IoT-enabled Particulate Matter Monitoring and Forecasting Method based on Cluster Analysis

Author(s):  
Jaeseok Yun ◽  
Jiyoung Woo
Author(s):  
Mohammad Azimi ◽  
Yasin Mansouri ◽  
Hamideh Mihanpour ◽  
Vida Rezai Hachasu ◽  
Morteza Mohammad Zadeh ◽  
...  

Background: Particulate matter air pollution is one of the most important risk factors for cardiovascular and respiratory diseases. By increasing the number of mineral industries in the two past decades, workers in these industries are exposed to pathogenic respirable particulate matter pollutants. Cluster analysis is a multivariate statistical analysis method. Clustering creates groups or classes that the difference between the sub-groups samples is less than the difference between the groups. Therefore, this study assigns the cluster analysis to air sampling data collected from the various units of a tile factory. Methods: In this observational study, sampling from the respiratory zone of 93 workers in a tile and ceramic factory for both respirable and inhalable particles were performed. Sampling of inhalable particle based on NIOSH_0500 protocol and respirable particles based on NIOSH_0600 was conducted. Data were analyzed by both R 3.2.2 software and hierarchical cluster analysis with Ward link. Results: 92.47% of Workers were exposed to respirable particles less than TLV and 39.8% of them were exposed to inhalable particles more than TLV. The maximum average exposure for respirable particles 13.04 mg/m3 and inhalable particles 84.88 mg/m3 is respectively reported for crusher unit. The lowest average exposures to respirable (0.41 mg/m3) and inhalable (min=1.74 mg/m3) particles were observed in the glaze line division. Conclusion: Since the workers are exposed to concentrations more than the threshold limit value of respirable particles, and especially inhalable particles in some units, appropriate control measures must be considered to prevent possible consequences


Author(s):  
Yusuf Aina ◽  
Elhadi Adam ◽  
Fethi Ahmed

The study of the concentrations and effects of fine particulate matter in urban areas have been of great interest to researchers in recent times. This is due to the acknowledgment of the far-reaching impacts of fine particulate matter on public health. Remote sensing data have been used to monitor the trend of concentrations of particulate matter by deriving aerosol optical depth (AOD) from satellite images. The Center for International Earth Science Information Network (CIESIN) has released the second version of its global PM2.5 data with improvement in spatial resolution. This paper revisits the study of spatial and temporal variations in particulate matter in Saudi Arabia by exploring the cluster analysis of the new data. Cluster analysis of the PM2.5 values of Saudi cities is performed by using Anselin local Moran’s I statistic. Also, the analysis is carried out at the regional level by using self-organizing map (SOM). The results show an increasing trend in the concentrations of particulate matter in Saudi Arabia, especially in some selected urban areas. The eastern and south-western parts of the Kingdom have significantly clustering high values. Some of the PM2.5 values have passed the threshold indicated by the World Health Organization (WHO) standard and targets posing health risks to Saudi urban population.


2018 ◽  
Vol 236 ◽  
pp. 591-597 ◽  
Author(s):  
Hsiao-Chi Chuang ◽  
Ruei-Hao Shie ◽  
Chia-Pin Chio ◽  
Tzu-Hsuen Yuan ◽  
Jui-Huan Lee ◽  
...  

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3223 ◽  
Author(s):  
Sachit Mahajan ◽  
Ling-Jyh Chen ◽  
Tzu-Chieh Tsai

Air pollution is a global problem and can be perceived as a modern-day curse. One way of dealing with it is by finding economical ways to monitor and forecast air quality. Accurately monitoring and forecasting fine particulate matter (PM2.5) concentrations is a challenging prediction task but Internet of Things (IoT) can help in developing economical and agile ways to design such systems. In this paper, we use a historical data-based approach to perform PM2.5 forecasting. A forecasting method is developed which uses exponential smoothing with drift. Experiments and evaluation were performed using the real-time PM2.5 data obtained from large scale deployment of IoT devices in Taichung region in Taiwan. We used the data from 132 monitoring stations to evaluate our model’s performance. A comparison of prediction accuracy and computation time between the proposed model and three widely used forecasting models was done. The results suggest that our method can perform PM2.5 forecast for 132 monitoring stations with error as low as 0.16 μ g/ m 3 and also with an acceptable computation time of 30 s. Further evaluation was done by forecasting PM2.5 for next 3 h. The results show that 90 % of the monitoring stations have error under 1.5 μ g/ m 3 which is significantly low.


Sign in / Sign up

Export Citation Format

Share Document