PM2.5 concentration prediction using deep learning in internet of things air monitoring system

Bei, Wei; Li, Fengying

doi:10.4491/eer.2021.456

Cited by 7 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Also, the smoothness of the junction points is ensured by connecting them with second-order continuous derivatives. As a result, cubic spline curve interpolation can maximally simulate the evolution process of gradual change of physical phenomena, leading to the ideal interpolation effect [48].…”

Section: Day Month Year Hourmentioning

confidence: 99%

A Prediction Hybrid Framework for Air Quality Integrated with W-BiLSTM(PSO)-GRU and XGBoost Methods

Chang,

Chen,

et al. 2023

Sustainability

View full text Add to dashboard Cite

Air quality issues are critical to daily life and public health. However, air quality data are characterized by complexity and nonlinearity due to multiple factors. Coupled with the exponentially growing data volume, this provides both opportunities and challenges for utilizing deep learning techniques to reveal complex relationships in massive knowledge from multiple sources for correct air quality prediction. This paper proposes a prediction hybrid framework for air quality integrated with W-BiLSTM(PSO)-GRU and XGBoost methods. Exploiting the potential of wavelet decomposition and PSO parameter optimization, the prediction accuracy, stability and robustness was improved. The results indicate that the R2 values of PM2.5, PM10, SO2, CO, NO2, and O3 predictions exceeded 0.94, and the MAE and RMSE values were lower than 0.02 and 0.03, respectively. By integrating the state-of-the-art XGBoost algorithm, meteorological data from neighboring monitoring stations were taken into account to predict air quality trends, resulting in a wider range of forecasts. This strategic merger not only enhanced the prediction accuracy, but also effectively solved the problem of sudden interruption of monitoring. Rigorous analysis and careful experiments showed that the proposed method is effective and has high application value in air quality prediction, building a solid framework for informed decision-making and sustainable development policy formulation.

show abstract

Section: Day Month Year Hourmentioning

confidence: 99%

A Prediction Hybrid Framework for Air Quality Integrated with W-BiLSTM(PSO)-GRU and XGBoost Methods

Chang,

Chen,

et al. 2023

Sustainability

View full text Add to dashboard Cite

show abstract

“…The spatiotemporal changes in PM2.5, owing to its dynamic and nonlinear nature, are influenced by meteorological variables, temporal features and other primary pollutants (Kristiani, 2022) (Bai and Li, 2023;) (Narkhede, 2023;Zhang, 2023). Classical models of PM2.5 prediction often rely solely on a single variable, namely PM2.5 concentration, and overlook the impact of meteorological variables and other pollutants (Liu, 2021).…”

Section: Introductionmentioning

confidence: 99%

“…Common statistical methods such as mean, median (Nguyen, M. H. et al, 2021), linear interpolation, and spline interpolation (Bai and Li, 2023) are frequently employed in studies due to their simplicity. The mean and median methods involve using a single value for missing data, resulting in reduced statistical power and potential bias (Acock, 2005).…”

Section: Introductionmentioning

confidence: 99%

A New Attention-Based a convolutional neural network-gated recurrent unit Model for Spatial-Temporal PM2.5 Prediction

Haghbayan,

Momeni,

Tashayo

2024

Preprint

View full text Add to dashboard Cite

Accurately predicting the spatiotemporal distribution of PM2.5 using real data is a challenge, as missing data must be imputed and a suitable modeling method selected. When imputing missing data, the relationships between the variables must be taken into account while preserving their inherent variability and uncertainty. In this study, the imputation of missing data considering the relationships between meteorological variables and other pollutants was performed using different machine learning methods. After handling the missing data, the selection of an appropriate modeling method becomes imperative. In this paper, we attempt to present an innovative spatiotemporal hybrid model based on GRU that aims to obtain and extract spatiotemporal relationships. A dataset collected from 2017 to 2023 in Tehran metropolis was used for the study. The proposed model with (R2 = 0.84) can well determine the changes in PM2.5 concentration for each station and visualize the level of air pollution on a diffusion map.

show abstract

“…One of the most harmful contaminants in the air is PM2.5, which is a type of a small particulate matter that has a diameter of less than 2.5 micrometers [3][4]. PM2.5 particles are incredibly lightweight and small, which allows them to remain suspended in the atmosphere for a considerable amount of time [5]. They are produced by a variety of sources, including motor vehicles, power plants, factories, and wildfires, and can be dispersed over enormous area by wind and atmospheric conditions [6].…”

Section: Introductionmentioning

confidence: 99%

“…5 propagation was presented. This study shows how a Lagrangian trajectory model with short path distance can be used to estimate PM2.5 concentrations by introducing a new mathematical method.…”

mentioning

confidence: 99%

Short Path Wind-Field Distance-Based Lagrangian Trajectory Model for Enhancing Atmospheric Dispersion Prediction Accuracy

R’Bigui,

Cho

2023

IEEE Access

View full text Add to dashboard Cite

Air pollution is a major global issue that not only threatens the safety of our planet but also poses risks to global health. Weather plays a crucial role in the rapid dispersion of air pollution. Various models have been used to predict air pollution; however, atmospheric pollution dispersion remains unpredictable, especially in relation to meteorological conditions. Our research scope focuses on developing an Air Diffusion Model using Future Wind and Pollutant sensing data to enhance prediction accuracy. In this paper, we present a new approach based on a mathematical model named the Short Path Distance based Lagrangian Trajectory Model (SPD-LTM). This model utilizes a trajectory approach and short path wind-field distance optimization to predict future air dispersion using pollutant sensing data. The framework developed in this work aims to model changes in Particulate Matter (PM2.5) and predict its concentration based on short path distance and time dependencies. The Lagrangian trajectory and concentration calculations are performed using the Hybrid Single-Particulate Lagrangian Integrated Trajectory algorithm (HYSPLIT). Then, we apply the short path distance algorithm using the Dijkstra algorithm. The obtained results demonstrate that the SPD-LTM outperforms the usual LTM and provides better accuracy to our predictive model.

show abstract

PM2.5 concentration prediction using deep learning in internet of things air monitoring system

Cited by 7 publications

References 19 publications

A Prediction Hybrid Framework for Air Quality Integrated with W-BiLSTM(PSO)-GRU and XGBoost Methods

A Prediction Hybrid Framework for Air Quality Integrated with W-BiLSTM(PSO)-GRU and XGBoost Methods

A New Attention-Based a convolutional neural network-gated recurrent unit Model for Spatial-Temporal PM2.5 Prediction

Short Path Wind-Field Distance-Based Lagrangian Trajectory Model for Enhancing Atmospheric Dispersion Prediction Accuracy

Contact Info

Product

Resources

About