEN LOOP SYSTEM
In an open loop system, the operator makes the decision on the amount of water
that will be applied and when the irrigation event will occur. This information is
programmed into the controller and the water is applied according to the desired
schedule. Open loop control systems use either the irrigation duration or a specific
applied volume for control purposes. Open loop controllers normally have a clock that is

used to start irrigations. Termination of the irrigation can be based on a pre-set time or
may be based on a specified volume of water passing through a flow meter. Open loop
control systems are typically low in cost and readily available from a variety of vendors.
They vary in design and complexity and often offer flexibility as to the number of zones
and how irrigations are scheduled. The drawback of open loop systems is their inability to
respond automatically to changing conditions in the environment. In addition, they may
require frequent resetting to achieve high levels of irrigation efficiency.
II. CLOSED LOOP SYSTEM
In closed loop systems, the operator develops a general control strategy. Once the
general strategy is defined, the control system takes over and makes detailed decisions of
when to apply water and how much water to apply. This type of system requires feedback
from one or more sensors. Irrigation decisions are made and actions are carried out based
on data from sensors. In this type of system, the feedback and control of the system are
done continuously. Closed loop controllers require data acquisition of environmental
parameters (such as soil moisture, temperature, radiation, wind-speed, etc) as well as
system parameters (pressure, flow, etc.). The state of the system is compared against a
specific desired state, and a decision whether or not to initiate an action based on this
comparison. Closed loop controllers typically base their irrigation decisions on the
sensors that measure soil moisture, temperature, humidity and evaporation and other
climatic data to estimate water requirement of a crop .
4.2 IMPLEMENTATION OF SYSTEM HARDWARE
This section presents proposed Fuzzy based Irrigation Control System
architecture using WSN for monitoring and controlling the irrigation in an agriculture
which is as shown in Figure 1. It consists of four basic components namely [A] Wireless
Sensor Network [B] Gateway Node / Sink Node [C] Fuzzy based Irrigation Controller
[D] Irrigation Pipe Network. The first component consists of Wireless Sensor Network
which sense physical and environmental parameters and send data to the gateway node.
Second component is application server which receives data from gateway and processes

it. The last component is irrigation pipe network which is laid over the irrigated areas and
the electric control valves are installed on pipelines.
4.3 WIRELESS SENSOR NETWORK
The proposed system implemented using the MEMSIC eKo Pro Series which is a
wireless agricultural and environmental sensing system for crop monitoring. The system
also provides an easy deployment of wireless monitoring system in an agricultural layout
for efficient collection of data about its needs from multiple locations.
The eKo Node is a fully integrated, rugged outdoor sensor package that uses an
energy-efficient radio and sensors for extended battery-life and performance. The eKo
Node integrates MEMSICs IRIS processor/radio board and antenna that are powered by
rechargeable batteries and a solar cell. An eKo Node is capable of an outdoor radio range
up to 2 miles depending on the deployment and the hardware configuration chosen. The
nodes themselves form a wireless mesh network that can be used to extend the range of
coverage. By simply adding an additional eKo Node, it is easy to expand the coverage
area. The nodes come pre-programmed and configured with MEMSICs XMesh lowpower networking protocol. This provides plug-and-play network scalability for wireless
sensor networks. The system can be easily enhanced with various sensors such as soil
moisture, ambient temperature and humidity, leaf wetness, soil water content, solar
radiation, etc.

Figure1. Fuzzy Based Irrigation Controller System Architecture

4.4 NEED FOR MODERN IRRIGATION SYSTEM

Water and electricity should be optimally utilized in an agricultural like India. The
development in the filed of science and technology should be appropriately used in the
field of agriculture for better yields. Irrigation has traditionally resulted in excessive
labour and non-uniformity in water application across the filed. Hence, an automatic
irrigation system is required to reduce the labour cost and to give uniformity in water
application across the field.
4.5 PHYSIOLOGICAL PROCESSING
In the irrigation system, plant take-varying quantities of water at different stages
of plant growth. Unless adequate and timely supply of water is assured, the physiological
activities taking place within the plant are bound to be adversely affected, thereby
resulting in reduced yield of crop. The amount of water to be irrigated in an irrigation
schedule depends upon the evapotranspiration(ET) from adjacent soil and from plant
leaves at that specified time. The rate of ET of a given crop is influenced by its growth
stages, environmental conditions and crop management. The consumptive use or
evapotranspiration for a given crop at a given place may vary through out the day,

through out the month and through out the crop period. Values of daily consumptive use
or monthly consumptive use are determined for a given crop and at a given place. It also
varies from crop to crop. There are several climatological factors, which will influence
and decide the rate of evaporation. Some of the important factors of eliminate influencing
the evaporation are radiation, temperature, humidity and wind speed. In this work, the
input variables chosen for the system are evapotranspiration and rate of change of
evapotranspiration called as error and the output variable is water amount a shown in
fig.1

Evapo
transpiration
Error

FUZZY LOGIC
CONTROLLER
(PLC)
SYSTEM

Water
Amount

Fig.1 Schematic of irrigation system

4.6 IRRIGATION PARAMETERS FOR EFFICIENT SYSTEM OPERATION

To ensure proper design and operation of an irrigation system, the following
parameters should be considered.
i) Rooting Depth of the crop.
ii) Sand Texture & Water Storage Capacities of Soil
iii) Water Usage Capabilities based on the type of crop
iv) Field Soil Moisture Content
v) Leaf Wetness, Environment Temperature and Relative Humidity
4.7 FUZZY LOGIC THEORY

Fuzzy Logic Control (FLC) system is based on fuzzy set theory. This set theory is
advanced version of classical set theory called crisp theory. In crisp set theory, an element
either belongs to or does not belong to a set. But fuzzy set supports a flexible sense of
membership of elements to a set. Many degrees of membership, between 0 and 1, are
allowed. The membership function is associated with a fuzzy set in such a way that the
function maps every element of the universe of discourse or the reference set to the
interval [0, 1]. In crisp logic, the truth values acquired by propositions or predicates are
two-valued, namely TRUE or FALSE which may be treated numerically equivalent to (0,
1). However, in fuzzy logic, truth values are multi valued such as absolutely true, partly
true, absolutely false very true, and so on and are numerically equivalent to any value in
the range 0 to 1. In general, a Fuzzy Logic System (FLS) is a nonlinear mapping of an
input data vector into a scalar output. Figure 3 depicts a FLS that is widely used in fuzzy
logic controllers. A FLS maps crisp inputs into crisp outputs, and this mapping can be
expressed quantitatively as y = f(x). It contains four components: fuzzifier, fuzzzy rules,
inference engine, and defuzzifier.
4.8 STRUCTURE OF FUZZY CONTROLLER
Here, the basic internal structure of a fuzzy logic controller is presented. The FLC
allows one to use a control strategy expressed in the form of linguistic rules for the
definition of an automatic control strategy. A typical fuzzy logic controller can be
decomposed into four basic components as shown in Fig.2.

Fuzzy controller

Knowledge Base
(FAM)

Fuzzification

Inference Unit
State (fuzzy)

State (Crisp)

Defuzzification
Control (fuzzy)

Process

Control (Crisp)

Fig 2. Internal structure of fuzzy logic controller

I. FUZZIFICATION UNIT
It converts a crisp process state into a fuzzy state so that it is compatible with the
fuzzy set representation of the process required by the inference unit.

II. KNOWLEDGE BASE

The Knowledge base consists of two components. A rule base, which describes
the behaviour of control surfaces, which involves writing the rules that tie the input
values to the output model properties. Rule formation can be framed by discussing with
the experts. A database contains the definition of the fuzzy sets representing the linguistic
terms used in the rules. The knowledge base is generally represented by a fuzzy
associative memory.

III. INFERENCE UNIT

This unit is the core of the fuzzy controller. It generates fuzzy control actions
applying the rules in the knowledge base to the current process state. It determines the
degree to which each measured valued is a member of a given labeled group. A given
measurement can be classified simultaneously as belonging to several linguistic groups.
The degree of fulfillment (DOF) of each rule is determined by applying the rules of
Boolean algebra to each linguistic group that is part of the rule. This is done for all the
rules in the system. Finally the net control action is determined by weighting action
associated with each rule by degree of fulfillment.

IV. DEFUZZIFICATION UNIT

It converts the fuzzy control action generated by the inference unit into a crisp
value that can be used to drive the actuators. The defuzzification methods such as
centroid method, center of maxima method have been predominant on fuzzy control.
Perhaps the most frequently used defuzzification method is the centroid method.

5. FUNCTIONAL AND TECHNICAL DETAILS

DESIGN PROCEDURE FLC FOR IRRIGATION CONTROL

The heart of the FLC is to form the knowledge base that can obtained form human
experts is that field. In designing FLC, the following five steps are to be followed.
Step 1 : Identification and Declaration of Inputs and Output
This is the basic step in which the inputs and output are identified. In the
controller design for irrigation control, the inputs are evapotranspiration and error and the
output is water amount. The process of declaring the values of inputs and output called
universe of discourse is shown in table 1.

TABLE 1. Universe of discourse

Name
Evapotranspiration
Rate of change of
Evapotranspiration
Water Amount

Input/Output

Min value %

Max value %

Input

100

Input

-50

+50

Output

100

Step 2 : Identification of Control Surfaces

In this step, the linguistic variables are identified and membership values for each
linguistic variable are calculated. In this FLC, five Linguistic variables for
evapotranspiration, five Linguistic variables for error and nine linguistic variables for
water output are used. They are very Low(VL), Low(L), Medium(M), High(H) and Very
High(VH) for evapotranspiration: More Negative(MN),Negative(N), zero(Z), Positive(P)
and More Positive(MP) for error; Drastic Low (DL), Very Low(VL), Low(L),Medium
Low(ML),Medium(M), Medium High(MH), High(H), Very High(VH),Drastic High(DH)

for water output. The input and output variables are represented by fuzzy membership
functions as shown in Fig 3a, Fig 3b and Fig 3c.

Step 3: Behaviour of Control Surfaces

Fuzzy rules are constructed in specify action for different conditions, that is the
control rules the associate the fuzzy output to fuzzy inputs are derived from general
knowledge of system behaviour. In this method, the rules are extracted form numerical
data and then combined with linguistic information collected for experts. The rule bas for
the said application is shown in Table 2. The weightage take for rules involving zero error
is reduced to 0.25 for facilitating over correcting problems.

Step 4 : Decision Making Logic Of Inference Logic

It infers a system of rules through the fuzzy operator. In inference mechanism
PRODUCT implication is superior to MIN implication-minimum clipping. Further a
SUM combiner is better that MAX combiner for aggregation. In this work, SUM
PRODUCT criteria are used to determine the outcome of rules.
Step 5: Defuzzification
For any given crisp input value, there may be fuzzy membership in several
input variables, and each will cause several fuzzy outputs cells to fire or to be activated.

TABLE 2 RULE BASE MATRIX

EVAPO TRANSPIRATION
ERROR

VL

VH

MN

DL

VL

ML

VL

ML

MH

ML

MH

ML

MH

VH

MP

MH

VH

DH

This brings the process of defuzzification of output to crisp value. Centriod weightage
method is used for defuzzification.
RESULTS AND DISCUSSION
This work has been carried out using MATLAB simulation tool, The developed
software for the proposed work was tested under different input condition and provided

good results in terms of accuracy and has a wide scope of being established in near
future.
By applying the fuzzy logic system, the results which were already observed
(referred from IETE Technical Review) which shows that this method requires less
amount of water for the same yield when compared to the method followed by the
traditional farmer. The results tend to move smoothly across the control surfaces. Thus
the result shown above ensures the effectiveness and accuracy of our proposed system.

6. ADVANTAGES AND DISADVANTAGES

The use of fuzzy logic clearly enables a human being to interface easier with an
automated system than in the conventional case. This is because human beings more or
less have a natural tendency towards uncertainty. Advantages therefore may result in all
cases where human beings are involved with systems, be it as a designer or as a user.
When a human being is seen as a user, a more natural system interface can be obtained in
fuzzy systems. This is because the system can directly communicate with the user via
natural language terms. In the design of systems that are less soft, fuzzy logic can be of
assistance because of the fact that in the design of such systems often human knowledge
can or must be used. One can think of expert knowledge from humans that already are
able to perform tasks that must be automated, like for instance train control, mortgage
analysis or target tracking. One can also think of fuzzy knowledge of expert system
designers. Mostly, the tasks that can be performed with fuzzy logic can also be done in a
non fuzzy way. The key idea of using fuzzy logic however is that precision is expensive
while not always necessary. People for instance are quite good at performing several
decision tasks using only non precise data and generating non precise actions. One of the
key reasons why fuzzy logic works well is the fact that many systems do not require very
critical tuning. In other words, when parameters are set sub-optimal, the performance will
not degrade very much.
Summarizing, the following benefits can be named:
- Fuzzy Logic describes systems in terms of a combination of numerics and linguistics

(symbolics). This has advantages over pure mathematical (numerical) approaches or pure
symbolic approaches because very often system knowledge is available in such a
combination.
- Problems for which an exact mathematically precise description is lacking or is only
available for very restricted conditions can often be tackled by fuzzy logic, provided a
fuzzy model is present.
- Fuzzy logic sometimes uses only approximate data, so simple sensors can be used.
- The algorithms can be described with little data, so little memory is required.
- The algorithms are often quite understandable.
- Fuzzy algorithms are often robust, in the sense that they are not very sensitive to
changing environments and erroneous or forgotten rules.
-

The reasoning process is often simple, compared to computationally precise systems, so

computing power is saved This is a very interesting feature, especially in real time
systems.
- Fuzzy methods usually have a shorter development time than conventional methods.
Although the above named advantages are very promising, one must be aware that fuzzy
logic does not fit to every problem. The following remarks must be made:
- As has been shown in section 3, fuzzy logic amounts to function approximation in the
case of Crisp-Input/Crisp-Output systems. This means that in many cases, using fuzzy
logic is just a different way of performing interposition In the light of the fact that system
knowledge is often available as a combination of numerics (quantitative) and linguistics
(quantitative or qualitative) this approach may even be advantageous.
- In areas that have good mathematical descriptions and solutions, the use of fuzzy logic
most often may be sensible when computing power (i.e. time and memory) restrictions
are too severe for a complete mathematical implementation.
- I am convinced that results obtained in successful fuzzy application,: that are given in
literature can be reached with a conventional approach as well, possibly taking longer
development time and possibly with the use of different interpolation methods. Careful
analysis of comparison examples, 'proving' the superiority of fuzzy logic often shows that
they compare the fuzz) approach with a very simple, non-optimized conventional
approach.

- Proof of characteristics of fuzzy systems is difficult or impossible in most cases because

of lacking mathematical descriptions; especially in the area of stability of control systems
this is an important research item. On the other hand, when solving practical problems,
this is often not a very severe restriction because when the system is tested the
characteristics will also be found.

7. APPLICATIONS
The applications of fuzzy logic, once thought to be an obscure mathematical
curiosity, can be found in many engineering and scientific works. Fuzzy logic has been
used in numerous applications such as facial pattern recognition, air conditioners,
washing machines, vacuum cleaners, antiskid braking systems, transmission systems,
control of subway systems and unmanned helicopters, knowledge-based systems for
multi objective optimization of power systems, weather forecasting systems, models for
new product pricing or project risk assessment, medical diagnosis and treatment plans,
and stock trading. Fuzzy logic has been successfully used in numerous fields such as
control systems engineering, image processing, power engineering, industrial automation,
robotics, consumer electronics, and optimization. This branch of mathematics has
instilled new life into scientific fields that have been dormant for a long time. Fuzzy logic
is extremely useful for many people involved in research and development including
engineers (electrical, mechanical, civil, chemical, aerospace, agricultural, biomedical,
computer, environmental, geological, industrial, and mechatronics), mathematicians,
computer software developers and researchers, natural scientists (biology, chemistry,
earth science, and physics), medical researchers, social scientists (economics,
management, political science, and psychology), public policy analysts, business
analysts, and jurists.

8. FUTURE SCOPE
Crop irrigation control is the most important concern in the domain of agriculture.
Shortage of water globally is also emphasizing the need of systems that not only control
the crop irrigation but also provide the intelligent way to provide water to only those
places where it is needed and in the required quantity. By monitoring soil moisture, Leaf
Wetness, Temperature and Relative Humidity and by taking into consideration the other
parameters like plant root depth , Sand Texture & Water Storage Capacities of Soil, Plant
Water Use Capabilities one can make efficient use of water resources and also in
achieving high yield. The work presented here brings out the potential advantages of
applying FLC technique for Irrigation System. The simulation result provides an exact
idea for water output for the prescribed agricultural field.

9. CONCLUSION
Based on crop growth by water stress, diseases such as forced, the characteristics
of environmental factors, a crop of conventional precision irrigation two-mode fuzzy
control model was designed in this paper. In order to overcome subjectivity regulations
on control the influence on the quality of fuzzy control model, the self learning function
was introduced in the structural design, a suitable for crop growth self-learning fuzzy
control algorithm was put forward, and a crop precise irrigation self-learning fuzzy
control model was established, and makes fuzzy control system has the self-perfection
sex. So the system as the work of change amendment rule to adapt the practical situation.
Simulation results show that this control strategy for overcoming the crops of fuzzy
control precision irrigation system exists when the normal amount of irrigation water
waste and precisely when the irrigation low efficiency, give water too much, can in the
normal amount of irrigation take safety and energy saving, precise irrigation take the
safety and efficiency for crops, precision irrigation intelligent control provides a control
strategies and methods.

10. REFERENCES
[1] Domingo Gmez-Melendez, Fuzzy irrigation greenhouse control system based on a
field programmable gate array.
[2] Edward C. Martin, Methods of Determining When to Irrigate Cooperative Extension,
College of Agriculture & Life Sciences, The University of Arizona .
[3] Environmental Optimization Of Irrigation Management With The Combined Use And
Integration Of High Precision Satellite Data, Advanced Modeling, Process Control And
Business Innovation .
[3] Hal Werner, Extension irrigation engineer, Measuring Soil Moisture for Irrigation
Water Management, Cooperative Extension Service .
[4]

http://www.memsic.com/products/wireless-sensornetworks.html

(Environmental

Monitoring System manual)

[5] Jumman A and Lecler NL,"A continuous soil water potential measurement System for
irrigation scheduling assessment" ,Proceedings of South African Sugarcane Technology
Association .
[6] M. Cogan. "Water measurement in soil and vines".