A Smart Flood Detection and Warning Prototype Using IoT Sensors and ML Model

Abstract

Floods are natural disasters that cause significant damage to infrastructure, agriculture, and human life. They can cause a lot of problems and people can lose their lives and homes. This is very sad. The systems we have now to monitor floods are very expensive. They are also made for places. Most of these systems use rules that do not work well and are hard to make bigger. This paper is talking about a system that uses the Internet of Things or IoT for short. This system uses devices called sensors to get information in real time. These sensors measure things like how high the water\'s how much rain is falling, how wet the soil is and the temperature and humidity. The ESP8266 microcontroller is what collects all this information. Then this information is sent to a server using something called HTTP.A special computer program called a machine learning model is trained on information about floods. This program can tell if a flood is likely to happen. The system can send warnings on time. It uses a webpage called a dashboard and also sends messages to peoples phones and has local signs, like a loud noise and a flashing light. By using IoT and machine learning this system is more accurate. It is also cheaper. Can be used in many places to monitor floods.

Introduction

The document describes a smart flood detection and early warning system that uses IoT and Machine Learning to reduce damage caused by floods, which are becoming more frequent due to climate change and urbanization. Traditional flood monitoring methods are costly and less accurate, so the project proposes a low-cost, real-time solution using an ESP8266 microcontroller connected to multiple sensors (ultrasonic water level sensor, rain sensor, soil moisture sensor, and DHT22 for temperature and humidity).

The system continuously collects environmental data and sends it to a server using HTTP. A Random Forest Machine Learning model processes this data to classify conditions as Safe, Warning, or Danger based on parameters like rainfall, water level, soil moisture, and humidity. The model is trained using Kaggle datasets and real sensor data, with preprocessing steps like normalization and handling missing values.

When a flood risk is detected, the system triggers multiple alerts: LEDs and a buzzer for local warning, and SMS alerts through a GSM module (SIM800L) for remote notification. A web dashboard built with Flask and Chart.js displays real-time sensor data and flood risk levels.

Conclusion

This paper presented a Smart Flood Detection and Early Warning Prototype. It effectively combines Internet of Things sensing with machine learning for time environmental monitoring and flood prediction. The system keeps collecting data from five sensors. These include water level, rainfall, soil moisture, temperature and humidity sensors. The data is sent to a server. There a Random Forest machine learning model classifies the flood risk level. When the system detects danger it sets off alarms. It also sends SMS alerts through GSM within seconds. The prototype works well. It achieved a testing accuracy of ninety-two point eight percent. The response time is between three and six seconds. It showed performance in indoor simulated flood conditions. The hardware cost is low one thousand nine hundred and fifty rupees. The modular design makes it a practical and scalable solution. It can be used for community-level deployment, in flood- regions. Future work will explore some features. These include LoRa-based long-range communication and solar-powered nodes. The team will also look into edge AI inference. They plan to integrate the system with disaster management platforms.

References

[1] K. Vinothini and S. Jayanthy, \"IoT Based Flood Detection and Notification System using Decision Tree Algorithm,\" 2019 Int. Conf. on Intelligent Computing and Control Systems (ICCS), Madurai, India, pp. 1481–1486, May.2019.[Online].Available: https://ieeexplore.ieee.org/document/9065799/ [2] A. A. Rashid, M. A. M. Ariffin, and Z. Kasiran, \"IoT-Based Flash Flood Detection and Alert Using TensorFlow,\" 2021 11th IEEE Int. Conf. on Control System, Computing and Engineering (ICCSCE), Penang, Malaysia, pp. 80–85, Nov. 2021. [Online]. Available: https://ieeexplore.ieee.org/document/9530926/ [3] A. Pravin, T. P. Jacob, and R. Rajakumar, \"Enhanced Flood Detection System using IoT,\" 2021 6th Int. Conf. on Communication and Electronics Systems (ICCES), Coimbatore, India, pp. 507–510, Jul. 2021. [Online]. Available: https://ieeexplore.ieee.org/document/9489059/ [4] F. S. Mousavi, S. Yousefi, H. Abghari, and A. Ghasemzadeh, \"Design of an IoT-Based Flood Early Detection System Using Machine Learning,\" 2021 26th Int. Computer Conf., Computer Society of Iran (CSICC), Tehran, Iran, pp. 1–7, Mar. 2021. [Online]. Available: https://ieeexplore.ieee.org/document/9420594/ [5] T. Sharma, A. Pal, A. Kaushik, A. Yadav, and A. Chitragupta, \"A Survey on Flood Prediction Analysis Based on ML Algorithm Using Data Science Methodology,\" 2022 IEEE Delhi Section Conf. (DELCON), New Delhi, India, pp. 1–7, Feb. 2022. [Online].Available: https://ieeexplore.ieee.org/document/9753396/ [6] L. Saravanan, D. Vijayanandh, J. R. Arunkumar, K. P. Chandran, and R. T. Prabhu, \"A Novel Approach for a Smart Early Flood Detection and Awareness System using IoT,\" 2022 8th Int. Conf. on Smart Structures and Systems (ICSSS), Chennai, India, pp. 1–6, Apr. 2022. [Online].Available: https://ieeexplore.ieee.org/document/9782286/ [7] M. Khalaf, H. Alaskar, A. J. Hussain, T. Baker, Z. Maamar, R. Buyya, P. Liatsis, W. Khan, H. Tawfik, and D. Al-Jumeily, \"IoT-Enabled Flood Severity Prediction via Ensemble Machine Learning Models,\" IEEE Access, vol. 8, pp. 70375–70386, Apr. 2020. [Online]. Available: https://ieeexplore.ieee.org/document/9057676/ [8] E. Deowan, S. Haque, J. Islam, Md. Hanjalayeamin, Md. T. Islam, and R. T. Meghla, \"Smart Early Flood Monitoring System Using IoT,\" 2022 14th Seminar on Power Electronics and Control (SEPOC), pp. 1–6, Nov. 2022.[Online].Available: https://ieeexplore.ieee.org/document/9976434/ [9] G. Furquim, G. P. R. Filho, R. Jalali, G. Pessin, R. W. Pazzi, and J. Ueyama, \"How to Improve Fault Tolerance in Disaster Predictions: A Case Study about Flash Floods Using IoT, ML and Real Data,\" Sensors (MDPI), vol. 18, no. 3, p. 907, Mar. 2018. [Online]. Available: https://www.mdpi.com/1424-8220/18/3/907 [10] K. Mosavi, P. Ozturk, and K. Chau, “Flood Prediction Using Machine Learning Models: Literature Review,” Water (MDPI), vol. 10, no. 11, pp. 1536, 2018. [Online]. Available: https://www.mdpi.com/2073-4441/10/11/1536

Copyright

Copyright © 2026 Vishveshvar Bidve, Darshan Kawale, Indrayani Raut, Prof. Shilpa Kale. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.