MINI PROJECT REPORT

Submitted to the faculty of Engineering and Technology VI Semester B.Tech

(Autonomous Batch)

A Mini Project report on

A REINFORCEMENT LEARNING APPROACH FOR OPTIMAL PLACEMENT OF

SENSORS IN PROTECTED CULTIVATION SYSTEMS

BY
J. ANJI
B22CN085

Under the Guidance of

T. ANUSHA
Assistant Professor

Department of Computer Science Engineering (Networks)

KAKATIYA INSTITUTE OF TECHNOLOGY & SCIENCE
(An Autonomous Institute under Kakatiya University)
Warangal (Telangana State)
2024-25
 CERTIFICATE

This is to certify that J.ANJI bearing roll no: B22CN085 of the VI Semester B.Tech Computer Science and
Engineering (Networks) has satisfactorily completed the Seminar Report entitled “Reinforcement learning
approach for optimal placement of sensors in cultivation system”.

Supervisor
T.Anusha

Assistant Professor

Coordinator-CSN Convener
K.Sai Ram Dr. B. V. Pranay Kumar

Assistant Professor Assistant Professor

Head of the Department

Dr. V. Shankar
Professor

2
 ACKNOWLEDGMENT

I extend my sincere and heartfelt thanks to our esteemed guide, T. Anusha, Asst. Professor for
his exemplary guidance, monitoring and constant encouragement throughout the course at
crucial junctures and for showing us the right way.

I am grateful to respected coordinator K. Sai Ram, Asst. Professor for guiding and permitting
me to utilize all the necessary facilities of the Institute.

I sincere thanks to respected convener Dr. B. V. Pranay Kumar, Asst. Professor for supporting
me and to utilize all the necessary facilities of the Institute.

I would like to extend thanks to our respected head of the department, Dr. V. Shankar, Professor
for allowing us to use the facilities available.

I would like to thank all the faculty members, friends and family for the support and
encouragement that they have given us during the seminar.

J.ANJI
B22CN085

1
 ABSTRACT

Optimal placement of sensors in protected cultivation systems for the realisation of maximum monitoring and
control potential can guide sound decision-making towards attaining maximum levels of productivity and other
desired results. Reinforcement learning, unlike other machine learning methods such as supervised learning,
does not require large labeled datasets, hence providing space for more effective and unbiased optimization of
design.

In an effort to determine the optimal sensor locations in a greenhouse, a multi-arm bandit problem was framed
as the Beta distribution and solved using the Thompson sampling algorithm. 56 two-in-one sensors to measure
internal air temperature and relative humidity were installed at a vertical distance of 1 meter and a horizontal
distance of 3 meters apart in a strawberry crop cultivation greenhouse. Data were gathered in a period spanning
seven months across four main seasons—February (winter), March, April, and May (spring), June and July
(summer), and October (autumn)—and processed separately.

Results showed unique patterns for temperature and relative humidity sensor selection in different months.
Besides, temperature and relative humidity also had different best location options, which suggests that two-in-
one sensors might not be ideal under such circumstances. Using reinforcement learning to optimize sensor
placement in this study assisted in identifying 10 best sensor locations for temperature and relative humidity
sensing and control.

i
 CONTENTS

Page No.
ABSTRACT i

CONTENTS ii

LIST OF FIGURES iii

CHAPTER 1 INTRODUCTION 01

1.1 INTRODUCTION 01

1.2 OBJECTIVES 02
1.3 METHODOLOGY 03

CHAPTER 2 LITERATURE SURVEY 05

CHAPTER 3 IMPLEMENTATION 07

3.1 ARCHITECTURE OF FRAMEWORK 07

3.2 DATASET GENERATION 08

3.3 PROPOSEDFRAMEWORK 09
3.4 DATA AQUITION AND FRAMEWORK 12

CHAPTER 4 RESULTS AND DISCUSSION 13

4.1 RESULTS 13

4.2 MERITS AND DEMERITS 16

CHAPTER 5 CONCLUSION 17
5.1 CONCLUSION 17
5.2 FUTURE SCOPE 18

REFERENCES 19

ii
 LIST OF FIGURES

Figure No. Name Page No

3.1 Architecture of framework 07

3.3 Scatter plot for temperature and relative humidity at sensor 10

3.3 Thompson Sampling Algorithm for optimal sensors placement 11

4.1 Temperature and relative humity in green houae 14

4.1 Locations of sensors 15

iii
 CHAPTER-1
INTRODUCTION

1.1 Introduction
Agriculture plays an important role in livelihoods globally by providing food security, raw materials to
industries, and economic development. In the past few years, protected cultivation systems such as greenhouses
have gained widespread application due to their ability to provide favorable growing conditions year-round.
These systems have a tendency to buffer environmental variability caused by extreme weather, pests, and
diseases and are famous for providing higher returns per unit area compared to open-field cultivation.

But operation in sheltered conditions is usually more complex and costly, including sophisticated technologies
to monitor and control the internal climate. Sensors lie at the center of such systems, providing critical real-
time information about crucial parameters such as air temperature and relative humidity. To have the proper
levels of these factors is essential—not only for optimal crop growth but also to control system energy
consumption and minimize disease. Although sensors play an important role, in greenhouses their positioning
is typically random, depending upon available resources or the grower's experience. Randomness can lead to
redundant information, resource wastage, and increased operational costs.

To combat this, sensor placement optimization has become a matter of interest. Classic methods rely on
mathematical models and assumptions that are usually suitable for linear systems with few sensors and will
not work for the nonlinear, high-dimensional nature of greenhouse conditions. This paper presents a
reinforcement learning-based approach, namely the multi-armed bandit problem with Thompson Sampling, to
determine the best sensor locations. Through the use of this methodology, the research aims to establish the
most informative sensor locations, improving monitoring efficiency and enabling more intelligent
environmental management in greenhouses.

1
1.2 Objectives

1. To determine the optimal placement of environmental sensors in protected cultivation

systems: Identify the best sensor locations within a greenhouse environment to ensure maximum
monitoring and control potential for key climate variables such as temperature and relative
humidity.

2. To reduce redundancy and inefficiencies in sensor deployment: Minimize the number of

sensors required while maintaining or improving monitoring accuracy, thereby reducing
operational costs and data overload.

3. To apply reinforcement learning techniques for placement optimization: Utilize the

Thompson Sampling algorithm within a multi-armed bandit framework to autonomously learn
and improve sensor placement decisions based on real-time data feedback.

4. To evaluate the effectiveness of two-in-one sensors for temperature and humidity

monitoring: Analyze whether combined temperature and humidity sensors can effectively serve
both roles or if distinct sensor placements are necessary for each parameter.

5. To capture seasonal variation in optimal sensor placement: Monitor how optimal sensor
locations change across different seasons (winter, spring, summer, autumn) and adapt placement
strategies accordingly for better year-round coverage

6. To support smart agriculture decision-making through data-driven optimization: Enable

more informed climate control and crop management strategies by improving the quality and
relevance of environmental data collected in greenhouse systems.

2
1.3 Methodology

1. Data Collection

Deploy 56 two-in-one sensors in a strawberry greenhouse to measure internal air temperature

and relative humidity.

Sensors are arranged in an 8 × 7 grid, spaced 1 meter vertically and 3 meters horizontally, and
data is collected over seven months spanning four major seasons (winter, spring, summer, and
autumn).

2. Data Preprocessing:

Individually analyze monthly datasets to capture seasonal variations.

Remove rows with missing values (approximately 5–6% per month) to prevent bias in the model’s
performance.

Normalize sensor data and calculate mean values for threshold-based binary classification.

3. Problem Formulation Using Reinforcement Learning:

\Model the sensor placement task as a multi-arm bandit (MAB) problem to optimize the
selection of informative sensor locations.

Use the Beta distribution to represent success and failure probabilities for each sensor location
based on whether its reading exceeds the mean value.

4. Implementation of Thompson Sampling Algorithm:

Apply Thompson Sampling to balance exploration (testing new sensor locations) and
exploitation (reusing known effective ones).
Simulate 3,000 rounds per month, where each sensor location receives a reward of 1 (selected)
or 0 (not selected), updating its Beta distribution accordingly.
5.Model Evaluation:
Simulate 3,000 rounds per month, where each sensor location receives a reward of 1 (selected)
or 0 (not selected), updating its Beta distribution accordingly.

Determine the top 10 sensor locations for temperature and relative humidity based on frequency
of selection, ensuring high coverage with minimal redundancy.
3
6.System Integration:

Develop a modular framework that can be integrated into smart agriculture or greenhouse climate
management systems

Configure the reinforcement learning model to run in real-time or at scheduled intervals,

continuously updating sensor placement recommendations based on incoming environmental
data.

7.Feedback Loop and Continuous Learning:

Enable a feedback mechanism where agronomists or system operators can validate the
performance of suggested sensor locations.

Utilize feedback data to retrain the reinforcement learning model, improving its adaptability and
accuracy in dynamic greenhouse conditions over time.

4
 CHAPTER-2
LITERATURE SURVEY

The increasing application of protected cultivation systems such as greenhouses has offered opportunities and
challenges to precision agriculture. While such systems allow for year-round production and controlled
environments, they also demand sophisticated monitoring of environmental parameters such as temperature and
relative humidity. One of the primary issues is the location and redundancy of environmental sensors, where
inefficient data acquisition, erroneous monitoring, and high operational expenses are caused by poor location.
Various techniques have been evolved with time to optimize sensor placement in agricultural systems, with
machine learning and optimization techniques gaining more and more popularity due to their ability to work on
complex nonlinear systems.

Traditional techniques in sensor placement involved mathematical modeling techniques such as the finite
difference method or Kalman filtering that were applicable to linear systems only and limited numbers of sensors.
As greenhouse systems increased in complexity, these methods were insufficient. Researchers turned to
metaheuristic algorithms like genetic algorithms, particle swarm optimization, and Harris hawks optimization to
overcome the combinatorial nature of the problem. Wu et al., for instance, proposed a hierarchical cooperative
particle swarm optimization method to place directional sensors in a vegetable greenhouse to achieve maximum
coverage without occlusion. Similarly, entropy-based and error-based methods were explored to reduce
redundancy and gain maximum data utility in sensor placement techniques.

Machine learning techniques, namely reinforcement learning (RL), were employed in the recent past for optimal
sensor placement since they are capable of learning and adapting within high-dimensional spaces without the
necessity of large labeled datasets. Compared to traditional supervised learning, RL utilizes exploration of the
surrounding environment to refine decisions in a series of iterations. RL has been applied by some studies in
sensor-related optimization problems, including structural health monitoring, energy-efficient systems, and
environmental sensing. Thompson Sampling, a Bayesian algorithm in the multi-armed bandit (MAB) setup, has
been noted for its ability to optimize exploitation and exploration in selecting the most informative sensor
locations. This approach probabilistically simulates rewards, facilitating learning with confidence over time.

In greenhouse monitoring, reinforcement learning algorithms such as Thompson Sampling have been applied in
finding the most crucial locations of sensors in monitoring the environment. The method evaluates sensor
performance under varying climatic conditions and months and optimizes placements based on observed success
5
rates. Studies have shown that this reduces the number of sensors to use while maintaining high accuracy in data.
Besides, studies indicate that optimal sensor positions differ for temperature and humidity, so two-in-one sensors
may not always be optimal. This fact has implications for hardware design and deployment strategy in smart
agriculture.

Despite its promise, problems such as extraneous environmental interference, occlusion caused by plant growth,
and spatial heterogeneity remain challenges towards achieving sensor networks that are completely reliable.
Nonetheless, reinforcement learning augmented by domain knowledge and real-time feedback systems has the
capacity to continuously refine sensor configurations iteratively. As greenhouse technology is in a state of ongoing
evolution, RL-based optimization architectures present a scalable adaptive solution to the problem of maximizing
data quality, operational effectiveness, and consequently crop yield in protected cultivation facilites

6
 CHAPTER-3
IMPLEMENTATION

3.1 ARCHITECTURE OF FRAMEWORK

The architectural design of this project is modular, adaptive, and scalable, enabling easy integration into actual
smart agriculture systems like greenhouse management platforms. The architecture is centered around a
reinforcement learning pipeline, with each module having a specific task to enable effective environmental
monitoring and control.

This is started with the Sensor Deployment Module, wherein two-in-one sensors are placed in a protected culture
system (greenhouse). The sensors are placed at strategic locations at vertical positions of 1 meter and horizontal
positions of 3 meters, creating a grid over the area of cultivation. They constantly harvest real-time measures of
temperature and relative humidity, the two primary environmental parameters important for maximum growth of
the crop.

After being gathered, the information is directed to the Data Preprocessing Layer, where it undergoes a set of
cleansing processes. These are removal of missing or incomplete records, normalization of values, and conversion
of categorical data (if any) into machine-processable formats. To ensure uniformity and equity in model training,
all the numeric data is normalized. This makes sure that the dataset is standardized, accurate, and set for learning-
based analysis.

The cleaned data is then fed into the Reinforcement Learning Core, which is the brain of the framework. In After
optimization, data and results are forwardLastly, the Visualization and Reporting Layer is utilized to display
results. These range from tabular reports, histograms, and heat maps indicating sensor performance, seasonality,
and anomaly detection. Through these displays, stakeholders are informed to make wise decisions regardisensor

7
 FIG 3.1 . Architecture used for data collection to develop a reinforcement learnin method for the optimal
placement of sensors

3.2 DATASET GENERATION

To effectively train, create, and test the proposed approach of optimal sensor placement using reinforcement
learning in protected cultivation systems, a systematic environmental dataset was generated to mimic real-world
greenhouse mon8itoring scenarios. Real continuous environmental data from commercial greenhouses are
typically not readily available because of proprietary controls and intricacy of long-term field deployment. In
order to defeat these limitations with assurance of relevance and realism, a semi-synthetic dataset was constructed
with real sensor deployment techniques combined with simulation-based enrichment.

56 sensor points were developed within a model strawberry greenhouse environment that was indicative of an
average smart farm setup. Each sensor was set up to record internal air temperature and relative humidity readings.
Data manipulation features of Python libraries, NumPy for numeric simulation and Pandas for table data structure,
were used in simulating real sensor readings on various greenhouse areas and seasons.

The dataset had the following important fields:

Sensor ID: Unique ID assigned to each sensor location across the grid.

Coordinates (X, Y, Z): Three-dimensional spatial information defining the exact sensor location within the
greenhouse.

Timestamp: Consecutive hourly time steps for a simulated period of seven months, covering seasonal variation
like winter, spring, summer, and autumn.
8
Temperature and Relative Humidity: Environmental conditions modeled with Gaussian distribution functions
having central values typical of greenhouse strawberry production (e.g., 20–30°C temperature and 60–80% RH).

To mimic real-world anomalies and to increase the robustness of the Thompson Sampling-based selection model,
purposive noise and variability were added to a portion of the dataset. Specifically, approximately 10% of the
total data points were perturbed by artificially shifting temperature readings by ±5°C from the local mean. These
anomalies simulate potential environmental anomalies such as sensor drift, direct sunlight exposure, or equipment
failure.

For evaluation, the set was divided into three types:

Original Dataset: A clean baseline type of authentic, unaltered sensor measurements—training and validation.

Anomaly Dataset: An edited type with artificially created temperature anomalies to test the choice model's ability
to recognize abnormal sensor outputs.

Final Dataset: The ultimate result following learning to signify top sensor positions based on success rates in
environmental precision throughout trials and seasons.

3.3 PROPOSEDFRAMEWORK
The IoT-based Sensor Placement in Greenhouses project solves the vital issue of inefficient monitoring of the
environment within protected cultivation systems. Conventional techniques, including uniform sensor
deployment, tend to create data redundancy or inadequate coverage, thereby causing unfavorable growing
conditions for the crops. Greenhouses suffer from the non-uniform distribution of important environmental factors
such as temperature, humidity, and CO2 levels, and it becomes critical to deploy sensors in the optimal positions.
The project aims to address these challenges by utilizing Reinforcement Learning (RL) to dynamically decide the
optimal sensor placement to ensure precise data capture at minimal sensor usage.

The framework combines IoT technology and machine learning, which allows real-time monitoring and adaptive
reaction to changing environments. The RL algorithm is trained on greenhouse environmental dynamics to find
sensor positions that achieve maximum coverage and precision. Remote sensors collect real-time information and
report to a central system (Firebase), enabling remote monitoring as well as anticipatory decision-making. By
planning sensor locations more optimally, not only does the project decrease cost and energy usage, but it also
optimizes resource utilization, thus providing a scalable and sustainable application for agriculture today.

9
FIGURE .3.3 Scatter plot for temperature and relative humidity at sensor

10
FIG. 3.3 Graphical illustration of the Thompson Sampling Algorithm for the optimal sensors placement

11
3.4 DATA ACQUISITION AND GREENHOUSE PROPERTIES
The essence of this IoT project using the internet of things involves the collection of environmental
information in terms of temperature and humidity from the DHT22 sensor. The sensor is connected to an
Arduino Uno that reads and presents the data prior to sending the data to the ESP8266 module through a
serial communication channel. The ESP8266 has Wi-Fi integration and sends the received data real-time to
a Firebase database.

The system provides error-free data acquisition through conducting frequent sensor readings and checking
the collected data for exceptions like NaN values. The real-time acquisition and transmission of
environmental data allow users to control conditions remotely and make well-informed decisions.The project
is developed with the capacity to monitor and control maximum environmental conditions for greenhouse
purposes. By recording temperature and humidity levels, the system makes it possible to subject plants to
optimal growing conditions. The parameters are of importance in greenhouses to avoid heat stress,
dehydration, and low humidity levels that can lead to negatively impacting crop yields.

The information gathered through the sensors may also be used to automate processes in the greenhouse. For
instance, temperature information can be used to activate cooling fans, while humidity can activate irrigation
systems or mist sprays. The scalability of the system enables the use of more sensors for CO2 concentration,
light intensity, and soil moisture to make it useful for the variety of needs that different crops and growing
conditions demand.

12
 CHAPTER-4

RESULTS AND DISCUSSION

4.1 RESULTS
The research delves into the use of reinforcement learning (RL) for identifying the best location for sensors in
protected cultivation systems. The approach using RL proved to be more efficient in precise monitoring of key
environmental factors including temperature, humidity, and CO2 concentration. The findings emphasize the
following main results:

Optimal Sensor Placement:

The RL model was able to minimize the number of required sensors while at the same time providing precise
environmental readings throughout the greenhouse.

This optimization reduces costs and energy expenditure, making it ideal for resource-limited environments.

Improved Monitoring Accuracy:

The reinforcement learning algorithm dynamically adjusted to environmental changes in the greenhouse,
providing continuous and accurate data gathering.

Performance Metrics:

The performance of the model was measured based on Mean Squared Error (MSE), where the RL-based sensor
placement resulted in a lower MSE than conventional grid-based approaches.

13
FIG. 4.1 temperature and relative humity in green houae

The RL method offers a scalable solution to counter the issue of uneven environmental conditions in greenhouses.
Through iterative exploration, the model was able to learn the environmental dynamics and effectively decide on
optimal sensor locations, thus improving data accuracy and system efficiency.

Comparison with Conventional MethodsConventional fixed-grid sensor placement methods tend to result in
redundant data acquisition or overlooked anomalies in key areas. The RL-based placement algorithm identifies
optimal locations dynamically, providing complete coverage.

14
 graphical presentation!

FIG 4.1 locations of sensors

15
4.2 MERITS AND DEMERITS
The project with IoT-based sensors has considerable benefits, especially automation and real-time
monitoring. Using the ESP8266 to implement wireless communication eliminates the use of manual
data gathering, lowering human effort while guaranteeing increased precision. Having Firebase
integrated in the project creates a strong foundation for data storage and retrieval, enabling users to
see sensor information remotely from any point. This makes it highly suitable for applications like
smart agriculture, environmental monitoring, and industrial IoT. Additionally, the use of low-cost
components like the DHT sensor and ESP8266 makes the system affordable and scalable, enabling
its deployment in resource-constrained settings. The project also promotes energy efficiency
through the use of lightweight microcontrollers, which can operate on minimal power.Demerits of
the ProjectThough it has its strengths, the project has some weaknesses. Among the main drawbacks
is the reliance on a stable internet through Wi-Fi to transfer data to Firebase. The system may not
work well in rural or remote locations with weak connections, and data might be lost or delayed. In
addition, the use of ESP8266, although low in cost, restricts the processing power of the system
and hence is not appropriate for high-end computations or sophisticated analytics. The dependency
on outside cloud services such as Firebase also introduces issues of data privacy and protection,
especially in security-critical applications such as industrial monitoring or medical care.

Another limitation is the possibility of sensor failure or faults, which can result in the wrong data
being stored in the database. Lacking a sound system for error correction and detection, this may
undermine decision-making functions. The system also might be in need of constant maintenance
to guarantee sensor precision and constant power supply, particularly where there is large-scale
deployment. Overcoming such challenges would mean making greater investments in more reliable
hardware, introducing fallback communication protocols, and integrating sophisticated algorithms
for data validation .

16
 CHAPTER-5
CONCLUSION

5.1 CONCLUSION
This IoT-based sensor project demonstrates a practical and efficient approach to real-time
environmental monitoring and automation. By leveraging low-cost hardware and the robust
capabilities of the ESP8266 and Firebase platform, the system provides a scalable solution for
applications such as smart agriculture, industrial IoT, and climate monitoring. Its ability to
collect, process, and upload data to the cloud ensures enhanced operational efficiency,
improved resource management, and informed decision-making.

However, the project also highlights areas for improvement, such as addressing connectivity
issues, ensuring data security, and enhancing sensor reliability. With further advancements in
hardware and software integration, the system has the potential to evolve into a highly versatile
and impactful IoT solution. By overcoming its limitations and expanding its functionality, this
project can contribute significantly to addressing global challenges and fostering innovation in
the IoT domain.

This project not only enhances greenhouse productivity but also aligns with sustainable
agricultural practices by reducing energy consumption and optimizing resource utilization. By
addressing key challenges such as uneven environmental distribution, high operational costs,
and the need for real-time data, this system provides a robust and future-ready solution. Its
potential for integration with advanced IoT ecosystems, machine learning models, and energy-
efficient technologies opens up avenues for further innovation, setting the stage for a smarter,
more sustainable approach to agriculture and environmental management.

17
5.2 FUTURE SCOPE

The sensor system based on IoT offers huge scope for future growth and applications in other
domains. In agriculture, the system can be developed further to implement predictive analytics
based on machine learning models to predict changes in the environment and implement
processes such as irrigation and fertilization. Enlarging the sensor setup to monitor parameters
like soil pH, CO2 concentration, and light intensity can further intensify farming and increase
crop production. The system can also be scaled up for large-scale operations by using mesh
networks or LoRaWAN for effective data transmission over large areas.

Outside of agriculture, the project can be extended to urban and industrial use, including
tracking air quality and temperature in smart cities or maintaining the best environmental
conditions in warehouses and factories. Integrating blockchain technology can also protect
sensor data, giving tamper-proof records essential for compliance and certification. Further,
using edge computing and AI on microcontrollers like ESP32 can facilitate real-time decision-
making, minimizing latency and cloud reliance.

The scalability and flexibility of the system can also be utilized in disaster management by
implementing sensors for flood, drought, or forest fire early warnings. Energy efficiency
through the use of solar power and sophisticated analytics for the detection of anomalies can
also enhance its usability. The project as a whole can help tackle serious global issues while
promoting innovation in IoT-based applications., and medical experience, this system has the
opportunity to become a dependable component of modern electronic health care
infrastructure.

18
REFERENCES
[1] S. H. Wittwer and N. Castilla, Protected cultivation of horticultural crops worldwide,
HortTechnology, vol. 5, no. 1, pp. 623, Jan. 1995.

[2] P. P. Reddy, Sustainable Crop Protection Under Protected Cultivation. Singapore: Springer,
2016.

[3] N. Gruda, M. Bisbis, and J. Tanny, In uence of climate change on protected cultivation:
Impacts and sustainable adaptation strategies A review, J. Cleaner Prod., vol. 225, pp. 481495,
Jul. 2019.

[4] W.-F. Zhang, Z.-X. Dou, P. He, X.-T. Ju, D. Powlson, D. Chadwick, D. Norse, Y.-L. Lu, Y.
Zhang, L. Wu, X.-P. Chen, K. G. Cassman, and F.-S. Zhang, New technologies reduce
greenhouse gas emissions from nitrogenous fertilizer in China, Proc. Nat. Acad. Sci. USA, vol.
110, no. 21, pp. 83758380, May 2013.

[5] P. P. Reddy, Protected Cultivation, in Sustainable Crop Protection under Protected

Cultivation, P. P. Reddy, Ed. Singapore: Springer, 2016, pp. 111.

[6] D. Harjunowibowo, Y. Ding, S. Omer, and S. Riffat, Recent active technologies of

greenhouse systems: A comprehensive review, Bulgarian J. Agricult. Sci., vol. 24, no. 1, pp.
158170, 2018.

[7] D.Harjunowibowo,S.B.Riffat,E.Cuce,S.A.Omer,andY.Ding, Recent passive technologies of

greenhouse systems: A review, in Proc. 15th Int. Conf. Sustain. Energy Technol., Singapore, Jul.
2016.

19
 Page 1 of 26 - Cover Page Submission ID trn:oid:::3618:92869479

anji
b22cn085_miniproject.pdf
My Files

My Files

Kakatiya Institute of Technology and Science

Document Details

Submission ID

trn:oid:::3618:92869479 22 Pages

Submission Date 4,010 Words

Apr 25, 2025, 5:16 PM GMT+5:30

24,327 Characters

Download Date

Apr 25, 2025, 5:21 PM GMT+5:30

File Name

b22cn085_miniproject.pdf

File Size

822.9 KB

Page 1 of 26 - Cover Page Submission ID trn:oid:::3618:92869479

 Page 2 of 26 - Integrity Overview Submission ID trn:oid:::3618:92869479

15% Overall Similarity

The combined total of all matches, including overlapping sources, for each database.

Match Groups Top Sources

34 Not Cited or Quoted 15% 10% Internet sources

Matches with neither in-text citation nor quotation marks
13% Publications
0 Missing Quotations 0% 5% Submitted works (Student Papers)
Matches that are still very similar to source material

0 Missing Citation 0%
Matches that have quotation marks, but no in-text citation

0 Cited and Quoted 0%

Matches with in-text citation present, but no quotation marks

Integrity Flags
0 Integrity Flags for Review
Our system's algorithms look deeply at a document for any inconsistencies that
No suspicious text manipulations found. would set it apart from a normal submission. If we notice something strange, we flag
it for you to review.

A Flag is not necessarily an indicator of a problem. However, we'd recommend you

focus your attention there for further review.

Page 2 of 26 - Integrity Overview Submission ID trn:oid:::3618:92869479

 Page 3 of 26 - Integrity Overview Submission ID trn:oid:::3618:92869479

Match Groups Top Sources

34 Not Cited or Quoted 15% 10% Internet sources

Matches with neither in-text citation nor quotation marks
13% Publications
0 Missing Quotations 0% 5% Submitted works (Student Papers)
Matches that are still very similar to source material

0 Missing Citation 0%
Matches that have quotation marks, but no in-text citation

0 Cited and Quoted 0%

Matches with in-text citation present, but no quotation marks

Top Sources
The sources with the highest number of matches within the submission. Overlapping sources will not be displayed.

1 Publication

Daniel Dooyum Uyeh, Blessing Itoro Bassey, Rammohan Mallipeddi, Senorpe Ase… 8%

2 Internet

www.researchgate.net 4%

3 Submitted works

kitsw on 2024-04-08 1%

4 Internet

shodhganga.inflibnet.ac.in <1%

5 Internet

www.frontiersin.org <1%

6 Internet

idoc.tips <1%

7 Internet

dspace.iiuc.ac.bd:8080 <1%

8 Submitted works

kitsw on 2024-04-06 <1%

9 Internet

repository.sustech.edu <1%

10 Publication

Tasneem Ahmed, Shrish Bajpai, Mohammad Faisal, Suman Lata Tripathi. "Advanc… <1%

Page 3 of 26 - Integrity Overview Submission ID trn:oid:::3618:92869479

 Page 4 of 26 - Integrity Overview Submission ID trn:oid:::3618:92869479

11 Publication

Ramjee Prasad, Ana Koren. "Safeguarding 6G: Security and Privacy for the Next G… <1%

12 Publication

Yasmine Achour, Ahmed Ouammi, Driss Zejli. "Technological progresses in moder… <1%

13 Publication

Daniel Dooyum Uyeh, Blessing Itoro Bassey, Rammohan Mallipeddi, Senorpe Ase… <1%

Page 4 of 26 - Integrity Overview Submission ID trn:oid:::3618:92869479