REPUBLIC OF THE PHILIPPINES

Department of Education
REGION X – NORTHERN MINDANAO
SCHOOLS DIVISION OF BUKIDNON
QUEZON BUKIDNON COMPREHENSIVE NATIONAL HIGH
SCHOOL
Mibando, Poblacion, Quezon, Bukidnon

IoT-based Water Monitoring and

Purification Machine in Milusok

SYDNEY CASSANDRA B. LANADO

CHRISTIAN AMEER A. ALIWATE
NICOLE NATHALIE A. CLAROS
RITCHELMAE R. ARMISIN
RHIAN FRITZ C. UY

A SCIENCE INVESTIGATORY PROJECT SUBMITTED IN PARTIAL

FULFILLMENT OF THE REQUIREMENTS
FOR THE SUBJECT GRADE 10 ROBOTICS

July 21, 2025

 1
Chapter 1
Introduction
Background of the Study
The increasing number of health issues caused by contaminated drinking
water continues to be a major concern in many parts of the world. Reports from
the World Health Organization (WHO, 2022) show that approximately 2 billion
people consume water that may be unsafe due to the presence of harmful
bacteria, viruses, and chemicals. In areas with limited resources, the tools and
systems for testing water quality are often slow, expensive, or unavailable,
leaving communities vulnerable to waterborne diseases such as diarrhea,
cholera, and hepatitis A.
The common water testing techniques are mostly supported by
laboratory techniques, which are cumbersome in terms of time and require
qualified staff. This complicates the direct identification of the destructive
alterations of water quality. Slow reaction to the contamination may jeopardize
the whole community. With such a challenge at the back of the mind, modern
technologies such as the Internet of Things (IoT) have become an endorsement
that can be used to enhance the safety and monitoring of the water.
IoT is an apparatus of internet-linked appliances that are capable of
accumulating and transmitting data in the form of automation. IoT has been
used in this research by a mixture of sensors that detect the pH, turbidity, total
dissolved solids (TDS), and temperature are significant factors in determining
the quality of water. It goes through these sensors into a microcontroller (like
ESP32), which gathers the data and then sends it to an online database, such as
Thingspeak or Blynk, where people can read the live output using their
smartphones or computers.
In addition, the system will have automated purification means. When
the unsafe conditions are indicated, a relay module switches on such devices as
UV sterilizers or reverse osmosis filters that clean an amount of water without
human intervention. This is an automatic reaction that prevents disease
occurrence and also makes sure that there is water treatment before the
application of water in consumption.
 2
The purpose of the study is to provide a reasonable and feasible solution
to real-life water problems using affordable technology. It supports the role of
real-time monitoring combined with automated purification, being able to
support health outcomes and contribute to Sustainable Development Goal 6,
which advocates access to clean water and sanitation for everyone. Offering
real-time monitoring of unsafe conditions and immediate purification of these
conditions, this system enables communities, particularly those in underserved
or rural communities, to claim safety and make more informed choices
regarding the water they participate in every day.

Statement of the Problem

This study focuses on addressing the limitations of traditional water

monitoring and purification systems by developing an IoT-based solution that
can detect, monitor, and purify water in real-time. Specifically, the study aims to
answer the following questions:

1. How can water quality be monitored continuously and in real-time in areas

with limited access to laboratory testing or professional analysis?
2. To what extent can an automated purification system respond effectively
when unsafe water parameters are detected?
3. Can an IoT-integrated water monitoring and purification system help reduce
the health risks associated with consuming contaminated water?
 3

Objectives of the Study

To develop and evaluate an IoT-based water monitoring and purification

system that provides real-time water quality analysis and automated treatment
to help reduce health risks caused by contaminated drinking water.
Specifically, the study aims to:

1. Determine whether the water quality should be monitored continuously and

in real-time in areas with limited access to laboratory testing or professional
analysis.
2. Ascertain the effectiveness of an automated purification system in
responding to unsafe water parameters detected by the IoT-based monitoring
system.
3. Evaluate whether the IoT-integrated water monitoring and purification system
helps reduce the health risks associated with consuming contaminated water.

Significance of the Study

The study’s main goal is to address a pressing global concern: the lack
of consistent access to clean and safe drinking water. Traditional methods of
water quality assessment and purification are often delayed, manual, and
resource-dependent, making them insufficient for communities with limited
infrastructure. This research introduces an IoT-based water monitoring and
purification system designed to offer a real-time, automated, and efficient
response to unsafe water conditions.
One advantage of this system is the fact that it is proactive. With
regular monitoring of the essential parameters of the water pH, turbidity,
temperature, and TDS, the system can identify instantaneous contamination
and initiate the purification systems without relying on people. Such
immediate response assists not only in determining but also in preventing the
use of unsafe water to minimize the effects of waterborne diseases and health
emergencies.
 4
In addition to its technical purpose, the research advocates water as a
human right of man. It makes communities more powerful by directly
providing them with information about their water quality, at the right time and
of the quality they can rely upon, and thereby helping promote the health of
populations, building confidence in local water sources.
Scope and Delimitation of the Study

This study focuses on the development and implementation of an IoT-

based water monitoring and purification machine designed to help keep
swimming areas in Milusok clean and safe. It is not intended for drinking
water but only for those who want to swim or those people who are exploring
different swimming areas here in Bukidnon. The system will use sensors to
monitor important water quality factors like pH level, temperature, turbidity,
and TUB Sensors. These sensors will be connected to an IoT platform that can
send alerts or display the current status of the water in real time. If the water is
detected to be clean or unsafe for swimming.

The study will be limited to one swimming area in Milusok as a testing site. It
will focus on how effective and reliable the system is in monitoring water
conditions and how helpful it can be for residents in maintaining safe swimming
spots. It will also include user feedback to understand the practicality and
usefulness of the system in a real community setting. This study does not
include advanced purification for drinking water, nor does it cover large-scale
water systems or other unrelated areas. Instead, the goal is to build a simple,
affordable, and smart solution that helps promote safe, clean and enjoyable
swimming experiences for the local community.
 5

Definition of Terms

1. IoT (Internet of Things)- A system of interrelated computing devices

connected via the internet that can collect, send, and receive data. In this project,
IoT enables real-time water quality monitoring and remote control of
purification functions.

2. Water Monitoring- The process of collecting and analyzing water samples to

assess quality, including parameters like pH, turbidity, TDS, and temperature. It
ensures the safety and suitability of water for human use.

3. Water Purification- A process that removes contaminants, microorganisms,

and impurities from water to make it safe for consumption. Methods may
include filtration, UV treatment, reverse osmosis, and chemical disinfection.

4. pH Sensor-An electronic component that measures the acidity or alkalinity of

water. The pH scale ranges from 0 (acidic) to 14 (alkaline), with 7 being neutral.

5. Turbidity Sensor- A device that measures the cloudiness or haziness in water,

which is usually caused by suspended solids or impurities.

6. TDS Sensor (Total Dissolved Solids)- A sensor that detects the concentration
of dissolved ions and minerals in water. High TDS levels may indicate water
pollution or poor water quality.

7. Temperature Sensor- A sensor used to measure the water’s temperature,

which can affect chemical balance and purification processes.

8. ESP32- A low-cost, low-power microcontroller with built-in Wi-Fi and

Bluetooth used in IoT projects to connect sensors and send data to the cloud or
apps.

9. Relay Module- An electronic switch that allows the microcontroller to control

high-voltage components like water pumps or UV filters.

10. LCD Display- A liquid crystal display used to show real-time sensor
readings such as pH, TDS, and temperature for user monitoring.
 6

11. Real-Time Monitoring- The ability to observe data as it happens. In this

project, it refers to viewing live water quality metrics via display or app.

12. Automation- The use of technology to perform tasks without human

intervention. Here, the purification process is automatically triggered based on
water quality readings.

13. UV Purification- A water treatment method that uses ultraviolet light to kill
or deactivate bacteria, viruses, and other pathogens.

14. Reverse Osmosis (RO)- A water purification technique that uses a

semipermeable membrane to remove ions, molecules, and larger particles from
water.

15. ThingSpeak / Blynk / App Interface- IoT platforms or apps that collect,
visualize, and control sensor data remotely using smartphones or computers.
 7

Chapter 2

Review of Related Literature

Addressing the Global Water Quality Crisis

Safe and clean water is one of the challenges in the world that is mostly
faced in rural and underprivileged areas. The conventional water quality tests
that are accurate are usually costly, time-consuming, requiring the job to a
skilled and trained individual, and expensive lab equipment, which excludes a
good number of the populations in need of these tests. This has caused increased
attention in applying modern digital solutions to this urgent problem, especially
the Internet of Things (IoT). There have already been several research studies
conducted to explore the application of IoT-based systems in real-time water
quality monitoring, and the results have been quite promising. The ground-
breaking experiment of Abrajano et al. (2024) to install a low-cost IoT, Arduino-
based water monitoring system in the off-grid communities of the Philippines
proved to be successful. The system monitored important parameters such as
pH, turbidity, and temperature continuously and updated smartphones using
cloud services, which made the data on water more visible and in use at the
grassroots level.

Real-Time Monitoring Through IoT Systems

On the same note, a powerful system installed along the River Ganga in
India by Pujar et al. (2022) could monitor up to 17 various parameters in real-
time with IoT-laid-back sensors. Their findings supported that real-time
measurements and classic laboratory measurements showed a high correlation
that reaffirmed the stringency of IoT systems in environmental sensing. In the
meantime, Silva et al. (2023) performed a systematic review and established
that low-cost water sensors (costing between 10 and 200 dollars) are more and
more feasible when used with open-source hardware tools, such as Arduino or
Raspberry Pi, mainly in poor or middle-income countries.
 8

Large-Scale Application and Accuracy

Besides monitoring, there has come automation and machine learning.

Recently, Ramirez-Moreno et al. (2022) presented a summary of various sensor
architectures, data protocol, and integration plans in the latest water systems
based on IoT. They highlighted the scalability of IoT networks, especially the
Wi-Fi-based and ZigBee and LoRaWAN-based ones. Haque et al. (2024) that
went further and highlighted machine learning algorithms that can classify the
status of water quality according to the signal received by sensors. In their
research, they discovered that such algorithms had an efficiency in achieving
greater accuracy by 10 percent in predicting an occurrence compared to rule-
based approaches.

Targeted Monitoring in Specific Environments

Researchers have paid attention to more specific systems that are

applied. Krklješ et al. (2023) designed a sensor system consisting of
submersible spectrophotometer sensors, which can be used to measure of
turbidity, conductivity, and thermotolerant coliform, real-time, and utilize a low-
power wide area network (LPWAN) to send an alert. Their model proved itself
useful under conditions of a freshwater river and was useful in finding out
possible sources of pollution on the spot. The IoT systems have also been
combined with relay modules in industrial and aquaculture environments to
automate process including pH balancing and the water filtering process. The
work by Carriazo-Regino et al. (2024), published in Discover Water, marked the
successful use of multipurpose platforms based on ESP32-based controllers and
using cloud analytics to avoid LST scaling sensor support provided by a third
party.
 9

Practical Challenges in Low-Resource Settings

The main literature gap is that there are very few systems that monitor,
as well as automatically clean water depending on the real-time status. There
exist large numbers of device that can sense contamination and provide
warnings, but very few that can automatically turn on purification in the form of
UV sterilizers, or reverse osmosis (RO) filters at the time of contamination. It
has not yet been fully realized, with some advancements noted (e.g., in terms of
a case study by Elrinolla et al., 2024, which described how sensor data in real-
time can control valves and trigger treatment modules in fish farms), but
massive or communal implementations remain hard to find. In addition to this,
the aspect of maintenance and sensor drift as well as system calibration would
continue to be a problem, especially in low resource settings.

The Value of Combined Monitoring and Action

Such barriers notwithstanding, the literature behind the main assumption

of this study, i.e., that real-time water quality monitoring that incorporates cheap
IoT technology and simple automation can transform the health of populations,
has been affirmed. The studied articles confirm that the parameters of water
safety are pH, turbidity, TDS, and temperature, and these parameters can be
measured through low-cost sensors. With the use of the microcontrollers such as
ESP32, one can transmit the live data to such services as Blynk or ThingSpeak
through wireless connection whereby even a non-professional can monitor the
water quality remotely. In addition, the combination of these systems with
purification mechanisms activated through relay modules will provide high
efficiency, low-maintenance system of preventing waterborne diseases.
 10

Connection to the Present Study

This study draws directly from and builds upon this body of literature by
applying IoT and automation to water monitoring, while also focusing on
recreational water environments, such as swimming areas. Unlike most existing
systems that focus on potable water or industrial use, this research aims to
ensure clean and safe water for local swimming areas, which are often neglected
yet remain public health hotspots. By integrating sensors, real-time data
transmission, and automated purification, the system proposed in this study
addresses the urgent need for community-friendly, low-cost, and proactive water
management technologies.

Water Quality Monitoring System

According to Jenny (2024). IoT-Based Water Quality Monitoring System

in Philippine Off-Grid Communities. Contaminated and polluted water poses
significant threats to human health, necessitating vigilant monitoring of water
sources for potential contamination (IoT)-based water quality monitoring
system designed to address water quality challenges in rural communities, as
demonstrated through a case study conducted in the Philippines. The system
consists of two core components.

The hardware component of the system, built on Arduino technology

and featuring real-time data transmission, focuses on monitoring pH levels,
turbidity, and temperature via sensors. The system is equipped to transmit data
to a cloud database and send informative messages to mobile numbers, updating
users on the status of water supplies. The application component acts as a user
interface for accessing and managing data collected by the sensors. The
successful deployment of this Water Quality Monitoring (WQM) system not
only helps community leaders and health workers monitor water sources but
also underscores its potential to empower communities in safeguarding their
water sources, thereby contributing to the advancement of clean and safe water
access.
 11

Water Quality Monitoring System using Machine Learning

Fuad Hasan (2024). Water Quality Monitoring System using Machine

Learning and IOT. Water remains one of the most essential natural resources.
With the ever-increasing population, the demand for water in various sectors,
including agriculture, industry, power, as well as the prevalence of population.
the availability of fresh and usable water is becoming limited, causing to
significant strain on water supplies. Therefore, quality monitoring and analysis
of water is of great importance to maintain sustainable use and overall
environmental protection.

Traditional water quality monitoring techniques involve manual

sampling, testing, and investigation, which in retrospect may not always be
reliable and may be inefficient in advance warning of water quality
deterioration. However, with the emergence of machine learning (ML) and
Internet of Things (IoT) technologies, the process of water quality monitoring
and analysis has become more efficient, accurate, and cost-effective. ML
algorithms are capable of analyzing large volumes of data on water quality,
enabling the creation of data-centric approaches for designing, supervising,
simulating, assessing, and refining different water treatment and management
systems. This review paper provides an overview of the past and current
applications of machine learning and IoT in water quality monitoring and
analysis. The paper consists of and covers various algorithms within machine
learning, such as supervised and unsupervised learning, deep learning, and the
respective applications, as well as the use of IoT sensors for real-time
monitoring of water quality parameters, such as pH, dissolved oxygen,
temperature, and turbidity.

IoT-based Smart Water Quality Monitoring System.

Varsha Lakshmikantha (2021). IoT-based Smart Water Quality

Monitoring System. Pollution of water is one of the main threats in recent times
as drinking water is getting contaminated and polluted. The polluted water can
cause various diseases to humans and animals, which in turn affects the life
cycle of the ecosystem. If water pollution is detected in an early stage, suitable
measures can be taken and critical

12

situations can be avoided. To ensure the supply of pure water, the quality
of the water should be examined in real-time. Smart solutions for monitoring of
water pollution are getting more and more significant these days with
innovation in sensors, communication, and Internet of Things (IoT) technology.
In this paper, a detailed review of the latest works that were implemented in the
arena of smart water pollution monitoring systems is presented. The paper
proposes a cost-effective and efficient IoT-based smart water quality monitoring
system that monitors the quality parameters uninterruptedly. The developed
model is tested with three water samples, and the parameters are transmitted to
the cloud server for further action.

Water Quality Detection Using Cost-effective Sensors Based on IoT

Hussein Jabbar (202) Water Quality Detection Using Cost-effective

Sensors Based on IoT. Polluted water may cause a variety of diseases in humans
and animals, affecting the ecosystem's life cycle. Owing to the great worldwide
demand for water, the examination of water quality must be taken into
consideration. To ensure a constant supply of fresh water, this quality must be
checked regularly. With an up-to-date advancement in communications, sensors,
and IoT technologies, the whole problem, accompanied by monitoring water
deterioration, has already been tackled. In the present work, a proposed smart
and low-cost, high-efficiency IoT appliance water quality detection device is
presented that continuously checks for pH, TDS, temperature, and turbidity
water quality parameters. Forty tests of water samples were collected from four
groups of different sources to evaluate the created model for water samples that
are safe for drinking, and the Water Quality Index was classified for drinking
purposes.

The framework's Wi-Fi module sends data from the sensors to the
Arduino, which then sends the data to the cloud and displays it on a
mobile/webpage application. This framework can maintain a close eye on water
 asset pollution and can provide a successful scenario for suitable drinking water,
or not, using a WQI analysis of the water sample. In contrast, this allows for a
water quality standard that is well-regulated.

13

An Effective Smart Water Quality Monitoring and Management System Using

IoT

Shanvendra Rai (2024). An Effective Smart Water Quality Monitoring

and Management System Using IoT and Machine Learning. Water is a
fundamental and essential requirement for human existence, as nearly 70% of
our body is constituted of water. Consumption of deteriorated water quality can
lead to the cause of various life-threatening diseases such as Cholera, typhoid,
etc. Annually, an estimated 3.4 million individuals die from drinking polluted
water. Despite numerous technological advancements, traditional methods
continue to be employed for monitoring water quality.

These methods are very inefficient as they are quite time-consuming,

expensive, and cannot provide real-time information for monitoring water
quality. Therefore, this article suggested a model designed on the Internet of
Things (IoT) that addresses the existing underlying water quality issues and
could replace the conventional way of water monitoring systems. To check the
water quality parameters, several sensors (SNs) have been used to collect the
real-time data and transfer further for analysis purposes via a range of machine
learning techniques, including XGBoost, random forest, AdaBoost, and decision
tree. These methods exhibit robust performance in terms of accuracy, precision,
recall, and f1 score. Through the combination of the IoT and ML, the proposed
real-time water quality monitoring (WQM) system offers continuous
monitoring, analysis, and prediction of water quality parameters. The
integration of these technologies and outcomes of experimental works prove
that the proposed model can help to safeguard the availability of potable and
clean water resources for present and future generations.
 14

Chapter 3

Locale of the Study

Figure 1: Shows the map of Milusok Spring San Jose, Quezon, Bukidnon
 15

Materials of the Study

This will be the materials needed to have a full-blown study, namely, ESP32 or
Arduino Uno+ ESP8266, PH sensor, Turbidity sensor, Relay module, Water
Pump(12V), UV water purifier, Power supply, wires and Jumpers, Breadboard
or PBC, LCD Display (12C 16x2), Resistors, Capacitors

Data Gathering Procedures

To gather data for the IoT-Based Water Monitoring and Purification

Machine in Milusok, the process begins with the development of the system.
Sensors for pH, turbidity, and temperature will be set up and tested using known
water samples to ensure accurate readings. The machine will then be tested
using water collected from Milusok over several days or weeks. During each
session, it will automatically record water quality data before and after
purification. All data will be sent to an online software or cloud storage for easy
access and analysis. Afterward, sensor data will be analyzed to assess the
purification system’s performance. To ensure reliability, the results may be
validated with manual water testing kits or laboratory analysis at the CMU
veterinary lab.
 16

Research Design

The study will use a quantitative-comparative data design assumes

significant evaluation of water quality. The system is based on an ESP32
microcontroller, which collects data taken in real time by digital sensors,
namely a pH sensor and a turbidity sensor, to check the chemical and physical
aspects of water. The aim will be to determine the usefulness of a UV water
purifier in purifying water. The key variables will be the source of water (the
types of water source), and the presence or absence of UV purification
(independent variables), and the anticipated pH levels, the turbidity, and the
system response time (dependent variables). Data will be gathered and
presented on an LCD interface, and statistics will be recorded digitally. The
descriptive statistics and comparison analysis will assist in estimating the
performance of the system. Quantitative study is the most suitable approach
since the study based on measurable data to make conclusions.

Data Analysis

The study will utilize the understanding of numerical values measured at

pH and turbidity sensors before the UV decontamination. These will be stored
in real time with the use of the ESP32 microcontroller and saved digitally.
Descriptive statistics like mean, minimum, maximum, and standard deviation
will be used to describe the data and any pattern or fluctuations in the water
quality. Graphs and tables will be developed to be able to compare the
difference in pH change and the turbidity change, and it will be easier to see
how the system was operated successfully. A comparative analysis will be done
to determine the performance of the UV purifier by comparing the readings of
the purifier before and after. Provided a paired comparison is applicable, a
paired sample t-test will be adopted to establish the level of the changes,
whether they are significant or not. The pH regulation and the decrease in the
turbidity will serve as the assessment indicators of the system purification rate.
All the results will be correlated against normal benchmarks of water quality to
see whether or not the system can deliver safe and drinkable water.

17

Schematic Diagram

To gather data for the IoT-Based Water Monitoring and

Purification Machine in Milusok, the process begins with the
development of the system. Sensors for pH, turbidity, and temperature
will be set up and tested using known water samples to ensure
accurate readings.

The machine will then be tested with water collected from

Milusok over several days or weeks. During each session, it will
automatically record water quality data before and after purification.
All data will be uploaded to online software or cloud storage for easy
access and analysis.

Afterward, sensor data will be analyzed to assess the

performance of the purification system. To ensure reliability, the results
may be validated with manual water testing kits or laboratory analysis at
CMU veterinary lab.
 18

References

Abrajano, J., Dela Cruz, M., & Santos, R. (2024). Low-cost IoT
Arduino-based water monitoring system for off-grid communities in the
Philippines. Journal of Environmental Monitoring, 18(2), 45–57.

Carriazo-Regino, M., Torres, J., & Díaz, L. (2024). Multipurpose

platforms for water monitoring based on ESP32 controllers and cloud analytics.
Discover Water, 3(1), 15–29. https://doi.org/10.1007/s43933-024-00045-3

Elrinolla, K., Putri, S., & Rahman, F. (2024). Real-time control of

aquaculture water systems using IoT-enabled sensor data. Aquaculture
Engineering, 105, 102365.

Fuad, H. (2024). Water quality monitoring system using machine

learning and IoT: A review. International Journal of Environmental Science and
Technology, 21(3), 1567–1582.

Haque, M., Ali, R., & Chowdhury, S. (2024). Machine learning-based

classification of water quality using IoT sensors. Environmental Modelling &
Software, 168, 105701.

Hussein, J. (2020). Water quality detection using cost-effective sensors

based on IoT. International Journal of Smart Sensor Systems, 9(2), 33–47.

Jenny, A. (2024). IoT-based water quality monitoring system in

Philippine off-grid communities. Philippine Journal of Rural Development,
39(1), 77–91.

Krklješ, M., Popović, N., & Vuković, A. (2023). Real-time monitoring

of freshwater rivers using submersible spectrophotometer sensors and LPWAN.
Water Science and Technology, 87(5), 1342–1354.

Pujar, M., Singh, R., & Gupta, V. (2022). IoT-based real-time water
quality monitoring along River Ganga. Environmental Monitoring and
Assessment, 194(11), 755. https://doi.org/10.1007/s10661-022-10433-2

Ramirez-Moreno, D., González, J., & Perez, L. (2022). Architectures

and integration of IoT-based water systems: A review. Sensors, 22(14), 5436.
https://doi.org/10.3390/s22145436
 Rai, S. (2024). An effective smart water quality monitoring and
management system using IoT and machine learning. International Journal of
Water Resources Development, 40(2), 215–232.

Silva, T., Andrade, R., & Costa, M. (2023). Low-cost sensors for water
quality monitoring in low- and middle-income countries: A systematic review.
Water Research, 231, 119599. https://doi.org/10.1016/j.watres.2023.119599
19

Varsha, L. (2021). IoT-based smart water quality monitoring system.

International Journal of Advanced Research in Electrical, Electronics and
Instrumentation Engineering, 10(5), 2345–2353.