Smart Irrigation System With Weather Prediction

1st Likith Chowdary Mallipeddi 2nd Naru Rohith Reddy

School of Information Technology and Engineering School of Information Technology and Engineering
Student Student
Vellore Institute of Technology Vellore Institute of Technology
Vellore, TamilNadu. Vellore, TamilNadu.
mallipeddilikith@gmail.com nrr8122000@gmail.com

4th Binyala Devang 3rd Pindukuri Rithvik

School of Information Technology and Engineering School of Information Technology and Engineering
Student Student
Vellore Institute of Technology Vellore Institute of Technology
Vellore, TamilNadu. Vellore, TamilNadu.
devangbinyala001@gmail.com rithupindikuri@gmail.com

6th Viswanadhuni Pavan Nikhil 5th Mukthineni Rahul

School of Information Technology and Engineering School of Information Technology and Engineering
Student Student
Vellore Institute of Technology Vellore Institute of Technology
Vellore, TamilNadu. Vellore, TamilNadu.
pavannikhil1223@gmail.com rahulmukthineni@gmail.com

7th A Saida Reddy 8th Ajith Kumar Santra

School of Information Technology and Engineering School of Information Technology and Engineering
Student Senior Professor
Vellore Institute of Technology Vellore Institute of Technology
Vellore, TamilNadu. Vellore, TamilNadu.
saidareddyakkala18@gmail.com ajitkumar@vit.ac.in

Abstract—Here in our paper an automatic plant watering I. O BJECTIVES

system is designed to facilitate the automatic supply of adequate
amount of water from a water source like small reservoirs or
tanks to the fields in the Gorantla Village in Guntur District
and can be used any time. As the crop we selected here is
paddy that grows in the clay soil type which means 80-100 is
the good moisture level and when it is from 60-80 the fields
must be irrigated below 60 considered to be dangerous to crop
as it can affect the yield. One of the objectives of this work
is to see how to optimize the use of water in the process of
irrigation. The method employed is to continuously monitor the
soil moisture level to decide whether irrigation is needed, and
how much water is needed in the soil or it also can be employed
with timer set irrespective of soil moisture. If the soil moisture
level is less than threshold, rain prediction is done. If there is no
occurrence of rain, a pumping grid mechanism is used to deliver
the needed amount of water to the soil. Therefore, the system
analyses the moisture content and the possibility of rainfall and
calculates the amount of water to be released to the field. The
work can be grouped into four subsystems namely, power supply,
sensing unit, control unit and pumping subsystems which make
up the automatic irrigation control system.
• To Predict the rainfall occurrence in Gorantla, Guntur
District, Andhra Pradesh using Machine Learning algo-
rithm.
• To irrigate fields automatically based on rain occurrence
and soil moisture level.
II. I NTRODUCTION
Since nowadays, in the age of advanced electronics and
technology, the life of human being should be simpler and
more convenient, there is a need for many automated systems
that can replace or reducing human effort in their daily
activities and jobs. Here we introduce one such system,
named as automatic plant watering system, which is a model
of controlling irrigation facilities that uses sensor technology
to sense soil moisture,temperature and humidity with a
micro-controller to make a smart switching device to help
millions of people. Since irregular watering leads to the
mineral loss in the soil and may end up with rotting the
plants, we can then somehow know if the soil really needs
to be watered and if so when required we water the plant
automatically. We can automatically water our field and
garden plants. This helps us to reduce human intervention
and can help with better water management practices.And in
the below figure we can see the architecture of the system.
III. L ITERATURE S URVEY
1) Vishal Saraswat and Rohith Gandhi, Ananth Rama
Raju M,Rajath Raj guru A (2020). Smart irrigation
system using machine learning. International
Journal of Progressive Research in Science and
Engineering, 1(3),186-189: Designing an automation
system to manage the irrigation system in the farmlands
and make sure that there is equal amount of water to
all the crops across the whole field. This project would
prove be better than manual operation, as it difficult to
operate due to possible human negligence and higher
maintained and staff cost. Generally, the farmers be
contingent on the rain to water their crops, then once
there is no rain they water their crops with available
water resource. This requires manual checking of the
fields and switching on/off the motor. This leads in
wastage of time and energy. The farmers need to check
their farms regularly to determine if enough water
is given to the farms. The introduction of automatic
irrigation helps the farmers to water the plants and save
the water resource without any human interpretation.
This project aims to help to help farmers in managing
the irrigation system and reduce the water wastage.
2) Nalluri Sravani, Somula Ramasubbareddy and
G. Kannayaram,(2019). Weather prediction using
clustering strategies in machine learning. Journal of
Computational and Theoretical Nanoscience,16(5/6),
1977-1981: One of the most challenging tasks in real
time environment is weather prediction. Understanding
the various factors that influences weather is as much
as important on knowing the specific methodology
on how to perform the prediction. Hence it is an
utmost necessary to identify the relationship among
different attributes used in the dataset. These datasets
can be obtained from various accessible databases
and information repositories. Even though it has
two different models (Predictive and Descriptive) for
extracting the information, used clustering method
of descriptive model, the justification for using such
method is explained in the methodology section. On
conforming to the above stated basic facts in this article
we perform forecasting.
3) Pratap N Kashyap, Preetham M S ,Rohan Venkatesh
Nayak, Uday B, and Radhika T V,(2020).Intelligent
Weather Forecasting Using Machine Techniques.
International Research Journal of Engineering and
Technology, 7(3): Weather forecasting has traditionally
been done by physical models of the atmosphere, which
are unstable to perturbations, and thus are inaccurate
for large periods of time. Since machine learning
techniques are more robust to perturbations, in this
paper we explore their application to weather forecasting
to potentially generate more accurate weather forecasts
for large periods of time. The scope of this paper
was restricted to forecasting the maximum temperature
and the minimum temperature for seven days, given
weather data for the past two days. A linear regression
model and a variation on a functional regression model
were used, with the latter able to capture trends in the
weather. Both of our models were outperformed by
professional weather forecasting services, although the
discrepancy between our models and the professional
ones diminished rapidly for forecasts of later days, and
perhaps for even longer time scales our models could
outperform professional ones. The linear regression
model outperformed the functional regression model,
suggesting that two days were too short for the latter
to capture significant weather trends, and perhaps
basing our forecasts on weather data for four or five
days would allow the functional regression model to
outperform the linear regression model.
4) Pradeep Hewage, Marcello Trovati, Ella Pereria,
Ardhendu Behera, (2020). Deep Learning-based
effective fine-graied Weather Forecasting model.
Pattern Analysis and Applications,24,343-366:
There is some evidence that better weather forecasts
can be produced by introducing big data mining
and neural networks into the weather prediction
workflow. Here, we discuss the question of whether it
is possible to completely replace the current numerical
weather models and data assimilation systems with DL
approaches. This discussion entails a review of state-of-
the-art machine learning concepts and their applicability
to weather data with its pertinent statistical properties.
We think that it is not inconceivable that numerical
weather models may one day become obsolete, but
several fundamental breakthroughs are needed before
this goal comes into reach.
5) Anisha Datta, Shukrity Si, and Sanket Biswas,(2020).
Complete Statistical Analysis to Weather
 Forecasting. Computational Intelligence in Pattern
Recognition, Book Chapter, 751-763: The primary
objective of the model applied in this work is to
predict the weather of a city named Austin in Texas
using supervised machine learning algorithms. In this
case, artificial neural networks and gradient boosting
classifier were implemented to build models to predict
weather and comparisons between these two models are
also made for this dataset. Here, average temperature,
average dew point, average pressure of sea level,
average percentage of humidity, etc. are the parameters
taken into consideration which influence the weather
of the place. Using these parameters, the trained
models performed a classification to predict whether
the weather is rainy (thunderstorm or not), not rainy,
snowy, or foggy.
6) Piyush Kapoor and Sarabjeet Singh Bedi,(2013).
Weather forecasting using sliding window algorithm.
ISRN Signal Processing, 2013(1), 1-5: To predict
the future’s weather condition, the variation in
the conditions in past years must be utilized. The
probability that the weather condition of the day in
consideration will match the same day in previous
year is very less. But the probability that it will match
within the span of adjacent fortnight of previous year is
very high. So, for the fortnight considered for previous
year a sliding window is selected of size equivalent to
a week. Every week of sliding window is then matched
with that of current year’s week in consideration. The
window best matched is made to participate in the
process of predicting weather conditions. The prediction
is made based on sliding window algorithm. The month
wise results are being computed for three years to check
the accuracy. The results of the approach suggested
that the method used for weather condition prediction
is quite efficient with an average accuracy of 92.2
7) Sudhan Murugan Bhagavathi, Anitha Thavasimuthu,
Aruna Murugesan, Charlyn Pushpa Latha George
Rajendran, Vijay A, Laxmi Raja, and Rajendran
Thavasimuthu,(2021). Weather forecasting and
prediction using hybrid C5. 0 machine learning
algorithm. International Journal of Communication
Systems, 34(1): In this research, a weather forecasting
model based on machine learning is proposed for
improving the accuracy and efficiency of forecasting.
The aim of this research is to propose a weather
prediction model for short-range prediction based on
numerical data. Daily weather prediction includes the
work of thousands of worldwide meteorologists and
observers. Modernized computers make predictions
more precise than ever, and earth-orbiting weather
satellites capture pictures of clouds from space.
However, in many cases, the forecast under many
conditions is not accurate. Numerical weather
prediction (NWP) is one of the popular methods
for forecasting weather conditions. NWP is a major
weather modelling tool for meteorologists which
contributes to more accurate accuracy. In this research,
the weather forecasting model uses the C5.0 algorithm
with K-means clustering. The C5.0 is one of the better
decision tree classifiers, and the decision tree is a great
alternative for forecasting and prediction.
8) Sam Cramer, Michael Kampouridis, Alex. A. Freitas
and Antonis K. Alexandridis,(2017). An extensive
evaluation of seven machine learning methods
for rainfall prediction in weather derivatives.
Journal of Expert Systems with Applications, 85,
169-181: Regression problems provide some of the
most challenging research opportunities in machine
learning, and more broadly intelligent systems, where
the predictions of some target variables are critical to
a specific application. Rainfall is a prime example,
as it exhibits unique characteristics of high volatility
and chaotic patterns that do not exist in other time
series data. This work’s main impact is to show the
benefit machine learning algorithms, and more broadly
intelligent systems have over the current state-of-the-
art techniques for rainfall prediction within rainfall
derivatives. We apply and compare the predictive
performance of the current state-of-the-art (Markov
chain extended with rainfall prediction) and six other
popular machine learning algorithms, namely: Genetic
Programming, Support Vector Regression, Radial
Basis Neural Networks, M5 Rules, M5 Model trees,
and k-Nearest Neighbours. To assist in the extensive
evaluation, we run tests using the rainfall time series
across data sets for 42 cities, with very diverse climatic
features. This thorough examination shows that the
machine learning methods can outperform the current
state-of-the-art. Another contribution of this work
is to detect correlations between different climates
and predictive accuracy. Thus, these results show the
positive effect that machine learning-based intelligent
systems have for predicting rainfall based on predictive
accuracy and with minimal correlations existing across
climates.
9) Wei-Chiang Hong,(2008). Rainfall forecasting by
technological machine learning models. Journal of
Applied Mathematics and Computation, 200(1),
41-57: Accurate forecasting of rainfall has been
one of the most important issues in hydrological
research. Due to rainfall forecasting involves a rather
complex nonlinear data pattern; there are lots of novel
forecasting approaches to improve the forecasting
accuracy. Recurrent artificial neural networks (RNNS)
have played a crucial role in forecasting rainfall data.
Meanwhile, support vector machines (SVMs) have been
successfully employed to solve nonlinear regression and
 time series problems. This investigation elucidates the
feasibility of hybrid model of RNNs and SVMs, namely
RSVR, to forecast rainfall depth values. Moreover,
chaotic particle swarm optimization algorithm (CPSO)
is employed to choose the parameters of a SVR model.
Subsequently, example of rainfall values during typhoon
periods from Northern Taiwan is used to illustrate the
proposed RSVRCPSO model. The empirical results
reveal that the proposed model yields well forecasting
performance, RSVRCPSO model provides a promising
alternative for forecasting rainfall values.
10) Gowtham Sethupathi. M, Yenugudhati Sai Ganesh
and Mohammad Mansoor Ali,(2021) Efficient
Rainfall Prediction and Analysis using Machine
Learning Techniques. Turkish Journal of Computer
and Mathematics Education (TURCOMAT), 12(6),
3467-3474: Rainfall prediction is a beneficiary one, but
it is a challenging task. Machine learning techniques
can use computational methods and predict rainfall
by retrieving and integrating the hidden knowledge
from the linear and non-linear patterns of past weather
data. Various tools and methods for predicting rain
are currently available, but there is still a shortage of
accurate results. Existing methods are failing whenever
massive datasets are used for rainfall prediction. This
study provides efficient rainfall prediction methods of
machine learning techniques: random forest and logistic
regression methods that provide an easy and accurate
prediction and determine which one is more effective
in comparison. This study would assist researchers in
analysing the most recent work on rainfall prediction
with an emphasis on machine learning techniques
and providing a reference for possible guidance and
comparisons. Anaconda framework is used, and the
coding language used is Python, which is portable and
dynamic. NumPy, matplotlib, seaborn, and pandas are
the libraries used for the implementation.
11) Fredy Angulo, Elsa Urueta, Gustavo Valverde
and Obert Paternina. O,(2017). Cartagena’s Water
Distribution System. International Journal of
Procedia Engineering, 186(1), 28-35: Water collecting
frameworks for a trim generation has an extensive
potential in semi-dry locales of Eastern and Southern
Africa; a potential which, to an expansive degree is
undiscovered. The explanation behind this is the solid
concentrate on in-situ water protection frameworks
and the restricted utilization of surge and capacity
water reaping. The improvement centers regarding
water assets administration have to a substantial degree
concentrated on an expansive scale, downstream found
frameworks, like water system plans. It appears that
the dominating size of water gathering improvement
at introducing in Eastern and Southern Africa is
on the family unit scale. There are significantly
fewer endeavors to plan and administration water
gathering frameworks on a sub catchment or catchment
scale. There is a significant arrangement of scattered
improvement endeavors on water reaping, however
apparently little investigate on the suitability of the
frameworks inside the setting of cultivating frameworks.
12) A M K El-Ghonemy,(2017). Performance test of
a sea water multi-stage flash distillation plant.
Alexandria Engineering Journal, 57(4), 2401-2413:
The water system has been the foundation of human
progress since man has begun horticulture. As the
age advanced, man created numerous techniques for
the water system to supply water to the land. In the
present situation on the preservation of water is of
high significance. Exhibit work endeavors to spare the
normal assets accessible to mankind. By constantly
observing the status of the dirt, we can control the
stream of water furthermore along these lines diminish
the wastage. The framework has a fused Bluetooth
for remote checking which lessens the issue of range
with NODE MCU system and spares SMS cost for the
rancher. The smoke sensors used to send crisis data to
the client in case of flame in the field or consuming of
the engine. The outline is low power, minimal effort,
and little size, vigorous and profoundly flexible. Along
these lines, this framework dodges over water system,
underwater system, topsoil disintegration and decrease
the wastage of water.
13) Katsunori Mizuno, Xiaofei Liu, Fuyuki Katase,
Akira Asada, Makato Murakoshi, Yosunobu Yagita,
Yoshiaki Watanabe, Tetsuo Shimada and Yasufumi
Fujimoto, (2016). Automatic non-destructive three-
dimensional acoustic coring system for in site
detection of aquatic plant root under the water
bottom. Journal of Nondestructive Testing and
Evaluation, 5, 1-8: A computerized plant observing
framework has been produced for plant water status
evaluation. In view of a dispersed flag preparing chain
of command, information lumberjacks were utilized
for neighborhood sensor excitation and flag preparing
to diminish the unpredictability of the computer-based
information obtaining framework. The framework can
be utilized to gather natural information, measure
shelter temperature, and distinguish morphological
changes influenced by trial medicines utilizing non–
contact sensors. Contact sensors, for example, soil
dampness test and micro– lysimeters are accessible
for deciding evapotranspiration rate. Wellsprings of
detecting and estimation blunder were breaking down
and evaluated, and measures were taken to enhance
the information precision. Exact soil dampness level
could be kept up utilizing a closed– circle water system
sub-framework.
14) H Etienne and M Berthouly, (2002). Temporary
immersion systems in plant micropropagation.
Plant Cell, Journal of Plant Cell,Tissue and Organ
Culture, 69, 215-231: The sensor with the brass
plate as the electrode and washed sand as porous
medium confirmed almost a consistent fashion within
the dating between resistance and soil moisture content
material in all trials. The automatic structures based
on soil resistance become observed to be working
effectively without frequent supervision and maintained
the pre-set moisture content inside the root sector.
The automatic device based on leaf-air temperature
differential maintained the pre-set leaf-air temperature
differential for the duration of the study duration.
15) Krishna S Nemali and Marc W van Iersel, (2006). An
automated system for controlling drought stress and
irrigation in potted plants. Scientia Horticulturae,
110(3), 292-297: Effect of Plant Density and Harvesting
Intervals In this paper, we’ve focused particularly on
physiological records. But a few crucial issues ought
to be addressed before putting in place an industrial
gadget for minituber production. A fully automated
system monitoring each nutrient answer and dynamic
parameters, which include concentration, pH, and
float rate would seem ideal. It needs to be geared up
with appropriate excessive- tech security devices and
alarms to control electric powered energy, pumping,
spraying, and chemical variables. Each the fee for those
additional facilities and such exertions-intensive tasks
as staking and guide harvesting of plant life must be
economically assessed.
16) Imma Farran and Angel M Mingo-Castel,(2006).
Potato minituber production using aeroponics:
effect of plant density and harvesting intervals.
American Journal of Potato Research, 83, 47-53:
Not like most computerized irrigation systems which
result in leaching and run-off, our system had very
little wastage of water. The machine required little
protection at some point of observation. No matter the
time of the day, the gadget irrigated the plants while
the substrate moisture fell beneath the target stage.
The controller additionally has the capability to be
used in drought strain studies, since it’s far possible to
control the amount of water to be had in the substrate
(or soil) and consequently the extent of stress that
the plant is uncovered to. The subsequent conclusions
may be drawn from this look at: In place of the
dry-down or common weighing approach for imposing
drought pressure, this machine-maintained u close to
the setpoint with little or no effect of surroundings and
plant length. The validation observes confirmed that u
maintained via the controller become dependable.
17) N Abraham , P S Hema, E K Saritha and S
Subramannian, (2000). Irrigation automation based
on soil electrical conductivity and leaf temperature.
Agricultural Water Management, 45(2), 145-157:
Transient immersion has fine consequences on all
tiers of shoot proliferation and somatic embryogenesis
with many plant species. Regenerated plantlets and
somatic embryos are of higher first-rate. Excellent
results have additionally been acquired at some stage
in acclimatization of plant cloth from temporary
immersion. Transient immersion combines the blessings
of strong culture media (most gasoline exchanges)
and liquid culturing. Their optimization, towards lower
values, outcomes in better organic yields through
more manage of morphogenesis, however also via the
manipulate of hyper hydricity Immersion times may be
as essential for each way of life degree as the length
of subcultures and the chemical composition of plant
boom media.
18) M Kacira and P Ling, (2001). Design and
development of an automated and Non–contact
sensing system for continuous monitoring of plant
health and growth. Transactions of the ASAE, 44(4),
989-996: We evolved and tested a non-unfavorable 3D
acoustic coring system under the water backside. A
discipline check becomes carried out in a pond and
the system worked with none troubles arising. The
sizing device can gather facts under the water backside
and reconstruct the spatial fame. The prevailing
observes showed that the natural lotus root (diameter of
0).020–zero. Half may be detected using the acoustic
coring technique. A measurement machine is a flexible
tool for the detection of lotus, the spread of which is
an environmental problem in EU trophied lakes.
19) D S Pavithra and M S Srinath,(2014). GSM based
automatic irrigation control system for efficient use
of resources and crop planning by using an Android
mobile. IOSR Journal of Mechanical and Civil
Engineering, 11(4), 49-55: The performance looks at
of a seawater multi-stage flash (MSF) desalination plant
is finished at a hundred load, 70percent load all through
operation in summer and iciness. Its miles concluded
that: To understand the manner and plant operation, the
plant key working parameters are supplied for the tested
cases (number of degrees, pinnacle brine temperature
(TBT), temperature and strain variant inside the levels,
and brine and flash steam flows. additionally, the
plant normal performance indicators are supplied for
the studied cases [performance ratio (PR) or gain
output ratio (GOR), and concentration ratio (CR). The
consequences confirmed that: For the same plant output
at a hundred capacity, the principal cooling water is
reduced from 47.1 percent to 20.1 percent for case-1
(summer) and case-2 (winter) respectively, which in
turns immediately reduces the pumping energy by
 means of the identical ratio.
20) J. Rockstrom,(2000). Water resources management
in smallholder farms in Eastern and Southern
Africa. Journal of Physics and Chemistry of the
Earth, part B: Hydrology, Oceans and Atmosphere,
25(3), 275-283: Cartagena’s WDS has skilled over
the last two decades, sizable changes in its bodily
configuration, going from a unified and radial network
with delivering especially via gravity, to a network
sectioned with waft prices and pressures monitored
and managed. This configuration has allowed the water
distribution optimization, lowering the unaccounted-for
water price and postponing investment for the system
enlargement. Mathematical modeling of the water
distribution machine has been essential to assess and
remedy troubles such as imperceptible leak detection;
water excellent; assessment of systems potential to meet
needs of recent city traits, and insurance extension of
the water distribution provider in areas above carrier
provision stage through the implementation of pressure
ranges and pumping systems; amongst others.
21) Bernard cardenas-Lailhacar,Michael D.Dukes,(2008).
Expanding disk rain sensor performance and po-
tential irrigation water savings. Journal of Irriga-
tion and Drainage Engineering, 134(1), 67-73: Rain
sensors (RSs) appear to be a useful tool for water
conservation at a relatively low cost. However, little
evidence related to RS performance and/or reliability
exists. Here the objective of this experiment is to:
a) evaluate two RS types with respect to the fol-
lowing: Accuracy of their set point, number of
irrigation cycles bypassed, and duration in bypass
mode.
b) quantify the amount of water that RSs could save.
c) estimate their payback period. Mini-click (MC) and
wireless rain-click (WL) rain sensor models were
monitored.
22) Leah Meeks, Michael D Dukes, Kati White
Migliaccio and Bernard Cardenas-Lailhacar, (2012).
Long term expanding-disk rain sensor accuracy.
Journal of irrigation and drainage engineering,
138(1), 16-20: Rain sensors are devices that connect
to automatic irrigation systems to interrupt scheduled
irrigation events because of rainfall. Here The accuracy
of expanding-disk rain sensors over nearly five years
was evaluated at a research site in north central Florida.
Seven rain sensor model/rainfall set point combinations
were evaluated. The seven configurations are as follows:
Wireless Rain-Click (WL) rain sensor, Mini-Click rain
sensor and 13 mm Hunter, 3 MC, 6 MC, Irritrol RFS
, and Toro TWRS. The MC and WL configurations
were evaluated for 1,182 days with four replicates. The
Hunter, Irritrol, and Toro configurations were monitored
1,150 days with eight replicates. Data were collected
for each configuration to determine the total time in
interrupt mode. This information was compared with
measured rainfall data on site. On average, the WL,
3MC, 6MC, 13MC, Hunter, Irritrol, and Toro Accuracy
ranged from 27 to 97 percent.
23) Yudi Irawan Chandra, Marti. Riastuti, Kosdiana and
Edo P. Nugroho, (2020). Automatic Garden Umbrella
Prototype with Light and Rain Sensor Based on
Arduino Uno Microcontroller. International Journal
of Intelligent Robotics, 2(2), 42-51: when the weather
is bright, and the sun is shining hot the sensors get
the data from the rain sensor and temperature sensor.
Therefore, an automatic umbrella tool was made that
functions as a shelter in the garden. In this there is
a light sensor module and a rain sensor, which is
controlled with the Arduino Uno microcontroller as an
input data processor and an L298N motor driver, which
functions to regulate the speed and direction of the DC
motor rotation (to the right and left) as an umbrella
drive. When the motor rotates to the right, the umbrella
will open, while when the motor rotates to the left, the
umbrella will close again.
24) Gad I and Hosahalli D, (2020). A comparative study
of prediction and classification models on NCDC
weather data. International Journal of Computers
and Applications,1-12: In this work, to examine
whether these models are robust to predict National
Climatic Data Center (NCDC) weather conditions, we
carried out to compare newly emerging models with
traditional meteorological models. In this paper, a set
of the most common machine learning techniques are
explored to generate robust weather forecasting model
for long periods of time. Moreover, the combinations of
all the model parameters are considered for simulations
and the performance results of each method using a
10-fold cross-validation procedure are presented. The
experimental results of the classifiers show that the
decision tree CART, XGBoost and AdaBoost models
exhibit better classification accuracy when compared
with the other methods and for regression task, the linear
regression method performs better in terms of R2 metric.
25) Bahzad Charbuty and Adnan Mohsin Abdulazeez,
(2021). Classification Based on Decision Tree
Algorithm for Machine Learning. Journal of
Applied Science And Technology Trends, 2(1),
20-28: Here in this they had taken a dataset which
contains various data and help to predict rain and after
that they had developed the machine learning model.
After that they do the same with the Decision tree and
then compare the accuracy and the results after which
we get to know that the decision tree will be one of the
good models to go with when we are making predictions

26) Saurav Barua,(2020). A Naı̈ve Bayes Classifier
Approach to Incorporate Weather to Predict
Congestion at Intersections. World Academics
Journal of Engineering Sciences, 7(2),72-76: Here in
this journal the author had worked upon the weather
prediction using the Naı̈ve Bayes after which the
results were pretty good, and the author had used the
various sensors to collect the data. The sensors include
the humidity sensor, temperature sensor, rain sensor,
precipitation sensor. All the data from the sensors are
logged and then the data is used for the prediction.
27) C. Karthikeyana, Gurram Sunithab, J. Avanijac,
K. Reddy Madhavid and E.S. Madhane,(2021).
Prediction of Climate Change using SVM and
Naı̈ve Bayes Machine Learning Algorithms.
Turkish Journal of Computer and Mathematics
Education,12(2),2134-2139: Here in this work, they
work on the Naı̈ve Bayes based weather prediction
model that stores the data on AWS and the model itself
is also deployed on the AWS which helps them to make
the model available 100 percent to the customers and
not only that they had taken a dataset and trained and
tested the model after that for the better accuracy they
had gone into dimensionality reduction.
28) I K Nti, J Aning, B K Ayawli, K Frimpong,
A Y Appiah and O Nyarko-Boateng, (2021). A
Comparative Theoretical and Empirical Analysis
of Machine Learning Algorithms. Webology, 17(1):
This paper examines the performance of twenty-
one (21) MLAs for classification and regression
tasks based on six datasets from different domains.
Empirically they compare their prediction results
based on accuracy, balanced accuracy, F-score, Area
Under the Curve (AUC), root mean square error,
r-squared and adjusted r-squared. The random forest
algorithm gave a consistent performance across all
the six datasets in classification and regression tasks.
However, on average, the XGBoost outperformed all
MLAs applied in this study. The dummy algorithm
and the linear regression were more moderate than the
rest of the applied MLAs in computational complexity.
Nevertheless, the overall study outcome shows that
MLAs algorithms efficiently solve everyday challenges
in elections outcome, financial fraud, network intrusion
detection, meteorological forecast and heart diseases
discovery; but their performance varies across domains
and dataset dimensions.
29) Raju N A, (2015). Land capability and suitability
in Vizianagaram district of Andhra Pradesh using
remote sensing and GIS techniques. IOSR J of
Humanities and Social Sci. (IOSR-JHSS),20,56-64:
Land evaluation procedure given by FAO for soil
site suitability for various land utilization types has
been used to assess the land suitability for different
crops and for generating cropping pattern for kharif
(summer) and rabi (winter) seasons in Vizianagaram
district of Andhra Pradesh. Vizianagaram suffers
from many types of land degradation such as such
as salinity, waterlogging, ravines, degraded hills and
rock quarries. The database on soil, land use/land
cover was generated from data derived from IRS -P6
remote sensing satellite and soil survey to perform
an integrated analysis in the geographic information
system environment. Agricultural and non-agricultural
lands were delineated using the Decision Tree Classifier
(DTC) and non-agricultural areas were masked for
removal from future analysis. Different soil chemical
parameters and physical parameters were evaluated
for different crops. Subsequently all of them were
integrated using a multi criteria decision making and
GIS to generate the land suitability maps for various
crops. Kharif and rabi season cropping patterns maps
were developed by integrating crop suitability maps
for the winter and summer seasons separately. Results
indicated that about 55 percent is highly suitable (S1)
for sugarcane and 60 percent, 54 percent and 48 percent
of the area are moderately suitable (S2) for cultivation
pearl millet, mustard and rice respectively. 50 percent
of the area is found to be marginally suitable (S3) for
growing maize. It was also found that better land use
options could be implemented in different land units
as the conventional land evaluation methods suffer
from limitation of spatial analysis for the suitability of
various crops.
30) Jaroslav Knápek , Tomas Králı́k , Kamila Vávrová
and Jan Weger,(2020). Dynamic biomass potential
from agricultural land. Renewable and Sustainable
Energy Reviews, 134: The potential of biomass from
agricultural land cannot be a constant value over time,
but rather as a dynamic one influenced by changes in
many factors and characteristics. The most important
include the amount of available agricultural land,
the structure of cultivated crops, allocation of energy
crops, the effect of a learning curve and the impact
of climate change. Our method used for determination
of biomass potential is based on usage of the GIS
model that includes key parameters such as soil and
climatic characteristics of farming plots and expected
crop yields related to them, changes in arable land,
planting area, utilization of biomass, and the influence
of climate change.
31) Yikuan Tang, Fan Zhang, Bernard A. Engel, Xiao
Liu, Qiong Yue and Ping Guo, (2020). Grid-scale
agricultural land and water management. Journal
of Cleaner Production, 265: In This paper they
developed a remote-sensing-based multiobjective
 (RSM) approach to formulate sustainable agricultural
land and water resources management strategies
at a grid scale. To meet the spatial resolution and
accuracy need of agricultural management, downscaled
precipitation data sets were obtained with the help of
global precipitation measurement (GPM) data and other
spatial information. Spatial crop water requirement
information was obtained via the combination use of the
Penman-Monteith method, remote sensing information
(MOD16/PET) and virtual water theory. Through
integrating these spatial data and considering the impact
of different spatial environments on crop growth,
a grid-based integer multiobjective programming
(GIMP) model was developed to determine best
suitable crop planting types at all grids. GIMP can
simultaneously consider several conflicting objectives:
crop growth suitability, crop spatial water requirements,
and ecosystem service value. Further, GIMP results
were inputted into a grid-based nonlinear fractional
multiobjective programming (GNFMP) model with
three objectives: maximize economic benefits, maximize
water productivity, and minimize blue water utilization,
to optimize irrigation-water allocation.
32) Elnashar, A., Abbas, M., Sobhy, H and Shahba,
M. (2021). Crop Water Requirements and
Suitability Assessment in Arid Environments.
Agronomy,11(2),260: In this study, twenty soil profiles
are collected to represent the soils of the study area.
Physical and chemical properties of soil, in addition
to irrigation water quality, provided data are utilized
by the Agriculture Land Evaluation System for Arid
and semi-arid regions (ALES-Arid) to determine crop
suitability. University of Idaho Ref-ET software is used
to calculate CWR from weather data while the Surface
Energy Balance Algorithms for Land Model (SEBAL)
is utilized to estimate CWR from remote sensing data.
The obtained results show that seasonal weather-based
CWR of the most suitable field crops (S1 and S2
classes) ranges from 804 to 1625 mm for wheat and
berseem, respectively, and ranges from 778 to 993
mm in the vegetable crops potato and watermelon,
respectively, under surface irrigation
33) Suraj Kumar, Thendiyath Roshni, Ercan Kahya
and Mohammad Ali Ghorbani,(2020). Climate
change projections of rainfall and its impact on the
cropland suitability for rice and wheat crops in the
Sone river command, Bihar. Journal of Theoretical
and Applied Climatology, 142, 433-451: The aim
of this study is to develop land suitability maps for
rice and wheat crop based on physical, chemical and
climatic factors of production using a multi-criteria
evaluation (MCE) and Geographic Information System
(GIS) approach. Biophysical variables of soil, climate
and topography have been considered for suitability
analysis. This work also evaluates the impact of climate
change (CC) on cropland suitability for rice and wheat
crops. Climatic variable scenarios of rainfall were
obtained from four GCMs (CAN-ESM2, MPI-ESM-
MR, CSIRO, CMCC-CMS) for three future time slices
(2011–2040, 2041–2070, 2071–2100) of representative
concentration pathways (RCP) RCP 2.6, RCP 4.5 and
RCP 8.5 scenarios.
34) A Gaddam, M Al-Hrooby and W F Esmael, (2020).
Designing a wireless sensors network for monitoring
and predicting droughts. International Journal
on Smart Sensing and Intelligent Systems, 7(5),
1-6: Global warming and lack of rain were the main
problems that caused increased drought around the
world. In New Zealand, according to National Institute
of Water and Atmospheric Research (NIWA) the
drought in 2012 and 2013 was the worst drought in the
last 70 years. Therefore, there is a need for technological
intervention to monitor basic information about the
weather and soil condition to identify and predict
drought conditions. Initial experiments have shown
that the proposed wireless sensor drought monitoring
system is capable of remote real-time monitoring
for extended periods. This monitoring can also help
identify drought in the early stages and thereby indicate
promptly when to take corrective measures. Intelligent
sensors in a wireless network monitor the soil condition.
35) Basudeba Behera, Smita Kumari, Alisha Kumari
and Ajay Kumar, (2020). Application of IoT and
Weather Prediction for Enhancement of Agricultural
Productivity. In International Conference on
Computational Intelligence, Security and Internet
of Things,215-230: This will help in monitoring the
various parameters such as temperature, humidity, soil
moisture, etc. required for the favorable growth of any
crop. The proposed system makes use of technologies
such as Wireless Sensor Networks (WSN) and Internet
of Things (IoT) for solving the major challenges faced
in agricultural fields. The WSN includes the various
sensors interfaced with Raspberry pi whereas with the
help of IoT we will be able to obtain the acquired data
globally
36) Sk. Fahmida Islam, Morium Akter and Mohammad
Shorif Uddin, (2021). Design and implementation
of an internet of things based low-cost smart
weather prediction system. International Journal
of Information Technology, 13, 2001-2010: In the
present era, the internet of things (IoT) plays an
important role in diverse domains, including weather
prediction. Accurate weather forecasting is very crucial
for the huge production of crops. Different methods
such as mathematical and statistical models have
been developed to predict the weather condition of
 Bangladesh. In this paper, we have proposed an IoT-
based low-cost smart weather prediction system using
an artificial neural network. The proposed system uses
an IoT platform for monitoring the weather parameters
like temperature, humidity, air pressure, rainfall, soil
moisture.
37) Vivtor Mokaya, (2019).Future of Precision
Agriculture in India using machine learning
and artificial intelligence. Journal of Environment,
Science And Technology, 5(2),37-44: The yield and
quality of crops depends on various biotic, abiotic
and management related parameters. In conventional
agriculture the farmers relied on their experiences. Due
to human perception often, there is uncontrolled use
of resources and inputs resulting in not only natural
resources wastage and environmental pollution but
also financial loss of farmers. Precision agriculture
uses technology such GPS, sensors, Internet of Things,
robotics, drones, machine learning al decision support
systems
38) Alka Rani, Arti Kumari, Ram Narayan Singh and
Kavita Kumari,(2021). Climate-Smart Agriculture:
An Integrated Approach for Attaining Agricultural
Sustainability. In Climate Change and Resilient
Food Systems Book, chapter 5, 141-189: Impacts
of climate change (CC) and climate variability (CV)
are felt throughout the world as it’s getting hotter.
As a result, irregular distribution of precipitation,
a progressive rise in the oceanic levels, and more
frequent occurrence of extreme weather events (EWE)
have become a common phenomenon in recent years.
Extended episodes of drought, floods, and shifts in
the agroclimatic areas are threatening agricultural crop
production (ACP) throughout the globe.
39) H. Swapna Rekha,Janmenjoy Nayak and Bighnaraj
Naik, (2020). Impact of IoT in agriculture: advances
and challenges. Internet of Things Book,127-154:
Due to the vast development of information and
transmission technologies, Internet of things (IoT) has
been emerging as a revolutionary technology in the
simulation of real-time data. IoT aims in connecting the
real world with the virtual world through the Internet
and the World Wide Web that helps in exchanging
information without human intervention. In the world,
most of the people depend on agriculture. The surge in
the world population has demanded the modernization
of agriculture to meet the rising demand for food in
terms of both quality and quantity.
40) Giang Truong Le, Nihat Minh Tran and Thang Viet
Tran, (2020). IoT System for Monitoring a Large-
Area Environment Sensors and Control Actuators
Using Real-Time Firebase Database. International
Conference on Intelligent Human Computer Inter-
action, Lecture Notes in Computer Science,12616, 3-
20: In this paper, we propose a real-time IoT (Internet
of Things) system with a built-in hardware security
function, allows the ambient environment monitoring
and control actuators via a wireless network at the
frequency of 433 MHz, a real-time database is used to
store and share environmental sensors data and users
with a proposed Mobile App can send commands to
turn on/off actuators through the Internet.
IV. P ROBLEM S TATEMENT
In daily operations related to farming watering plays a major
role irrelative to the climatic conditions, it might be either too
hot and dry or too cloudy and wet , we have to be able to
control the amount of water that reach your plant. To water
the fields in small scale would be way easier than the larger
scale fields. In the large scale fields watering would involve
large amount of labour work, which eventually increases in
the expenses for the farmers. Even though we spend money,
insufficient amount of water would be supplied to each plant
by manually. So our project would help to overcome this
problem (Manual work and cost) by detecting or sensing the
humidity scale to be present in that particular temperature zone
as per predefined records from the plant’s water need, and
this data would be stored in the cloud named firebase and
our program would help to supply water to the zones where
the humidity levels are low as per the records stored in our
firebase.
V. P ROPOSED M ETHODOLOGY
This project aims to deliver real time application by
simplifying the lives of the people whose occupation comes
under agriculture by automating the irrigation system. The
proposed system delivers a way to supply water to the
agricultural field automatically. The system analyses the
moisture content and the rainfall and calculates the amount
of water to be released to the field. we use Gaussian NB
To predict rain the Soil Moisture sensor and Rain sensor
checks the soil moisture and density of rain. Based on these
values the system will be switched on/off. And real time
humidity, temperature and rain data is sent to cloud (Firebase).
As when we are going to make the first prediction and
after that we will be monitoring the temperature levels and
the humidity levels and note down the rain occurance and
these data will be included to Firebase and then after some
days we will have our own data available at the Firebase. As
when we are going to find the rain occurrence the supervised
learning algorithms helps a lot but, in this scenario, we cannot
use the regression model so we are going with the Gaussian
NB model through which we can be able to predict the rainfall.
A local weather station is created to predict the occurrence
of rain by giving humidity and temperature, Humidity can
be predicted from rain and temperature and temperature can
be predicted from rain and humidity. This helps us to stop
providing the water to the crop as there could be notification
when there is an occurance of rainfall.
We will be using the sensor data that has been stored
in the Firebase for the rain prediction after the rainfall we
check the moisture of the soil if the rain is not sufficient,
we will provide the crop with water if it has the abundant
amount of moisture, we will wait till it drops below required
level of moisture and then go for irrigation again. As we are
going to work on a field here, we will make a grid of size 1
Feet x 1 Feet and then arrange the sensors as such after that
we are set and then we provide the water to less moisture
areas.
A. Description of the Components used
The NODE MCU module is connected to Arduino which
helps in passing the messages. Humidity sensor is connected
to Arduino which tells about the climatic conditions. Soil
moisture sensor is connected to Arduino which tells about
the water moisture level in the soil that helps in watering
the field with sufficient amount of water. Water motor is
connected which helps in pumping the water when request is
given. Rain sensor helps to weathercast when the rain comes
and stops water motor from watering the field.
Arduino Board: Arduino is an open-source computer
hardware and software company, project and user community
that designs and manufactures micro-controller based kits for
building digital devices and interactive objects that can sense
and control the physical world.It uses various 8-bit or 32-bit
microprocessors. These systems provide sets of digital and
analog I/O pins that can be interfaced to various expansion
boards (”shields”) and other circuits.There are various types
of Arduino board. In this project we are using Arduino UNO.
Soil Moisture Sensor: Soil moisture sensors measure
the volumetric water content in soil. Since the direct
gravimetric measurement of free soil moisture requires
removing, drying, and weighting of a sample, soil moisture
sensors measure the volumetric water content indirectly by
using some other property of the soil, such as electrical
resistance, dielectric constant, or interaction with neutrons,
as a proxy for the moisture content. The relation between
the measured property and soil moisture must be calibrated
and may vary depending on environmental factors such as
soil type, temperature, or electric conductivity. Reflected
microwave radiation is affected by the soil moisture and
is used for remote sensing in hydrology and agriculture.
Portable probe instruments can be used by farmers or
gardeners.Soil moisture sensors typically refer to sensors that
estimate volumetric water content. Another class of sensors
measure another property of moisture in soils called water
potential; these sensors are usually referred to as soil water
potential sensors and include tensiometers and gypsum blocks.
Humidity Sensor: A humidity sensor senses, measures
and reports both moisture and air temperature. The ratio of
moisture in the air to the highest amount of moisture at a
particular air temperature is called relative humidity. Relative
humidity becomes an important factor, when looking for
comfort. Humidity sensors work by detecting changes that
alter electrical currents or temperature in the air. There are
three basic types of humidity sensors: capacitive,resistive and
thermal. All three types of sensors monitor minute changes in
the atmosphere in order to calculate the humidity in the air.
Rain Sensor: A rain sensor is a switching device activated
by rainfall. There are two main applications for rain sensors.
The first is a connected to an automatic irrigation system
that causes the system to shut down in the event of rainfall.
The second is a device used to protect the interior of an
automobile from rain and to support the automatic mode of
windscreen wipers. An additional application in professional
satellite communications antennas is to trigger a rain blower
on the aperture of the antenna feed, to remove water droplets
from the Mylar cover that keeps pressurized and dry air
inside the wave-guides.
2 Channel Relay for Arduino: This is a 5V 2-Channels
Relay module, it can be controlled directly by a wide range
of microcontrollers such as Arduino, AVR, PIC, ARM and
MSP430. 2 relays are included in this module, with “NC”
ports means “Normally connected to COM” and “NO” ports
means “Normally open to COM”. It is also equipped with 4
LEDS to show the status of relays.
Jumper Wires: A jump wire is an electrical wire or
group of them in a cable with a connector or pin at each end,
which is normally used to interconnect the components of a
breadboard or other prototype or test circuit, internally or with
other equipment or components, without soldering. Individual
jump wires are fitted by inserting their ”end connectors” into
the slots provided in a breadboard, the header connector of a
circuit board, or a piece of test equipment. Individual jump
wires are fitted by inserting their ”end connectors” into the
slots provided in a breadboard, the header connector of a
circuit board, or a piece of test equipment.
Water Motor (With Ac Pump): It is used to pump
water from the reservoir or tank to irrigate the plant.
VI. P LATFORM AND T OOLS
• Arduino to controll all the components by automated
commands
• Firebase as the database to store the sensor data
• Spyder IDE to run our Machine Leanring model to
predict rainfall
• Flask to develop web interface for the user
• python programming language
 • Moisture level Sensors to sense the moisture levels of 7) Process Specification
the soil
• Rain Sensor, Humidity sensor, Temperature Sensor for
recording data and to make rain prediction
• Motor to pump water to the fields

VII. OVERVIEW A ND P LANNING

A. Challenges

• The flow of information to and from the centralized

gateway (here, the microcontroller) has to be supported 8) Domain Model Specification
by stable internet services which can be difficult to be
found in some argricultural areas.
• A smart phone is required to receive SMS about the
irrigation schedules.
• The system has to be ON all the time and the farmer
must have awareness of basics of the system.

B. Architecture Specifications

• The NODE MCU module is connected to Arduino which

helps in passing the messages.
• Humidity sensor is connected to Arduino which tells
about the climatic conditions.
• Soil moisture sensor is connected to Arduino which tells
about the water moisture level in the soil that helps in
watering the field with sufficient amount of water.
• Water motor is connected to Arduino which triggers when
the irrigation is needed and supply the water to the field.

9) Information Model Specification

C. IOT Design Methodologies

1) Purpose: A Smart Irrigation System that allows op-

timization of water provided with accurate watering
schedules for the plants.
2) Behaviour: The system measures the amount of mois-
ture in the soil and predicts the possibility of rain and
gives the schedules.
3) Data Analysis Requirements: Data (Humidity and
temperature) is sent to ThingSpeak, data is visualized
here(Using matlab we point the data), then the data is
taken to the computer and analysis is performed here to
predict possibility of rain.
4) System Management Requirements: The system
should be ON all the time, provide remote monitoring
and control functions. 10) Service Model Specification
5) Security Requirements: Basic authentication capability
is required.
6) User Interface: System has to be deployed on a device
locally, notifications regarding the watering schedules
are received by this device.
 Serial.println(”rain analog”);
Serial.println(analogRead(A0));*/
int sensorReading = analogRead(A2);
int range = map(sensorReading, sensorMin, sensorMax, 0, 3);
Serial.println(range);
// range value:
switch (range) {
case 0: // Sensor getting wet
//Serial.println(”Raininggggggggggggggg”);
break;
case 1: // Sensor getting wet
//Serial.println(”Drizzleeeeeeeeeeeeee”);
VIII. I MPLEMENTATION
break;
A. Aurdino Code case 2: // Sensor dry
//Libraries //Serial.println(”Nooooooooooooooooooo
#include DHT.h/; Rainnnnnnnnnnnnnnnnnnnnnnnnn”);
//DHT break;
#define DHTPIN 7 // what pin we’re connected to }
#define DHTTYPE DHT22 // DHT 22 (AM2302) //moisture value= analogRead(sensor pin);
DHT dht(DHTPIN, DHTTYPE); //// Initialize DHT sensor // moisture value = map(moisture value,550,0,0,100);
for normal 16mhz Arduino //Serial.print(”Mositure : ”);
//Variables //Serial.print(moisture value);
int chk; //Serial.println(”%”);
float hum; //Stores humidity value val = digitalRead(8); //Read data from soil moisture sensor
float temp; //Stores temperature value //– Serial.println(”soil moisture”);
//Soilmoisture //– Serial.println(val);
int sensor pin = A0; if(val == LOW)
int ACWATERPUMP = 13; {
int moisture value ; //– Serial.println(”offffffffffffffffffffffffffffffffffffffffffffffffff”);
int sensor = 8; digitalWrite(13,LOW); //if soil moisture sensor provides
int val; LOW value send LOW value to relay
const int sensorMin = 0; }
const int sensorMax = 1024; else
void setup() { {
Serial.begin(9600); //– Serial.println(”onnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnnn”);
pinMode(8,INPUT); digitalWrite(13,HIGH); //if soil moisture sensor provides
pinMode(7,INPUT); HIGH value send HIGH value to relay
pinMode(A0,INPUT); }
pinMode(13,OUTPUT); //Set pin 13 as OUTPUT pin, to send delay(1000);}
signal to relay
dht.begin();
}
void loop(){
//Read data and store it to variables hum and temp
hum = dht.readHumidity();
temp= dht.readTemperature();
//Print temp and humidity values to serial monitor
//Serial.print(”Humidity: ”);
Serial.print(hum);
Serial.print(” ”);
//Serial.print(” %, Temp: ”);
Serial.print(temp);
Serial.print(” ”);
//Serial.println(” Celsius”);
/*Serial.println(”rain digi”);
Serial.println(digitalRead(7));
B. Aurdino Code To Upload Sensor Data To Firebase

IX. B ENCHMARKING I NNOVATIONS

A. System-01
In one of the existing systems the data for the system is
collected from weather stations i.e., wind, temperature, solar
radiation, humidity and rainfall from one or other system.
It takes time to collect data from the stations and feed it to
the system.To avoid this problem we are going to use a soil
moisture sensor to collect data and directly feed it to the
system, this is one of the main advantages of the planned
system.
https://patents.google.com/patent/US7403840B2/en?
q=PLANT+WATERING&q=SYSTEM&oq=PLANT+
WATERING+SYSTEM
B. System-02
In this existing system hundreds of plotted plants rapidly
get amounts of water, with the supervision of nursery
employees. Sometimes the employees may not be available
every time. Or they might have been mistake. So to overcome
this problem in our system we have given option for regular
watering as well as for interval planting. So the crops or
plants might not get affected or died.
https://patents.google.com/patent/US3108400A/en?q=
PLANT+WATERING&q=SYSTEM&oq=+PLANT+
WATERING+SYSTEM/
C. System-03
In this existing system the watering takes place on range
based irrigation i.e. works in certain radius. Due to this there
may be problems like no proper connection and every part of
the area cannot be covered. To overcome this in our project
we have added NODE MCU module.
https://patents.google.com/patent/US20120036091A1/en?
q=automated&q=PLANT+WAERING&q=SYSTEM&oq=
+automated+PLANT+WATERING+SYSTEM
D. System-04
In this existing system there is no output to display the
amount of water is used till date and number of hours the
system is been running. The proposed system is equipped
with a digital LCD monitor that displays the amount of water
used for irrigating the area in specified units and time for
which the system is running.
https://www.google.co.in/patents/US4684920?dq=automated+
watering+system&hl=en&sa=X&ved=0ahUKEwjz6Of9m
vYAhXEso8KHaC1CvYQ6AEIQjAE

E. System-05
In this system the watering is done by directly measuring
the moisture content of the soil there is no specific operation
to stop or start watering the area. The moisture is measured from the above image we can say that our humidity sensor
by the moisture sensor and sent to the system for operation. and the temperature sensor are working fine as per the data
In the proposed system we will be providing an option to stored in the firebase. Now we have the storage part and done
set intervals of watering the system or manual on/off the and now here we have to check how the rain varies with
system via a mobile. The instructions are sent to the system temperature and all.
by accessing the NODE MCU module.
https://www.google.co.in/patents/US6079433?dq=automated+
watering+system&hl=en&sa=X&ved=0ahUKEwjz6Of9m
vYAhXEso8KHaC1CvYQ6AEIUDAG

X. D ESCRIPTION OF T EST C ASES

Here the main test cases include the functionality of our

sensors and how reliable are they. so here we step out to test
the connection of our sensor to the Firebase and see if the now lets see the variation of rain under various circumstances.
data is generated or not and the if the data is generated then
we have to test it for the data is stored in the Firebase or not.

A. Description of Test Case-01

here we will be first going trough the set up of the sensors
and their connection.

B. Description of Test Case-02

from the above we can know that all the sensors are
working fine and all the data that have been generated is
stored to the firebase now here we have to talk about the
web interface developed for the user puprose that whill help
him find the value of humidity or temperature or the rainfall
when other two are given as input

after all the connections are made then we can say that our
mini weather station and the irrigating system has been
built and now we will be going to use the sensor to store
the data to do that we will use the firebase as the storage
and check if any data has been recorded in the database or not.

The graphs that we had seen in the above test case are the
part of the output of the python machine learning model
that runs in the background of this flask interface which is
responsible for the rain prediction and also from the above
two test cases we can say that we have the pretty good results
so that we can relay on our model to make it work on large
scale.
XI. R ESULTS AND D ISCUSSION
Here the approach we take to get the data about the rainfall
is different as a part of data collection here we will be
generating our own data from the sensors we c=will setup
the system at the workplace ahead of the installation the
irrigation system which in turn stores the daily humidity and
the temperature and the prediction is made using the rain
sensor we get to know our model prediction is correct or
wrong when it is correct the data will be pushed to firebase
and when the prediction is wrong we will alter the rain column
by replacing with correct value and push the data into firebase
as time goes by we will have lots of data with accurate values
and we will be able to attain the accurate results on the rainfall
prediction.
Other than the rainfall prediction here we will be using the
moisture sensor which worked pretty good in the small scale
experiments when going with the large scale applications we
must have to go with the different type of moisture sensor that
could help us do the work accurately. Here when we has done
this expirement we had the all the components working and
we can say that the Aurdino is the heart of the system which
gives commands to the motor when to irrigate and when not
to and also it manages the data that have been generated by
the sensor and the duty of uploading the data to the firebase
one of the major duty that the aurdino does. Finnaly, we can
that the developed system worked will with the experiments
and had shown pretty good results by which we can have a
confidence to headstart a large scale system production.
XII. C ONCLUSION AND F UTURE S COPE
We can use this product in agricultural fields for watering
the plants at the regular interval of time, and to know when
the plants are to be watered. This product can also be used
at house for feeding pet animals or plants at regular time
intervals by the house owners. This product can also tell you
the climatic conditions and weather reporting so you can come
to know when to water the field and when it’s not needed. This
information is helpful for both farmers and house owners. In
future we develop this product in such a way that it can be
used in large scale and we minimize the product size using
the nano technology concepts.
R EFERENCES
[1] Vishal Saraswat and Rohith Gandhi, Ananth Rama Raju M,Rajath Raj
guru A (2020). Smart irrigation system using machine learning. Inter-
national Journal of Progressive Research in Science and Engineering,
1(3),186-189.
[2] Nalluri Sravani, Somula Ramasubbareddy and G. Kannayaram,(2019).
Weather prediction using clustering strategies in machine learning.
Journal of Computational and Theoretical Nanoscience,16(5/6), 1977-
1981
[3] Pratap N Kashyap, Preetham M S ,Rohan Venkatesh Nayak, Uday B,
and Radhika T V,(2020).Intelligent Weather Forecasting Using Machine
Techniques. International Research Journal of Engineering and Technol-
ogy, 7(3).
[4] Pradeep Hewage, Marcello Trovati, Ella Pereria, Ardhendu Behera,
(2020). Deep Learning-based effective fine-graied Weather Forecasting
model. Pattern Analysis and Applications,24,343-366.
[5] Anisha Datta, Shukrity Si, and Sanket Biswas,(2020). Complete Sta-
tistical Analysis to Weather Forecasting. Computational Intelligence in
Pattern Recognition, Book Chapter, 751-763.
[6] Piyush Kapoor and Sarabjeet Singh Bedi,(2013). Weather forecasting
using sliding window algorithm. ISRN Signal Processing, 2013(1), 1-5.
[7] Sudhan Murugan Bhagavathi, Anitha Thavasimuthu, Aruna Murugesan,
Charlyn Pushpa Latha George Rajendran, Vijay A, Laxmi Raja, and
Rajendran Thavasimuthu,(2021). Weather forecasting and prediction
using hybrid C5. 0 machine learning algorithm. International Journal
of Communication Systems, 34(1).
[8] Sam Cramer, Michael Kampouridis, Alex. A. Freitas and Antonis K.
Alexandridis,(2017). An extensive evaluation of seven machine learning
methods for rainfall prediction in weather derivatives. Journal of Expert
Systems with Applications, 85, 169-181.
[9] Wei-Chiang Hong,(2008). Rainfall forecasting by technological machine
learning models. Journal of Applied Mathematics and Computation,
200(1), 41-57.
[10] Gowtham Sethupathi. M, Yenugudhati Sai Ganesh and Mohammad
Mansoor Ali,(2021) Efficient Rainfall Prediction and Analysis using
Machine Learning Techniques. Turkish Journal of Computer and Math-
ematics Education (TURCOMAT), 12(6), 3467-3474.
[11] Fredy Angulo, Elsa Urueta, Gustavo Valverde and Obert Paternina.
O,(2017). Cartagena’s Water Distribution System. International Journal
of Procedia Engineering, 186(1), 28-35.
[12] A. M. K. El-Ghonemy,(2017). Performance test of a sea water multi-
stage flash distillation plant. Alexandria Engineering Journal, 57(4),
2401-2413.
[13] Katsunori Mizuno, Xiaofei Liu, Fuyuki Katase, Akira Asada, Makato
Murakoshi, Yosunobu Yagita, Yoshiaki Watanabe, Tetsuo Shimada
and Yasufumi Fujimoto, (2016). Automatic non-destructive three-
dimensional acoustic coring system for in site detection of aquatic plant
root under the water bottom. Journal of Nondestructive Testing and
Evaluation, 5, 1-8.
[14] H. Etienne and M. Berthouly, (2002). Temporary immersion systems
in plant micropropagation. Plant Cell, Journal of Plant Cell,Tissue and
Organ Culture, 69, 215-231.
[15] Krishna S.Nemali and Marc W. van Iersel, (2006). An automated system
for controlling drought stress and irrigation in potted plants. Scientia
Horticulturae, 110(3), 292-297.
[16] Imma Farran and Angel M.Mingo-Castel,(2006). Potato minituber pro-
duction using aeroponics: effect of plant density and harvesting intervals.
American Journal of Potato Research, 83, 47-53.
[17] Abraham.N, Hema.P.S, Saritha.E.K and Subramannian S,(2000). Irriga-
tion automation based on soil electrical conductivity and leaf tempera-
ture. Agricultural Water Management, 45(2), 145-157.
[18] M. Kacira and P. P. Ling, (2001). Design and development of an
automated and Non–contact sensing system for continuous monitoring
of plant health and growth. Transactions of the ASAE, 44(4), 989-996.
[19] Pavithra, D. S and M. S. Srinath,(2014). GSM based automatic irrigation
control system for efficient use of resources and crop planning by using
an Android mobile. IOSR Journal of Mechanical and Civil Engineering,
11(4), 49-55.
[20] J. Rockstrom,(2000). Water resources management in smallholder farms
in Eastern and Southern Africa. Journal of Physics and Chemistry of the
Earth, part B: Hydrology, Oceans and Atmosphere, 25(3), 275-283.
[21] Bernard Cardenas-Lailhacar and Michael D. Dukes,(2008). Expanding
disk rain sensor performance and potential irrigation water savings.
Journal of Irrigation and Drainage Engineering, 134(1), 67-73.
[22] Leah Meeks, Michael D Dukes, Kati White Migliaccio and Bernard
Cardenas-Lailhacar, (2012). Long term expanding-disk rain sensor ac-
curacy. Journal of irrigation and drainage engineering, 138(1), 16-20.
[23] Yudi Irawan Chandra, Marti. Riastuti, Kosdiana and Edo P. Nugroho,
(2020). Automatic Garden Umbrella Prototype with Light and Rain
Sensor Based on Arduino Uno Microcontroller. International Journal
of Intelligent Robotics, 2(2), 42-51.
[24] Gad I and Hosahalli D, (2020). A comparative study of prediction and
classification models on NCDC weather data. International Journal of
Computers and Applications,1-12.
[25] Bahzad Charbuty and Adnan Mohsin Abdulazeez, (2021). Classification
Based on Decision Tree Algorithm for Machine Learning. Journal of
Applied Science And Technology Trends, 2(1), 20-28.
[26] Saurav Barua,(2020). A Naı̈ve Bayes Classifier Approach to Incorpo-
rate Weather to Predict Congestion at Intersections. World Academics
Journal of Engineering Sciences, 7(2),72-76.
[27] C. Karthikeyana, Gurram Sunithab, J. Avanijac, K. Reddy Madhavid
and E.S. Madhane,(2021). Prediction of Climate Change using SVM
and Naı̈ve Bayes Machine Learning Algorithms. Turkish Journal of
Computer and Mathematics Education,12(2),2134-2139.
[28] Nti.I.K, Aning.J, Ayawli.B.B.K, Frimpong.K, Appiah.A.Y and Nyarko-
Boateng.O, (2021). A Comparative Theoretical and Empirical Analysis
of Machine Learning Algorithms. Webology, 17(1).
[29] Raju N A, (2015). Land capability and suitability in Vizianagaram
district of Andhra Pradesh using remote sensing and GIS techniques.
IOSR J of Humanities and Social Sci. (IOSR-JHSS),20,56-64.
[30] Jaroslav Knápek , Tomas Králı́k , Kamila Vávrová and Jan Weger,(2020).
Dynamic biomass potential from agricultural land. Renewable and
Sustainable Energy Reviews, 134.
[31] Yikuan Tang, Fan Zhang, Bernard A. Engel, Xiao Liu, Qiong Yue and
Ping Guo, (2020). Grid-scale agricultural land and water management.
Journal of Cleaner Production, 265.
[32] Elnashar.A, Abbas.M, Sobhy.H and Shahba.M, (2021). Crop Water
Requirements and Suitability Assessment in Arid Environments. Agron-
omy,11(2),260.
[33] Suraj Kumar, Thendiyath Roshni, Ercan Kahya and Mohammad Ali
Ghorbani,(2020). Climate change projections of rainfall and its impact
on the cropland suitability for rice and wheat crops in the Sone river
command, Bihar. Journal of Theoretical and Applied Climatology, 142,
433-451.
[34] A Gaddam, M Al-Hrooby and W F Esmael, (2020). Designing a wireless
sensors network for monitoring and predicting droughts. International
Journal on Smart Sensing and Intelligent Systems, 7(5), 1-6.
[35] Basudeba Behera, Smita Kumari, Alisha Kumari and Ajay Kumar,
(2020). Application of IoT and Weather Prediction for Enhancement of
Agricultural Productivity. In International Conference on Computational
Intelligence, Security and Internet of Things,215-230.
[36] Sk. Fahmida Islam, Morium Akter and Mohammad Shorif Uddin,
(2021). Design and implementation of an internet of things based low-
cost smart weather prediction system. International Journal of Informa-
tion Technology, 13, 2001-2010
[37] Vivtor Mokaya, (2019).Future of Precision Agriculture in India using
machine learning and artificial intelligence. Journal of Environment,
Science And Technology, 5(2),37-44.
[38] Alka Rani, Arti Kumari, Ram Narayan Singh and Kavita Kumari,(2021).
Climate-Smart Agriculture: An Integrated Approach for Attaining Agri-
cultural Sustainability. In Climate Change and Resilient Food Systems
Book, chapter 5, 141-189.
[39] H. Swapna Rekha,Janmenjoy Nayak and Bighnaraj Naik, (2020). Impact
of IoT in agriculture: advances and challenges. Internet of Things
Book,127-154.
[40] Giang Truong Le, Nihat Minh Tran and Thang Viet Tran, (2020). IoT
System for Monitoring a Large-Area Environment Sensors and Control
Actuators Using Real-Time Firebase Database. International Conference
on Intelligent Human Computer Interaction, Lecture Notes in Computer
Science,12616, 3-20.