Artificial Neural Network Based Prediction of PM2.5 Mass Concentration DOI

Carolina Dwi Sundari,

Aurellia Nadira Putri,

I.W. Kinanti

et al.

Published: Nov. 8, 2023

Economic growth and human activities affect the increase of particulate matter $(\text{PM}_{2.5})$ concentration. In addition to sources emissions, mass concentration xmlns:xlink="http://www.w3.org/1999/xlink">$\text{PM}_{2.5}$ may be impacted by meteorological factors such as temperature, humidity levels, atmospheric pressure, precipitation, speed/direction wind. A previous study established a monitoring system for air quality at Tokong Nanas (GKU) Deli Building Telkom University, located in Bandung, using microsensor technology. Various forecasting techniques were also employed predict , considering its factors. However, found that not all parameters give significant results forecasting, with an RMSE value 27 xmlns:xlink="http://www.w3.org/1999/xlink">$\mu\mathrm{g}/\mathrm{m}^{3}$ . Hence, this optimized Artificial Neural Network Backpropagation method. Only few taken into consider-ation system, which has impact on forecast quality, rainfall intensity, relative humidity, wind speed. As result, best network model 4-9-12-9-1 architecture learning 0.2, whereas is 4-20-9-9-1 0.3 value. The MAPE performances generated GKU models 8 xmlns:xlink="http://www.w3.org/1999/xlink">$\mu \mathrm{g}\mathrm{m}^{3}$ 37% 13 \mathrm{g}/\mathrm{m}^{3}$ 15%, respectively. Additional investigation required scrutinize conduct contaminated atmosphere tackle predicament purity Bandung Metropolitan forthcoming times.

Language: Английский

Use of machine learning to identify risk factors for insomnia DOI Creative Commons
Alexander A. Huang, Samuel Y. Huang

PLoS ONE, Journal Year: 2023, Volume and Issue: 18(4), P. e0282622 - e0282622

Published: April 12, 2023

Sleep is critical to a person's physical and mental health, but there are few studies systematically assessing risk factors for sleep disorders.The objective of this study was identify disorder through machine-learning assess methodology.A retrospective, cross-sectional cohort using the publicly available National Health Nutrition Examination Survey (NHANES) conducted in patients who completed demographic, dietary, exercise, health questionnaire had laboratory exam data.A physician diagnosis insomnia outcome study. Univariate logistic models, with as outcome, were used covariates that associated insomnia. Covariates p<0.0001 on univariate analysis included within final model. The machine learning model XGBoost due its prevalence literature well increased predictive accuracy healthcare prediction. Model ranked according cover statistic Shapely Additive Explanations (SHAP) utilized visualize relationship between these potential insomnia.Of 7,929 met inclusion criteria study, 4,055 (51% female, 3,874 (49%) male. mean age 49.2 (SD = 18.4), 2,885 (36%) White patients, 2,144 (27%) Black 1,639 (21%) Hispanic 1,261 (16%) another race. 64 out total 684 features found be significant (P<0.0001 used). These fitted into an AUROC 0.87, Sensitivity 0.77, Specificity 0.77 observed. top four highest by cover, measure percentage contribution covariate overall prediction, Patient Questionnaire depression survey (PHQ-9) (Cover 31.1%), 7.54%), recommendation exercise 3.86%), weight 2.99%), waist circumference 2.70%).Machine models can effectively predict laboratory, exam, lifestyle key factors.

Language: Английский

Citations

32
Assessing the impact of global carbon dioxide changes on atmospheric fluctuations in Iran through satellite data analysis DOI Creative Commons
Seyed Mohsen Mousavi, Naghmeh Mobarghaee Dinan, Saeed Ansarifard

et al.

Journal of Water and Climate Change, Journal Year: 2024, Volume and Issue: 15(6), P. 2774 - 2791

Published: April 3, 2024

ABSTRACT Atmospheric Carbon Dioxide (CO2), a significant greenhouse gas, drives climate change, influencing temperature, rainfall, and the hydrologic cycle. This alters precipitation patterns, intensifies storms, changes drought frequency timing of floods, impacting ecosystems, agriculture, water resources, societies globally. Understanding how global CO2 fluctuations impact regional atmospheric levels can inform mitigation strategies Facilitate resources management. The study investigates affect concentrations (XCO2) in Iran from 2015 to 2020, aiming against change. XCO2 data OCO-2 satellite surface flux Copernicus Atmosphere Monitoring Service (CAMS) were analyzed. Over 6 years, increased steadily by 12.66 ppm, mirroring rises. However, Iran's decreased, with slight increases anthropogenic emissions but decreased natural total fluxes. Monthly patterns exhibited variations, reaching its zenith spring dipping lowest point during summer, while peaked summer months. results reveal discrepancy between trends. While barely 2015–2020, fluxes decreased. over this period, indicating dominant rather than local factors on XCO2. Curbing worldwide gas output is imperative disrupt current trajectory Reporting plans, reducing combat warming minimize impacts

Language: Английский

Citations

10
Determining the influence of meteorological, environmental, and anthropogenic activity variables on the atmospheric CO2 concentration in the arid and semi-arid regions: A case study in the Middle East DOI
Seyed Mohsen Mousavi, Naghmeh Mobarghaee Dinan,

Korous Khoshbakht

et al.

Atmospheric Research, Journal Year: 2025, Volume and Issue: unknown, P. 108009 - 108009

Published: Feb. 1, 2025

Language: Английский

Citations

1
Spatial Distribution of Particulate Matter in Iran from Internal Factors to the Role of Western Adjacent Countries from Political Governance to Environmental Governance DOI
Faezeh Borhani, Ali Asghar Pourezzat,

Amir Houshang Ehsani

et al.

Earth Systems and Environment, Journal Year: 2024, Volume and Issue: 8(1), P. 135 - 164

Published: Jan. 1, 2024

Language: Английский

Citations

7
Estimate Ground-based PM2.5 concentrations with Merra-2 aerosol components in Tehran, Iran: Merra-2 PM2.5 concentrations verification and meteorological dependence DOI
Faezeh Borhani,

Amir Houshang Ehsani,

Majid Shafiepour Motlagh

et al.

Environment Development and Sustainability, Journal Year: 2023, Volume and Issue: 26(3), P. 5775 - 5816

Published: Feb. 8, 2023

Language: Английский

Citations

15
Examining the Role of the Main Terrestrial Factors Won the Seasonal Distribution of Atmospheric Carbon Dioxide Concentration over Iran DOI
Seyed Mohsen Mousavi, Naghmeh Mobarghaee Dinan, Saeed Ansarifard

et al.

Journal of the Indian Society of Remote Sensing, Journal Year: 2023, Volume and Issue: 51(4), P. 865 - 875

Published: Feb. 19, 2023

Language: Английский

Citations

15
Current Status and Future Forecast of Short-lived Climate-Forced Ozone in Tehran, Iran, derived from Ground-Based and Satellite Observations DOI Open Access
Faezeh Borhani, Majid Shafiepour Motlagh,

Amir Houshang Ehsani

et al.

Water Air & Soil Pollution, Journal Year: 2023, Volume and Issue: 234(2)

Published: Feb. 1, 2023

Language: Английский

Citations

12
Examining and predicting the influence of climatic and terrestrial factors on the seasonal distribution of ozone column depth over Tehran province using satellite observations DOI
Faezeh Borhani,

Amir Houshang Ehsani,

Savannah L. McGuirk

et al.

Acta Geophysica, Journal Year: 2023, Volume and Issue: 72(2), P. 1191 - 1226

Published: Oct. 3, 2023

Language: Английский

Citations

12
Machine learning models for modeling the biosorption of Fe(III) ions by activated carbon from olive stone DOI
Ayman Massaoudi, Fraj Echouchene, Mossaad Ben Ayed

et al.

Neural Computing and Applications, Journal Year: 2024, Volume and Issue: 36(21), P. 13357 - 13372

Published: May 7, 2024

Language: Английский

Citations

4
Combined Prediction of PM10 Concentration at Smart Construction Sites Based on Quadratic Mode Decomposition and Deep Learning DOI Open Access
Xiaogang Li, Xin Li,

Kaikai Kang

et al.

Sustainability, Journal Year: 2025, Volume and Issue: 17(2), P. 616 - 616

Published: Jan. 15, 2025

The accurate prediction of PM10 concentrations at smart construction sites is crucial for improving urban air quality, protecting public health, and advancing sustainable development in the industry. are influenced by interaction intensity environmental meteorological factors, resulting nonlinear volatile data. To improve accuracy, this paper presents a two-stage mode decomposition method that integrates Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) Variational (VMD). This combined Bidirectional Long Short-Term Memory (BiLSTM) neural network, optimized using Sparrow Search Algorithm (SSA), to establish hybrid model forecasting sites. Initially, CEEMDAN decomposes original sequence into several Intrinsic Functions (IMFs). sample entropy each component then calculated, K-means clustering used group them. VMD applied further decompose high-frequency components obtained after clustering. SSA employed optimize parameters BiLSTM which models all predictive model. predicted values aggregated generate final forecast. Real-time monitoring data from Construction Site A Nanjing case study validation. empirical results demonstrate proposed outperforms comparison on evaluation metrics, offering scientific foundation automated dust reduction decision-making sites, thereby facilitating shift toward greener, smarter, more digitized practices.

Language: Английский

Citations

0