Iot Based Smart Irrigation System Using Machine Learning Algorithms
Gomathy B1Karpagapriya D 2, Mohamed Aasif M 3and ParthaSarathy E 4
1 Professor/CSE, KPR Institute of Engineering and Technology, Coimbatore, [email protected]
2,3,4
UG Scholars/CSE, KPR Institute of Engineering and Technology, Coimbatore, [email protected]
Abstract: A smart irrigation system is a combination of hardware and software applications with various technologies. It has a data analysis method with a variety of techniques and models to read data directly. Information technology plays an important role. An application for machine learning helps to maintain a proper irrigation system.
For example, measuring hand-held interventions, increasing water use and providing water and fertility in the field improves crop production. It helps farmers to adapt to the right system according to their needs.Machine learning prediction algorithm is added for predicting the temperature and humidity in the environment.
.
Index Terms: Machine learning ,Arduinouno, Sensors, IoT
I. Introduction
Agriculture is a large number of GDP (Gross Domestic Product) not only for developing countries but also for many developed countries. Therefore, upgrading and improving existing agricultural technologies is appropriate. It will not only contribute to the sustainable development of human, plant and animal life but also address global challenges such as climate change and the drought-like epidemic. Technology needs to be available at a lower cost to make its impact on billions of people around the world. Therefore, it will help to protect conditions such as hunger and malnutrition.
1.1 Arduino:
The micro control is small and costs less. It is an open source microcontroller board. The important part is that the little controller contains a processor and memory, and some input / output pins that you can control. (GPIO - Standard Input Anchors). Uno means "one" in Italian and marks the release of Arduino Software (IDE). ArduinoBoard , and a reference model for the Arduino platform; For a comprehensive list of current, past or expired boards
see the Arduino .
Fig - 1: Arduino
1.2 Temperature Sensor:
A device that measures natural temperature. Here we use the LM35 temperature sensor. LM35 is an integrated region, the effect of which varies according to the temperature around us. It is used to measure temperatures between -55 ° C to 150 ° C. It measures the temperature of a particular environment.
Fig - 2: Temperature sensor 1.3 Soil Moisture Sensor:
Soil sensors are often used to check volumetric water volume in the soil. The relationship between the measured material and the soil moisture should be tagged and should vary according to natural factors such as soil temperature or electrical object. The microwave radiation shown is filled with wet soil and is used for distant sensations in geophysics and agriculture. Soil sensors often visit sensors that balance excessive water content. Some groundwater is known as water; these square sensors are sometimes referred to as groundwater sensors and accept tensiomers and mineral blocks
Fig – 3 :Soil Moisture Sensor
1.4Humidity Sensor:
Moisture is the availability of water content in the air. The amount of steam in the air will affect human comfort in the same way as several industrial processes. The presence of the vapor effect has a variety of physical, chemical, and biological processes. Industrial moisture level is important because it should have an impact on business prices and the health and safety of employees. Here we have a tendency to use an opposing type of moisture detector.
Fig – 4 :Humidity Sensor 1.5Node Mcu:
The node MCU firmware open-source board design is available. The firmware uses light- weight language, high-level, multi-paradigm. ESP8266 Core Arduino IDE used for software development.
Fig – 5 : Node MCU 2.LiteratureSurvey:
Comprehensive book reviews have been done and a few working technologies and algorithms that support book reviews and testing are suggested within the paper when there is a Smart Farm Monitoring System.
RuanYue et al has proposed a paper called water quality monitoring system based on a network of solar power and solar energy [1] A novel system architecture composed of many distributed sensor nodes and a base station is proposed to tr ack water quality in different field sites and in real-time. WSN technology is used to link the nodes and base station. A prototype system based on WSN technology and a single solar -powered node has been developed.The pH, turbidity, and oxygen density data collected by various sensors on the node side are sent to the base station via WSN. The system has many benefits, including low carbon emissions, low power consumption, and greater deployment flexibility.
ZesongFei et al, published the paper Survey of Multi-Objective Optimization in Wireless Sensor Networks-Metrics, Algorithms and Open Problems [2] —Wireless sensor networks (WSNs) have sparked a lot of interest in the research community, particularly for monitoring and surveillance applications. However, striking convincing trade-offs among the numerous competing interests is difficult energy dissipation, packet- loss rate, coverage, and lifetime are all optimization parameters. This paper uses the multi-objective methodology to include a tutorial and survey of recent research and development initiatives addressing this topic. Furthermore, we summarise a number of recent MOO studies in the sense of WSNs, with the aim of providing researchers with useful guidance for comprehending the references with literary work Finally, we explore a number of open issues that will be addressed by future studies.
The leaf spot of phaeosphaeria was considered to be one of the real diseases that weaken the sound of corn production in tropical and subtropical areas of Africa .Elhadi Adam et al, published the paper Detecting the First Stage of Phaeosphaeria Leaf Spot Infestations in Maize Crop Using in Hyperspectral Data and Guided Regularized Random Forest Algorithm. [3]
3. Proposed System
The proposed system provides the best solution for farm and irrigation requirements depending on the various open source information available online and in the machine learning algorithm. The solution available for smaller controllers is Arduino Uno, the choice of a small controller depends on their calculation capabilities, cost and ease of availability.
With the use of a variety of sensors, variable parameters will be regularly monitored and irrigated appropriately and to determine the type of plant to be produced .Local water level will also be determined. Temperature details will be uploaded to the cloud and accessed through a machine learning algorithm and run.
4. Conclusion
Machine learning in a sensible irrigation system is a useful and effective way to reduce water resources for agricultural crops. This method is worn to cultivate in areas where there is a shortage of water thus improving comfort. within the existing system human mediation is required to chase the system for a minute, but during this irrigation system the machine can be read by contributing to the weather that can lead to the weather prediction. The system is familiar and economical.
References
[1] RuanYue, Tang Ying, ―The Internet of Things: a survey, ―A water quality monitoring system based on wireless Sensor network & solar power supply‖, 2011 IEEE International
Cloudstora ge
Node mcu
ard
uino Humidi
ty
Sen sor
Power
suppl y
Temper ature
sens or
Soilmoi sture
sens
or
Conference on Cyber Technology in Automation, Control, and Intelligent Systems
[2] ZesongFei, Chengwen Xing, ―the Survey of Multi-Objective Optimization in Wireless Sensor Networks-Metrics, Algorithms and Open Problems,‖ IEEE Communications Surveys & Tutorials ( Volume: 19, Issue: 1, Firstquarter 2017)
[3] Elhadi Adam, Houtao Deng, John Odindi, Elfatih M. Abdel-Rahman and OnisimoMutanga,
―Detecting the first Stage of Phaeosphaeria Leaf Spot Infestations in Maize Crop Using in place Hyperspectral Data and Guided Regularized Random Forest Algorithm. ‖ Research Article | Open Access Volume 2017 |Article ID 6961387
[4] V. Ramasamy, B. Gomathy, Joy Lal Sarkar, Chhabi Rani Panigrahi,BibudhenduPati, AbhishekMajumderDCQSH: Dynamic Conflict-Free Query
Scheduling in Heterogeneous
Networks during EmergencyComputación y Sistemas, Vol. 25, No. 1, 2021, pp. 117–128doi:
10.13053/CyS-25-1-3455 Authors’ Profiles
Dr.B.Gomathyreceived the B.E degree in Computer Science and Engineering from Sethu Institute of Technology, Madurai, Madurai Kamaraj University, India in the year 1999 and the M.E, degree in Computer Science and Engineering from Bannari Amman Institute of Technology, Sathyamangalam, Anna University, India in the year 2009. She has completed her Ph.D. in Anna University, Chennai, India in the area of Data Mining and Soft Computing.
Currently, she is working as Professor in the Department of Computer and Science Engineering, KPR Institute of Engineering and Technology, Coimbatore, India. She has published 38 articles in National and International Journals and more than 23 papers in International and National Conferences. She is member of ISTE,CSI.
Karpagapriya D is currently pursuing the B.E degree in Computer Science and Engineering in KPR Institute of Engineering and Technology, Coimbatore, India. She is the currently the active member of Computer Society of India (CSI)
Mohamed Aasif M is currently pursuing the B.E degree in Computer Science and Engineering in KPR Institute of Engineering and Technology, Coimbatore, India.he is the currently the active member of Computer Society of India (CSI)
ParthaSarathy E is currently pursuing the B.E degree in Computer Science and Engineering in KPR Institute of Engineering and Technology, Coimbatore, India.he is the currently the active member of Computer Society of India (CSI)
