Jyothy Charitable Trust®
Jyothy Institute of Technology
Tataguni, off Kanakapura road, Bengaluru-560082
Approved by the All India Council for Technical Education (AICTE) - New Delhi;
Affiliated to Visvesvaraya Technological University (VTU), Belagavi;
Department of Computer Science &
Engineering
Accredited by the National Board of Accreditation (NBA) -New
Delhi,
Project Phase 1 (BCS685)
Water Quality Monitoring System
Likhitha.G: 1JT22CS087
Likhitha.K.V: 1JT22CS088
Keshava Reddy: 1JT22CS081 Mr Srinidhi Kulkarni
Kanagavelu: 1JT22CS071 Assistant Professor
Institution Vision & Mission
Vision
To be an Institute of Excellence in Engineering education, Innovation, Research and work towards evolving
great leaders for the country's future and meeting global needs
Mission
The Institution aims at providing a vibrant, intellectually and emotionally rich teaching learning
environment with the State of the Art Infrastructure and recognizing and nurturing the potential of each
individual to evolve into one's own self and contribute to the welfare of all.
Department of CSE., Jyothy Institute of Technology, Bengaluru
Department of Computer Science & Engineering
Vision
To be a center of excellence in Computer Science and Engineering education, focus on research,
innovation and entrepreneurial skill development with professional competency.
Mission
M1: To provide state-of-the-art ICT infrastructure and innovative, research-oriented teaching-learning
environment and motivation for self-learning & problem-solving abilities by recruiting committed
faculty.
M2: To encourage Industry Institute Interaction & multi-disciplinary approach to problemsolving and
adapt to ever-changing global IT trends.
M2: To encourage Industry Institute Interaction & multi-disciplinary approach to problemsolving and
adapt to ever-changing global IT trends.
Department of CSE., Jyothy Institute of Technology, Bengaluru
OVERVIEW
• INTRODUCTION
• LITERATURE SURVEY
• PROBLEM IDENTIFICATION
• PROBLEM STATEMENT
• OBJECTIVES
• HARDWARE AND SOFTWARE REQUIREMENTS
• REFERENCES
Department of CSE., Jyothy Institute of Technology,
INTRODUCTION
• Water is one of the most essential natural resources for human life, agriculture,
and industry.
• Many people still do not have access to clean and safe drinking water,
especially in rural and developing areas.
• Access to clean water is essential for health and well-being.
• Our project uses Artificial Intelligence [AI] and Machine Learning [ML]to
monitor and predict water quality in real-time.
• This system helps detect contamination early and suggests suitable purification
methods, ensuring safe water for communities
Department of CSE., Jyothy Institute of Technology, Bengaluru
LITERATURE SURVEY
Title Main Methodology Findings Reference
Contribution
Soft Sensor for Real-time soft sensors SVR, Cubist, QRNN, SVR best for COD Soft Sensor Using Machine
OWTS WQ using ML to predict PLS with 56 field (R² = 0.96), CUB Learning Algorithms for
Prediction COD, TSS, E. coli samples best for TSS (R² = Predicting the Water Quality
0.99) of an Onsite Wastewater
Treatment System.
[Link]
vironau.2c00072
IoT-Based WQ Mobile app + ESP32 + Arduino with sensors Detected best IoT-Enabled Advanced Water
Monitoring for sensors for pond WQ (pH, TDS, turbidity), (Birkona) and Quality Monitoring System
Ponds tracking cloud storage worst (Budha) for Pond Management and
ponds Environmental Conservation.
[Link]
SS.2024.3391807
AI/ML for WQ Review of AI/ML in Survey on ML/AI + IoT AI enhances Machine Learning and AI-
Monitoring & real-time water quality integration in WQ accuracy, enables Driven Water Quality
Treatment systems management early Monitoring and Treatment
contamination [Link]
detection nf/202450503012
Title Main Methodology Findings Reference
Contribution
Water Quality Classification using Dataset with pH, RF best (70.4%), DT Water Quality Prediction Using
Prediction Using Random Forest, KNN, turbidity, DO, etc.; lowest (58%) Machine Learning Models. E3S
ML Models Decision Tree accuracy comparison Web of Conferences.
[Link]
459601025596
Deep Learning DNN models for Sentinel-2 Rrs data + DNN had lowest Deep Learning for Water Quality
for Water remote sensing-based GLORIA; DNN, XGB, RMSE: Chl-a Multivariate Assessment in Inland
Quality in China Chl-a, SDD, TSS SVM, RF, MDN (14.18), SDD Water Across China.
prediction (0.07m) [Link]
04078
Smart Water IoT-WSN with AI ML regressors: AdaBoost best Smart Water Quality Monitoring
Quality sensor predictions (E. AdaBoost, RF, SVM, (MAE 14.37), with IoT Wireless Sensor Networks.
Monitoring with coli) KNN, Ridge stochastic GB worst Sensors, 24(9), 2871.
IoT [Link]
Deep Learning Compared ARIMA vs ARIMA vs 4 deep DL outperformed Assessing the Performance of Deep
for Short-Term deep learning for models; ARIMA (MAPE < Learning Algorithms for Short-Term
WQ Prediction time-series WQ univariate/multivariat 20% vs 27–400%) Surface Water Quality Prediction.
e setups [Link]
90
PROBLEM IDENTIFICATION
• Water contamination often goes undetected due to manual, slow
testing and lack of real-time monitoring, leading to delayed
treatment and health risks.
• Existing systems are isolated, depend on human interpretation,
lack data sharing, and limit community awareness, reducing early
detection and response.
To address these challenges, there is a need for an intelligent and automated system that:
• Use water quality sensors for continuous and accurate detection.
• Apply machine learning to predict future water quality trends.
• Recommend purification methods based on contamination type.
• Enable community participation for reporting and validation.
• Provide real-time dashboards and alerts for faster action.
Department of CSE., Jyothy Institute of Technology, Bengaluru
PROBLEM STATEMENT
Access to clean and safe water is essential for human health, , yet
contamination often goes undetected due to traditional testing methods being manual
and time-consuming. The lack of real-time monitoring, predictive analysis, and public
involvement leads to delayed responses, health risks, and inefficient management of
water resources, especially in rural and industrial areas where early detection and
prevention are critical.
Our project is to design and implement a smart system that integrates IoT-based water
quality sensors, AI-powered predictive models, and community interaction tools to
enable real-time monitoring, automated contamination detection, predictive analysis,
and immediate purification recommendations — ultimately promoting safe water
practices and timely decision-making.
Department of CSE., Jyothy Institute of Technology, Bengaluru
PROJECT OBJECTIVES
• Accurate and Real-Time Detection of water contamination
using IoT-based sensors.
• Intelligent Prediction of future water quality trends through
advanced machine learning models.
• Automated Purification Recommendations tailored to
specific contamination types.
• Community-Driven Reporting and Validation to promote
proactive participation and awareness.
• Dynamic Dashboards and Instant Alerts for actionable
insights and timely decision-making.
Department of CSE., Jyothy Institute of Technology, Bengaluru
HARDWARE AND SOFTWARE OVERVIEW
Hardware:
• Water quality sensors are devices used to measure
parameters like pH, turbidity, Total Dissolved Solids (TDS),
dissolved oxygen, and temperature, helping detect
contamination in real-time.
Software:
• Python for programming, Scikit-learn for model training,
Database for data storage, and Streamlit for a real-time
dashboard.
Department of CSE., Jyothy Institute of Technology, Bengaluru