LoRaWAN Based Smart System for Unsafe Event Detection in Mining Industry Using ML

Abstract

Mining environments are highly hazardous due to toxic gases, extreme temperatures, fire risks, landslides, and physically demanding conditions. This paper presents a real-time monitoring and alert system integrating IoT sensors, ESP32 microcontroller, WiFi and LoRa communication, and Machine Learning algorithms. The system monitors worker health parameters (pulse, SpO?, body temperature) and environmental conditions (gas, humidity, fire, landslides, earthquakes). Data is transmitted to a cloud platform and analyzed using Random Forest models for hazard prediction. Alerts are displayed on LCD, web dashboard, and mobile application. Experimental evaluation demonstrates reliable detection of unsafe events with >90% prediction accuracy and <1s latency. The proposed system enhances worker safety and provides a scalable framework for industrial hazard monitoring.

Introduction

This paper presents an IoT- and Machine Learning-based smart mining safety system that continuously monitors both miners' health and environmental conditions to improve safety in hazardous mining environments. Unlike conventional mining safety systems that primarily monitor environmental hazards, the proposed system also tracks workers' vital signs and uses predictive analytics to detect accidents and health risks in real time.

Background

Mining workers face numerous dangers, including:

• Toxic gas exposure
• High temperature and humidity
• Fire hazards
• Landslides and earthquakes
• Low oxygen levels
• Health emergencies such as abnormal heart rate or low blood oxygen

Advances in IoT, wireless communication (WiFi and LoRa), and Machine Learning enable continuous monitoring of these conditions and allow rapid emergency response.

Literature Survey

Previous studies have explored:

• IoT-based mine safety monitoring using gas, temperature, and humidity sensors.
• Wireless sensor networks (WSNs) and ZigBee for underground monitoring.
• Machine learning for predicting hazardous industrial conditions.
• ESP32-based IoT systems for embedded monitoring.
• Regression and predictive models for mining applications.

However, most existing systems focus mainly on environmental monitoring rather than integrating worker health monitoring with intelligent accident prediction.

Proposed System

The proposed system integrates:

• ESP32 microcontroller
• Health sensors (pulse, SpO?, body temperature)
• Environmental sensors (gas, smoke, fire, temperature, humidity)
• Accelerometer for worker fall detection
• WiFi and LoRa communication
• Machine Learning algorithms (Random Forest and Gradient Boosting)
• Blynk IoT cloud platform
• Streamlit web dashboard

The system continuously collects sensor data, predicts unsafe conditions such as:

• Toxic gas leakage
• Fire accidents
• Landslides
• Earthquakes
• High temperatures
• Worker falls
• Health risks

It then generates real-time alerts for workers and supervisors.

Advantages

• Simultaneous monitoring of worker health and environmental conditions.
• Early accident prediction using Machine Learning.
• Long-range underground communication through LoRa.
• Remote monitoring via IoT cloud and web dashboard.
• Immediate alerts for emergency situations.

System Workflow

The system operates as follows:

1. Sensors collect environmental and health data.
2. ESP32 processes and transmits the data.
3. WiFi uploads data to the cloud, while LoRa supports long-distance underground communication.
4. Machine Learning analyzes sensor readings and predicts hazardous conditions.
5. Results are displayed on an LCD, mobile app, and Streamlit dashboard.
6. Buzzer and dashboard notifications alert workers and supervisors during emergencies.

Main Modules

• Sensor Module: Measures gas concentration, temperature, humidity, fire, heart rate, SpO?, body temperature, and worker movement.
• ESP32 Module: Central controller that processes sensor data and manages communication.
• Communication Module: Uses WiFi and LoRa for reliable data transmission.
• Machine Learning Module: Predicts health risks and hazardous events using Random Forest and Gradient Boosting.
• IoT Cloud Module: Stores and visualizes real-time data using Blynk.
• Web Dashboard: Displays live readings, predictions, alerts, and historical data through Streamlit.
• Alert Module: Provides LCD displays, buzzer alarms, and dashboard notifications.

Testing and Results

The system underwent:

• Hardware testing
• Sensor calibration
• Communication testing
• Machine Learning model evaluation
• System integration testing
• Real-time performance testing

The prototype successfully monitored health and environmental parameters, detected unsafe events such as fire accidents, and displayed real-time information on both a web dashboard and the Blynk mobile application. The mobile app also showed live health data such as heart rate (64 BPM) and blood oxygen level (98% SpO?).

Conclusion

The LoRa-Based Smart System for Unstable Event Detection in Mining Industry Using Microcontroller provides an efficient, low-cost, and reliable solution for enhancing mining safety. 1) Compared to traditional wired and short-range wireless systems, the proposed model offers: 2) Greater communication range. 3) Lower maintenance cost. 4) Improved safety. 5) Scalability for large mining area

References

[1] A. Kumar and R. Singh, “IoT based coal mine safety monitoring and alerting system,” International Journal of Engineering Research & Technology (IJERT), vol. 9, no. 5, pp. 123–126, May 2020. [2] S. Li, L. D. Xu, and S. Zhao, “The internet of things: A survey,” Information Systems Frontiers, vol. 17, no. 2, pp. 243–259, Apr. 2015. [3] J. Wang, G. Song, and Y. Cui, “Underground coal mine monitoring system based on wireless sensor network,” IEEE Sensors Journal, vol. 19, no. 2, pp. 792–799, Jan. 2019. [4] P. S. Bhatia and R. Sharma, “Worker health monitoring system using IoT,” International Journal of Advanced Research in Computer Engineering, vol. 10, no. 3, pp. 45–50, Mar. 2021. [5] M. Hasan, M. Islam, and M. Rahman, “Gas detection and monitoring system using IoT,” International Journal of Scientific & Engineering Research, vol. 10, no. 6, pp. 789–793, Jun. 2019. [6] H. B. Kekre and T. K. Sarode, “Machine learning approach for predicting industrial hazards,” International Journal of Computer Applications, vol. 175, no. 12, pp. 22–26, Oct. 2020. [7] Espressif Systems, “ESP32 Technical Reference Manual,” Espressif Systems, 2022. [8] LI JG, Z. K. (2018). Intelligent mining technology for an underground metal mine based on unmanned equipment [J]. Engineering, 4(3), 381. [9] Aosong Electronics, “DHT11 Temperature and Humidity Sensor Datasheet,” 2020. [10] Maxim Integrated, “MAX30100 Pulse Oximeter and Heart-Rate Sensor Datasheet,” 2021. [11] Nan, W., & Xue-li, S. Research on WSN nodes location technology in coal mine. In 2009 International Forum on Computer Science-Technology and Applications (Vol. 3, pp. 232-234). (2009, December). IEEE. [12] Mohutsiwa, M., & Musingwini, C. Parametric estimation of capital costs for establishing a coal mine: South Africa case study. Journal of the Southern African Institute of Mining and Metallurgy, 115(8), (2015) 789-797. [13] Paraszczak, J., Gustafson, A., & Schunnesson, H. Technical and operational aspects of autonomous LHD application in metal mines. International Journal of Mining, Reclamation and Environment, 29(5), (2015) 391-403. [14] Li-min, Y., Anqi, L., Zheng, S., & Hui,L. (2008, October). Design of monitoring system for coal mine safety based on wireless sensor network. In 2008 IEEE/ASME International Conference on Mechtronic and Embedded Systems and Applications(pp. 409-414). IEEE. [15] Lynch, J. P., & Loh, K. J. (2006). A summary review of wireless sensors and sensor networks for structural health monitoring. Shock and vibration digest, 38(2), 91-130. [16] Zhang, H., Li, B., Karimi, M., Saydam, S., & Hassan, M. (2023). Recent Advancements in IoT Implementation for Environmental, Safety, and Production Monitoring in Underground Mines. IEEE Internet of Things Journal. [17] Nourali, H., & Osanloo, M. (2020). A regression-tree-based model for mining capital cost estimation. International Journal of Mining, Reclamation and Environment, 34(2), 88-100.

Copyright

Copyright © 2026 Anitha S, Priyanka G, Harini R, Sabareeshwaran A, Gnanasekaran K. 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.