Spatiotemporal PM2.5 forecasting via dynamic geographical Graph Neural Network

Environmental Modelling & Software, Journal Year: 2025, Volume and Issue: 186, P. 106351 - 106351

Published: Feb. 6, 2025

Language: Английский

---

Exploring the significance of temporal, meteorological, and previous states parameters in $$\hbox {PM}_{2.5}$$ concentration predictions: a neural network sensitivity study for Aguascalientes, Mexico

Modeling Earth Systems and Environment, Journal Year: 2025, Volume and Issue: 11(3)

Published: March 8, 2025

Language: Английский

---

Multiscale coherence analysis of PM2.5 and PM10 concentrations of four Indian urban areas using wavelet transform

Urban Climate, Journal Year: 2024, Volume and Issue: 56, P. 102056 - 102056

Published: July 1, 2024

Language: Английский

---

A Short-Term Air Pollutant Concentration Forecasting Method Based on a Hybrid Neural Network and Metaheuristic Optimization Algorithms

Sustainability, Journal Year: 2024, Volume and Issue: 16(11), P. 4829 - 4829

Published: June 5, 2024

In the contemporary era, global air quality has been adversely affected by technological progress, urban development, population expansion, and proliferation of industries power plants. Recognizing urgency addressing pollution consequences, prediction concentration levels pollutants become crucial. This study focuses on short-term nitrogen dioxide (NO2) sulfur (SO2), prominent emitted Kerman Combined Cycle Power Plant, from May to September 2019. The proposed method utilizes a new two-step feature selection (FS) process, hybrid neural network (HNN), Coot optimization algorithm (COOT). combination FS COOT selects most relevant input features while eliminating redundant ones, leading improved accuracy. application HNN for training further enhances accuracy significantly. To assess model’s performance, two datasets, including real data different parts Plant in Kerman, Iran, 1 2019 30 (namely dataset A B), are utilized. Subsequently, mean square error (MSE), absolute (MAE), root deviation (RMSE), percentage (MAPE) were employed obtain FS-HNN-COOT. Experimental results showed MSE FS-HNN-COOT NO2 ranged 0.002 0.005, MAE 0.016 0.0492, RMSE 0.0142 0.0736, MAEP 4.21% 8.69%. Also, MSE, MAE, RMSE, MAPE 0.0001 0.0137, 0.0108 0.0908, 0.0137 0.1173, 9.03% 15.93%, respectively, SO2.

Language: Английский

---

Deep learning PM 2.5 hybrid prediction model based on clustering- secondary decomposition strategy

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 29, 2024

Accurate prediction of PM_2.5 concentration is important for pollution control, public health and ecological protection. To this end, paper proposes a deep learning hybrid model based on clustering secondary decomposition, aiming to achieve accurate concentration. The utilizes the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) decompose sequences into multiple intrinsic modal function components (IMFs), clusters re-fuses sub-sequences similar complexity by permutation entropy (PE) K-means clustering. For fused high-frequency performed using whale optimization algorithm (WOA) optimized variational (VMD). Finally, two basic frameworks combined long short-term memory neural network (LSTM). Experiments show that proposed exhibits good stability generalization ability. It does not only make predicts in short term, but also captures trends long-term prediction. There significant performance improvement over four baseline models. Further comparisons existing models outperform current state-of-the-art

Language: Английский

---

A Deep Learning PM2.5 Hybrid Prediction Model Based on Clustering–Secondary Decomposition Strategy

Electronics, Journal Year: 2024, Volume and Issue: 13(21), P. 4242 - 4242

Published: Oct. 29, 2024

Accurate prediction of PM2.5 concentration is important for pollution control, public health, and ecological protection. However, due to the nonlinear nature data, accuracy existing methods suffers performs poorly in both short-term long-term predictions. In this study, a deep learning hybrid model based on clustering quadratic decomposition proposed. The utilizes complete ensemble empirical mode with adaptive noise (CEEMDAN) decompose sequences into multiple intrinsic modal function components (IMFs), clusters re-fuses subsequences similar complexity by permutation entropy (PE) K-means clustering. For fused high-frequency sequences, secondary performed using whale optimization algorithm (WOA) optimized variational (VMD). Finally, temporal features are captured long- memory neural network (LSTM). Experiments show that proposed exhibits good stability generalization ability. It does not only make accurate predictions short term, but also captures trends prediction. There significant performance improvement over baseline models. Further comparisons models outperform current state-of-the-art

Language: Английский

---

Particulate Matter 2.5 concentration prediction system based on uncertainty analysis and multi-model integration

The Science of The Total Environment, Journal Year: 2024, Volume and Issue: 958, P. 177924 - 177924

Published: Dec. 9, 2024

Language: Английский

---