Cloud Hosting

In mining operations, the generation of dust is a frequent phenomenon, leading to the presence of airborne dust suspended in the mine atmosphere. This airborne dust primarily comprises mineral particles and, in the presence of moisture, gives rise to particulate matter, which consists of a complex mixture of solid and liquid components. The size of these particles ranges from 10 to 2.5 , rendering them invisible to the naked eye. Inhalation of such particles can pose significant health hazards to workers, especially upon chronic exposure. Particulate matter is composed of a combination of organic and inorganic particles, including dust, pollen, soot, smoke, and liquid droplets, making it extremely hazardous to human respiratory health. Thus, monitoring the levels of particulate matter in mining sites is of utmost importance for ensuring the safety and well-being of workers. This monitoring plays a vital role in the prevention and prediction of health hazards associated with inhalation12.

Understanding particulate matter (PM) in mining environments is essential for recognizing its sources, characteristics, and the potential impact it poses in the mining context. PM originates from both natural and anthropogenic sources, encompassing sea salt, pollen, volcanic eruptions, airborne dust, and various industrial activities. Among industrial operations, mining significantly contributes to PM emissions due to processes such as drilling, blasting, transportation, and handling of materials. Drilling operations generate suspended airborne dust particles, while blasting releases particles and gas emissions, including NOx, which can pose health risks13. Additionally, open-pit coal mining contributes to elevated PM levels, facilitated by wind-driven dispersion of coal dust, necessitating the implementation of effective mitigation strategies to safeguard human health and the environment14.

Based on their formation mechanisms, PM is classified into various types, including dust, smoke, fumes, fly ash, mist, and spray (see Table 1). These different PM types exhibit distinct size ranges, with fine and ultrafine particles capable of reaching the alveoli in the respiratory system, while PM10-sized particles primarily settle in the upper airways. The percentage of inhaled airborne particles that enter the respiratory tract is represented by total inhalable dust15. Other measures, such as thoracic and respirable dust, refer to particles that pass through the larynx into the thoracic cavity and reach the gas exchange region of the lungs, respectively. Hazardous dusts can also chemically interact with the respiratory system, allowing toxic substances like lead and arsenic to pass through alveolar walls into the bloodstream16. A comprehensive understanding of these PM classifications is crucial for assessing their impact on human health.

Exposure to particulate matter (PM) poses significant health risks to miners, as they inhale ambient air in their workplace. PM's mineralogical composition can lead to severe health issues, such as asbestosis and silicosis3. Effective monitoring of PM is crucial not only for environmental permits and planning but also for safeguarding miners' health. However, current monitoring systems in mining areas encounter limitations, necessitating the implementation of fast and accurate air monitoring systems. Inadequate monitoring of PM dust concentration (ranging from PM 2.5 to PM 10 ) can lead to worker exposure and various health complications, including respiratory problems, lung diseases, breathing difficulties, non-fatal heart attacks, and cardiac arrhythmias. Therefore, comprehensive and precise monitoring systems are essential for ensuring the well-being of miners17,18.

Monitoring particulate matter (PM) in mining sites involves collecting air quality data while considering wind direction. This monitoring can be divided into three parts: (1) monitoring the mine atmosphere away from equipment operations but within the site, (2) monitoring PM dust at operating sites, including drilling, blasting, loading, transportation, and facilities, and (3) monitoring PM dust outside the mining area19.

In mining operations, various dust-forming activities occur at different locations, necessitating the monitoring of particulate matter (PM) concentrations at multiple sites. The rapid advancement of Internet of Things (IoT) technology has led to the development of IoT-based PM monitoring systems, which serve as a promising alternative to traditional monitoring methods20. Conventional monitoring systems often require significant human intervention, are time-consuming, and may result in manual errors, emphasizing the need for improved monitoring solutions. IoT-based PM monitoring systems collect data through measurement devices (sensors) and transmit it via the network, making them more efficient and reliable. These systems are designed to enable mine operators to promptly inspect dust-causing sites and implement necessary preventive measures. To be effective, these systems should be easy to install at multiple sites and exhibit sufficient endurance, considering that the main dust-generating areas may change over time, and workers are exposed to harsh outdoor conditions during mining operations. This study investigates the performance of IoT measurement devices and the network in existing operations, including an open-pit mine site.

A multitude of studies has explored the application of the Internet of Things (IoT) in tracking traffic flow and monitoring air quality. For instance, a study in 2022 introduced an inexpensive IoT-based system for tracking traffic flow and determining the air quality index (AQI)21. This study utilized machine learning methods, which eliminated the need for complex calibration, allowing the measurement of pollutant gases and accurate determination of AQI. Similarly, another study in 2020 demonstrated an IoT-based indoor air quality monitoring platform, storing data in the cloud and providing resources for further indoor air quality studies22.

In line with this, researchers in 2020 developed an IoT system for monitoring air quality, capable of monitoring local air quality and providing data for user analysis via an integrated buzzer23. Additionally, another study in 2020 discussed the use of IoT in the mining field, highlighting how IoT serves as a wireless network for collecting information from electronic devices and sensors24.

Over the past decade, advances in wireless sensor networks (WSN), radio frequency identification (RFID), and cloud computing have facilitated the integration of the Internet of Things (IoT) in harsh work environments like mining25. This integration has significantly improved the accuracy, efficiency, cost-effectiveness, and real-time capabilities of the monitoring process. Notably, these advancements have enabled automatic event detection, control, and remote data exchange, making monitoring feasible in otherwise inaccessible locations. Several successful implementations of WSN-based monitoring systems have been reported, such as early detection of fires in coal mines and detection of toxic mine gases in the environment. Furthermore, IoT technology has enabled the accurate measurement of particulate matter within a short time. Given that time and cost are crucial factors in managing these projects, this work aims to develop a low-cost IoT-based PM monitoring device capable of monitoring pollutants of less than 2.5 . By utilising these technologies, mining operations can be made safer and more efficient, while simultaneously reducing costs and environmental impacts26.

Numerous studies have proposed various machine learning algorithms for the prediction of airborne particulate matter. Li et al. introduced a real-time prediction approach based on weighted extreme learning machine (WELM) and adaptive neuro-fuzzy inference system (ANFIS)27. Choubin et al. developed machine learning models, including Random Forest (RF), Bagged Classification and Regression Trees (Bagged CART), and Mixture Discriminant Analysis (MDA), for forecasting PM10-induced risk28. Rutherford et al. utilized excitation-emission matrix (EEM) fluorescence spectroscopy and a machine learning algorithm to localize PM sources29.

In the context of PM2.5 prediction, Just et al. proposed a new strategy using machine learning techniques30. Yang et al. put forward hybrid models by combining different deep learning approaches31. Stirnberg et al. developed a method integrating satellite-based Aerosol Optical Depth (AOD) with meteorological and land use factors for predicting PM10 concentrations32. Additionally, Gilik et al. constructed a supervised model for air pollution prediction using sensor data and explored model transferability between cities33.These comprehensive studies collectively demonstrate the potential and effectiveness of machine learning in air pollution prediction, providing valuable insights for future research and applications in this field.

In the context of this literature review, the section prominently highlights the novelty and scientific contribution of the current research workan IoT-based monitoring and ML powered dust prediction system. The proposed system not only offers real-time monitoring of various PM particle sizes, including PM1.0, PM2.5, PM4.0, and PM10.0, but also integrates a efficient prediction model to ensure precise and accurate PM measurements. With hardware integration and robust software protocols, the system addresses the limitations of traditional monitoring techniques, facilitating efficient and comprehensive monitoring of PM dust concentration in mining environments. This research work aims to significantly contribute for improving mine air quality by effectively monitored and prediction of PM dust pollution in surface mine sites by utilising the cutting edge technology like IoT and ML. The proposed IoT-based Dust Monitoring System stands as a novel and practical solution that advances the field of air quality monitoring and holds promising potential for widespread implementation in mining and beyond.

Read more:
Integrated smart dust monitoring and prediction system for surface mine sites using IoT and machine learning ... - Nature.com