See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/317633874

A Smart IoT Fuzzy Irrigation System

Article in IOSR Journal of Engineering · June 2017

DOI: 10.9790/3021-0706011521

CITATIONS READS

33 6,361

3 authors, including:

George Kokkonis S. Kontogiannis

International Hellenic University University of Ioannina
73 PUBLICATIONS 633 CITATIONS 126 PUBLICATIONS 964 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by George Kokkonis on 16 October 2017.

The user has requested enhancement of the downloaded file.

IOSR Journal of Engineering (IOSRJEN) www.iosrjen.org
ISSN (e): 2250-3021, ISSN (p): 2278-8719
Vol. 07, Issue 06 (June. 2017), ||V1|| PP 15-21

A Smart IoT Fuzzy Irrigation System

George Kokkonis1, Sotirios Kontogiannis2, Dimitrios Tomtsis1
1
Dept. of Business administration, Western Macedonia University of Applied Studies, Greece,
(gkokkonis,tomtsis)@teiwm.gr
2
Dept. of Mathematics, University of Ioannina, Greece, [email protected]

Abstract: This paper proposes a novel fuzzy computational algorithm for IoT smart irrigation
systems. It describes all the possible sensors, actuators and microcontrollers that could be used in the
irrigation systems. Based on these sensors a novel system architecture is presented. The irrigation
system continuously monitoring air temperature and humidity, and ground moisture. The ground
humidity sensors are interspersed all over the field. The measurements are send in a microcontroller
that applies a fuzzy computational algorithm and decides whether to open a servo valve or not. All the
data collected from the microcontroller are sent to a cloud database for statistical information and
processing.

Keywords: Iot, irrigation systems, fuzzy algorithms, precision farming.

I. INTRODUCTION
Water scarcity is on one of the major concerns of today‟s agriculture. Efficient irrigation of agricultural
land could be accomplished with the help of smart irrigation systems. Smart irrigation systems have as primary
goal the minimization of water consumption and the maximization of quantity and the quality of the crops. A
trade of between these three amounts could be made [1]. Another concern that irrigation systems have is the
power consumption of the system. As most of agricultural land is located in isolated places without stable
electrical powerline the power of the system is an important factor. Smart irrigation systems use actuators and
sensors to control the ground moisture of the agricultural land. The data collected from the sensors could be
send through the Internet of Things (IoT) to data servers for statistical information and processing. On this basis
we propose our smart - fuzzy irrigation system of arable areas. With the help of multiple soil moisture sensors,
effective irrigation of agricultural land that increase production and minimize water consumption is achieved. A
fuzzy computational algorithm is proposed that determines the amount of the opening of the central servo valve
of the irrigation system. This fuzzy algorithm has as input interspersed in the field multiple soil moisture
sensors. Soil moisture is constantly monitored. The data collected from the sensors is sent to server for statistical
analyses. The rest of the paper is organized as follows. Section 2 outlines the related work on smart irrigation
systems. Section 3 presents the wireless technologies that are used in IoT irrigation systems. Section 4 describes
the sensors and the actuators that our irrigation system uses. Section 5 presents the proposed system architecture
of our system. Section 6 analyzes the proposed fuzzy watering algorithm . Section 7 identifies conclusions and
future work.

II. RELATED WORK

A smart IoT fuzzy irrigation system is defined as a system that controls field irrigation without user
interference. The decision for watering is based on data that are collected through the Internet of Things (IoT)
from weather websites and with the help of sensors and actuators. Based on the process for the watering
decision the irrigation systems are divided into the IoT weather based irrigation systems and the sensor based
irrigation systems. In the weather based irrigation systems the decision for irrigation is taken based on future
weather conditions that are collect through the Internet. The system gathers data for temperature, rain, air
humidity, [2]. Based on the data that are gathered the decision for irrigation is made. The system monitor the air
temperature in order to avoid irrigation in extreme weather conditions, as if the temperature drops down 0 oC, so
as to prevent the plants from freezing. Furthermore the system monitors the rain fall in the close future in order
to avoid irrigation and save water if rain is about to start. Moreover, humidity and wind measurements are
needed to estimate the dehydration of the plants [3]. The system accesses the Internet through 4G/3G/GPRS
gateways and it collects the data from weather websites [4]. All the data gathered from the websites are stored in
databases for future processing. Smart phones play a major role in these systems as the farmer is informed
through its phone for notifications and alerts from the system[5].

International organization of Scientific Research 15 | P a g e

 Paper preparation guidelines for IOSR Journal of Engineering

Another system architecture for smart irrigation systems is based on sensors that are placed in the field.
They monitor soil moisture with the help of ground humidity sensors [6] and try to decide if the field needs
irrigation or not. This type of irrigation systems outperform the weatherbased systems as the measure with
precision the humidity of the field. The only disadvantage they have is that they cannot predict the weather in
the close future. A hydride system that combines information for future weather conditions for website and close
field environmental conditions from local sensors is a good solution for the proper irrigation of the field with as
watering saving as possible [7]. Some smart irrigations systems have already been proposed. A smart irrigation
system that uses a soil sensor and a central controller is offered by Koubachi company. A self-learning watering
algorithm with statically assigned weights is used in [8]. An arduino based smart irrigation system was
presented by Parameswaran and Sivaprasath. It monitors the PH, the temperature and the moisture levels in the
field [9]. An Automation of Irrigation System based on Wi-Fi Technology and IOT was presented by Kranthi
Kumar and Srenivasa [10]. Harun et al. proposed a IoT precision irrigation system for green houses with the
help of a wireless sensor network [11].
Rajalakshmi and Mahalakshmi developed a system to monitor crop-field using sensors (soil moisture,
temperature, humidity, Light) and automate the irrigation system. They used a NRF24L01 module for wireless
data transmission between the field and the web server. They choose NRF24L01 because of the low power
consumption. NRF24L01 requires 1.9-3.6V and uses 2.4GHz [23]. On the other hand Srishti Rawal proposes an
automated irrigation system which monitors and maintains the desired soil moisture content via automatic
watering. He choose GSM-GPRS SIM900A modem to update regularly a webpage through which a farmer can
check whether the water sprinklers are ON/OFF at any given time. The GSM modem is a highly flexible plug
and play quad bandSIM900. It Supports features like Voice, SMS, Data/Fax, GPRS and integrated TCP/IP stack
[24].
.
III. WIRELESS TECHNOLOGIES USED BY EXISTING IOT BASED SMART
IRRIGATIONS SYSTEMS.
For the purposes of proper and efficient operation of the smart irrigations systems, a wireless
communications system capable of controlling the actuators and sensors is essential. This ability share RFM69,
IEE802.15.4 Zigbee, LoRa and LoRaWaN technologies.

TABLE 1. The Comparative table of RFM69, IEEE802.15.4 Zigbee, LoRa and LoRaWAN
IEEΕ
RFM6
802.15.4 LoRa LoRaWAN
9
Zigbee
Frequency(MHz) 433 2.400 433 868
Data rate (kbps) 1.2-300 20-250 0.3- 22
Transmit/ Receipt
100 63 25
Power (mW)
Coverage LOS
1.600 10-100 2.000-5.000
Distance (m)
Tree, star or
Topology Star Star-of-stars
mesh

RFM69 operates at 433GHz. Can cover a star topology network within 1,6Km. The data rate ranges
between 1.2-300kbps, however, the power consumption for transmission and reception is higher than Zigbee
and Lora. Α much smaller surface of land (up to 100m) can serve the IEEE802.15.4 Zigbee. The Zigbee
modules operate within the ISM 2.4 GHz frequency band. A Zigbee network can be a star, tree or mesh
topology. The MAC layer of Zigbee attempts to achieve energy saving by exclusive use of medium access
control facilities so that higher layers in the protocol stack are unaffected and unaware of this. A typical LoRa or
LoRaWAN network is a „star-of-stars‟ topology. Both can cover the distance between 2 – 5km, i.e. much
greater distance than that covered by the aforementioned technologies while consuming less power. The
essential difference between Lora and LoRaWAN is, that LoRaWAN standardizes and extend the LoRa physical
communication layer onto internet network. Based on Table 1 the critical factors for selecting the appropriate
wireless technology are the necessary transmission range, the data rate and the power consumption. If the
distance between the Internet gateway and the microcontroller is greater than 100 meters then IEEE 802.15.4
Zigbee should not be chosen. If the distance is greater than 5000 meters then the 3G/GPRS cellular network is
the only solution. If the data rate of the irrigation system is greater than 22 kbps then the LoRaWan technology
should not be chosen. The authors in order to test the wireless technologies in various distances measured the
signal power that RFM69, IEEE802.15.4 Zigbee, LoRa wireless modules transmit. The outcomes of these
experiments for 50m, 1500m and 4000m are depicted in Table 2, 3 and 4.

International organization of Scientific Research 16 | P a g e

 Paper preparation guidelines for IOSR Journal of Engineering

TABLE 2. Signal Power for Cover Distance 50m

Transceiver Range (m) Signal Power (dBm)
RF69HCW 50 -45
SX1276 LoRa 50 -45
Xbee 63mW 50 -56

TABLE 3. Signal Power for Cover Distance 1500m

Transceiver Range (m) Signal Power (dBm)
RF69HCW 1500 -94
SX1276 LoRa 1500 -89
Xbee 63mW 1500 -

TABLE 4. Signal Power for Cover Distance 4000m

Transceiver Range (m) Signal Power (dBm)
RF69HCW 1500 -
SX1276 LoRa 1500 -104
Xbee 63mW 1500 -

It is obvious that if the distance range is shorter than 50m all technologies transfer IoT data
successfully. If the distance is close to 1500m then only the SX1276 LoRa and the RF69HCW module manage
to transfer data. If the distance range is equal to 4000m only the SX1276 LoRa module manage to transfer data.

IV. SENSORS, ACTUATORS AND MICROCONTROLLERS IN SMART IRRIGATION

SYSTEMS
The sensors, the actuators and the microcontrollers that are usually used in a smart irrigation system ,
are the following:
1. Arduino microcontroller. It is a single board microcontroller that collects the data from the sensors. It is
responsible to perform statistical analysis and controls in order to decide whether to open the valve for
watering or not. It has multiple digital and analog input and output pins and consumes little power energy.
All the sensors, the actuators and the data transceivers are connected to the microcontroller. If the control
unit demands more computational power then the Raspberry Pi microcomputer is proposed [14]. In that
case a analog to digital (A2D)converter is used since the RPi doesn‟t have analog inputs. A common A2D
converter is the MCP3008.
2. Water flow sensors. These sensors help the microcontroller to measure the amount of water that is used for
irrigation. The amount of the water depends on the water pressure, the diameter of the pipe and the watering
time.
3. Soil moisture sensors [13]. These sensors measure the moisture of the ground. They are placed near the
roots of the plant in order to be more accurate. They are analog or digital sensors. Their value range is
[0..1024]. As ground humidity varies from place to place, multiple soil moisture sensors are used, in order
to have a safer outcome. Figure 1 depicts the sensitivity and border conditions of two soil moisture sensors
over rainfall that are placed in two diagonal edges of one square meter flat test-bed field. These sensors
should be calibrated in order to establish the desirable moisture.

International organization of Scientific Research 17 | P a g e

 Paper preparation guidelines for IOSR Journal of Engineering

Figure 1. Soil sensors measurements over rainfall placed in two diagonal edges of a flat 1 square meter test-bed
field
4. Temperature sensors. These sensors are used to measure the air temperature. The air temperature give
important date in order to calculate the dehydration of the plants. Many plants are sensitive to air
temperature during irrigation. Especially in extreme weather conditions, very hot or very cold, the
irrigation should be avoided.
5. Air humidity sensors. Air humidity is also a crucial factor for plant dehydration. Air humidity also affects
growth, flowering, and water relations of the plants.
6. The water valve. It is responsible for opening and closing the water flow. It is controlled from the
microcontroller with the help of a power mosfet transistor switch. If the water systems under pressure, then
a solenoid valve is often used. If we want to control the water flow then a servo valve is chosen.
7. Internet gateway. If the field is located in a remote location then the 4G/3G/GPRS network is used to access
the internet. In that case a 3G dongle/shield is used to connect the microcontroller to the cloud. If the field is
close to an Internet gateway, then other communication technologies , as Bluetooth LE, Zigbee and WiFi,
could be chosen. The wireless technology that is chosen is based on Table 1.
8. Power system. Since most of the field do not have a power line available, the solar energy is often chosen.
A solar system is composed by a solar panel, a battery and a charge regulator. Since the input voltage of its
component is deferent, DC-DC duck or boost step up voltage regulators are used to change the available
input voltage. If the field is connected to the powerline then a simple AC-DC voltage adapter is used.
V. PROPOSED SYSTEM ARCHITECTURE OF THE IRRIGATION SYSTEM
The proposed system architecture of the irrigation system is depicted in Figure 2. The structural parts
of the watering system as the microcontroller, the sensors, the actuators and the internet gateway and the power
system are demonstrated. All the individual parts of the system architecture are analyzed in the previous section.

Figure 2. High Level System Architecture of a Smart Irrigation System.

VI. THE FUZZY PROPOSED WATERING SYSTEM ALGORITHM

A. Input- Outputs
As the above system has multiple sensors and as input and one actuator, the valve, as output there must
be an algorithm that decides when the valve should open, how much should be opened and when to close. The
authors propose a fuzzy algorithm for the control of the watering valve.
As input the system has:
 eight ground moisture sensors scattered in the field.
 One air humidity sensor.
 One temperature sensor
As output the system has:
 One servo valve that irrigates the field.

B. Membership Functions
As membership function for all the inputs we used the Gaussian function.

International organization of Scientific Research 18 | P a g e

 Paper preparation guidelines for IOSR Journal of Engineering

For the eight ground moisture sensors the state, the range, the Standard Deviation, the mean value for
the gaussian function are given to Table 5.
TABLE 5. Membership Functions of the ground moisture sensors.
Sensor Name Membership Range Standard Mean
of MF function Deviation value
Ground moisture Dry Gaussian [0-1024] 256 0
sensors
Ground moisture Normal Gaussian [0-1024] 256 512
sensors
Ground moisture Wet Gaussian [0-1024] 256 1024
sensors

For the temperature sensor the state, the range, the Standard Deviation, the mean value for the
gaussian function are given to Table 6
.
TABLE 6. Membership Functions of the temperature sensor.
Sensor Name Membership Range Standard Mean
of MF function Deviation value
Temperature sensor Cold Gaussian [-20, 40] 15 -20
Temperature sensor Normal Gaussian [-20, 40] 15 10
Temperature sensor Hot Gaussian [-20, 40] 15 40

For the humidity sensor the state, the range, the Standard Deviation, the mean value for the
gaussian function are given to Table 7.

TABLE 7. Membership Functions of the temperature sensor.

Sensor Name Membership Range Standard Mean
of MF function Deviation value
Humidity sensor Dry Gaussian [0, 100] 25 0
Humidity sensor Normal Gaussian [0, 100] 25 50
Humidity sensor Wet Gaussian [0, 100] 25 100

For the output of the Fuzzy system there was the opening of the servo valve. The range of the
degrees that a servo valve can be opened are [0, 90]. There were three state of our valve based on
Table 8. The function that described the opening of the servo valve was the triangle function.

TABLE 8. Membership Functions of the Servo valve.

Output Name of MF Membership function Range MF Parameters
Servo valve Close Trapezoid-Shaped [0, 90] [-1, 0, 30, 45]
Servo valve Half -open Triangle [0, 90] [30, 45, 60]
Servo valve Open Trapezoid-Shaped [0, 90] [45, 60, 90, 91]
C. Rules
The rules of the Fuzzy system are set with the help of the farmer. The general rule of the
Fuzzy irrigation system are given to Table 9.
TABLE 9. Membership Functions of the Servo valve.
Input Parameters Output State
More than 3 Ground moisture Servo valve Open
sensors are Dry
Temperature sensor is Cold Servo valve Close
Less than 4 Ground moisture sensors Servo valve Half -open
are Normal and Temperature sensor
is Hot and Humidity Sensor is Dry
Less than 4 Ground moisture sensors Servo valve Open
are Dry and Temperature sensor is
Hot and Humidity Sensor is Dry
Less than 3 Ground moisture sensors Servo valve Half -open
International organization of Scientific Research 19 | P a g e
 Paper preparation guidelines for IOSR Journal of Engineering

are Dry and Temperature sensor is

Hot and Humidity Sensor is Dry
In all other combinations of the input values the Servo valve is closed. As a defuzzification method,
the center of gravity of the set is chosen.

VII. CONCLUSIONS
This paper presents a smart IoT Fuzzy irrigation system. It uses low-cost off the shelf sensors
and actuators for continuously monitoring environmental conditions of arable areas. Our system
proposes an integrated solution that maximizes the quality and the quantity of the agriculture
production. Air temperature, moisture and soil humidity are continuously monitored. With the help of
servo valve the ground humidity is controlled. The high-level architecture of the system and its
individual parts are presented and analyzed. It is based on open source programming environments
and languages such as Linux, PHP, and MySQL. Moreover, a novel fuzzy computational algorithm
responsible for water irrigation is proposed. As input for that algorithm has the air humidity and
temperature and the soil moisture from multiple ground humidity sensors. Each of the input has three
levels in the fuzzy system. The output of the algorithm controls the opening of the central servo valve
of the irrigation system.

REFERENCES
[1] S. R. Kumbhar, Arjun P. Ghatule, “Microcontroller based Controlled Irrigation System for Plantation”,
Proceedings of the International Multi Conference of Engineers and Computer Scientists, Vol. 2, March
2013.
[2] Halahan, P. B., McIntyre, J. P., Coopersmith, M., & Puckett, M., "System and Method for Smart
Irrigation", U.S. Patent Application No. 14/612,229, 2015
[3] Ko, J., & Piccinni, G. (2009). "Corn yield responses under crop evapotranspiration-based irrigation
management.", Agricultural Water Management, vol. 96, pp 799-808., 2009
[4] Allen, R. G., Pereira, L. S., Raes, D., & Smith, M., "Crop evapotranspiration-Guidelines for computing
crop water requirements FAO Irrigation and drainage paper 56.", FAO, Rome, 300(9), D05109, 1998
[5] Migliaccio, K. W., Morgan, K. T., Fraisse, C., Vellidis, G., & Andreis, J. H. (2015). Performance
evaluation of urban turf irrigation smartphone app.Computers and Electronics in Agriculture, 118, 136-
142.
[6] Majone, B., Viani, F., Filippi, E., Bellin, A., Massa, A., Toller, G., ... & Salucci, M. (2013). Wireless
sensor network deployment for monitoring soil moisture dynamics at the field scale. Procedia
Environmental Sciences, 19, 426-435.
[7] Navarro-Hellín, H., Martínez-del-Rincon, J., Domingo-Miguel, R., Soto-Valles, F., & Torres-Sánchez, R.
(2016). A decision support system for managing irrigation in agriculture. Computers and Electronics in
Agriculture, 124, 121-131.
[8] Koubachi Company. , "The Plant Sensor for your Home and Garden", April 2015.
[9] Parameswaran, G., and K. Sivaprasath. "Arduino Based Smart Drip Irrigation System Using Internet of
Things.", International Journal of Engineering Science, vol 5518, 2016.
[10] Kumar, M. Kranthi, and Srenivasa Ravi. K., "Automation of Irrigation System based on Wi-Fi
Technology and IOT.", Indian Journal of Science and Technology, vol. 9, issue 17, 2016.
[11] Harun, A. N., Kassim, M. R. M., Mat, I., & Ramli, S. S., "Precision irrigation using Wireless Sensor
Network."IEEE Int. Conf. on. Smart Sensors and Application (ICSSA 2015), pp. 71-75, 2015.
[12] FAO Corporate Document Repository, "Irrigation Water Management: Training Manual No. 1 -
Introduction to Irrigation", June 1985.
[13] LU, Rong-jian, Pin Li, and Zhou Sun. "Application of SHT10 sensor in humidity-and-temperature
monitoring system", Transducer and Microsystem Technologies vol 9, pp 0-40, 2012.
[14] Raspberry Pi Foundation. The raspberry pi, the low cost embedded hardware computer system, 2011.
[15] Gardena soil moisture sensor, An automatic water saving irrigation component for small field irrigation,
http://www.gardena.com/int/water-management/water-controls/soil-moisture-sensor, 2015.
[16] Hunter Industries, Soil clik moisture sensor irrigation system, http://www.hunterindustries.com/irrigation-
product/sensors/soil-cliktm,2014.
[17] Mihajlov, B., and Bogdanoski, M.."Overview and Analysis of the Performances of ZigBeebased Wireless
Sensor Networks", International Journal of Computer Applications, vol. 29, no. 12, 2011.
[18] Augustin, A., Yi, J., Clausen, T., and Townsley, W. M. (2016). "A Study of LoRa: Long Range & Low
Power Networks for the Internet of Things", Sensors, 16, 1466.

International organization of Scientific Research 20 | P a g e

 Paper preparation guidelines for IOSR Journal of Engineering

[19] Baronti, P., Pillai, P., Chook, V.W.C.,Chessa, S., Gotta A., Hu, Fun Y.. "Wireless sensor networks: A
survey on the state of the at and the 802.15.4 and ZigBee standards", Computer Communications 30,
1655–1695, 2007
[20] ZigBee Alliance, "ZigBee Specification", ZigBee Document 053474r17, 2008
[21] Orange Connected Objects & Partnerships, "LoRa Device Developer Guide", 2016
[22] Lora Alliance, "LoRaWAN- What is it?- A technical overview of LoRa and LoRaWAN", 2015
[23] Rajalakshmi, P., Mahalakshmi, S. D., "IOT Based Crop-Field Monitoring And Irrigation Automation ",
Proceedings of the 10th International Conference on Intelligent Systems and Control, ISCO 2016
[24] Mortensen, Leiv M. "Effects of air humidity on growth, flowering, keeping quality and water relations of
four short-day greenhouse species." Scientia horticulturae 86.4 (2000): 299-310.

International organization of Scientific Research 21 | P a g e

View publication stats