Novel hybrid extreme learning machine and multi-objective optimization algorithm for air pollution prediction

Bai, Lu; Zhi, Liu; Wang, Jianzhou

doi:10.1016/j.apm.2022.01.023

Cited by 26 publications

(7 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The results found, for the three ML models applied and for each set of features selected, reveal accuracies > 87% and precisions > 86%. In (Bai et al, 2022), for accurately predicting the PM2.5 concentrations, the authors proposed a hybrid model based on a deterministic prediction module introduced as Random Fourier Extreme Learning Machine (RF-ELM) combined with an interval prediction module. This later was used to provide effective intervals of pollutant concentrations based on the upper and lower bounds determined in the deterministic prediction stage.…”

Section: Introductionmentioning

confidence: 99%

Air pollution forecasting and modelling using hybrid machine learning for Craiova City in Romania

Udriștioiu

MGHOUGHI

Yıldızhan

2022

Preprint

View full text Add to dashboard Cite

Poor air quality affects human health and exacerbates climate change, inducing temperature variations. Therefore, it is important to understand at local level the interactions between climate variability and air pollution to mitigate the effects of air pollution on human health. The aim of this paper is to explore the correlations between meteorological parameters, temperature (T), humidity (H) and air pressure (P) and particulate matter concentrations (PM1, PM2.5, PM10). Also, it is investigated the correlation between PM 2.5 and PM10, between noise, CO2, and PM concentrations. For this purpose, five hybrid Machine Learning models were employed to forecast the PMs concentrations and then the Air Quality Index (AQI). The data set was provided by an independent network of sensors for the period of September 22, 2021 – February 17, 2022. In the results we found, R² exceeds, in general, 0.96 and, in most cases, they can reach 0.99. The humidity was found to be the less significant variable on the PMs concentrations, and the best accuracy was found by combining the pressure with the temperature. In addition, PM10 concentrations were found effectively related to the PM2.5 concentrations and with a bit strongly to the PM1. The PM10 concentration is not strongly correlated to the Noise and CO2 data. Finally, several new relationships have been built to forecast the PMs concentrations and the AQI based on the best combinations of predictor variables found.

show abstract

Section: Introductionmentioning

confidence: 99%

Air pollution forecasting and modelling using hybrid machine learning for Craiova City in Romania

Udriștioiu

MGHOUGHI

Yıldızhan

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Bai and Lu et al suggested a new method for deterministic and interval pollutant concentration forecasting [17]. This approach predicts air pollutant concentration intervals with enhanced deterministic accuracy.…”

Section: Related Workmentioning

confidence: 99%

Developing the Optimal Hybrid Neural Network for Predicting the Factor of Air Pollutants

Viswanathan

2024

Preprint

View full text Add to dashboard Cite

Urban air pollution can be reduced via precise air pollutant forecasts.For that, the air quality index (AQI) quantifies air quality.In this manner, accurate and trustworthy air quality index (AQI) estimates are essential for preserving the natural environment and the general population's health. Using the backpropagation (BP) algorithm, this study describes a method for enhancing the performance of neural networks. Using a network optimized with natural swarm intelligence, a novel optimal-hybrid model approachto Nature Swarm Intelligence (NSI), predicting the Air Quality Index (AQI), is possible. This NSI comprises the optimization algorithms Dove Swarm optimization (DSA) and Bat Algorithm (BA), which aim to optimize the weight of the Backpropagation neural network (BPNN) to promote the air quality prediction. The constructed optimal-hybrid modelcaptured the characteristics of the AQI series and produced a more accurate AQI forecast according to exhaustive comparisons using a set of evaluation indicators. Experiments conducted verify the proposed modelis validfor application when attempting to forecast the AQI. This is because it receives a high RMSE, MAPE, Error Absolute total, and Accuracy value from the simulation. This is because the simulation results suggest that the network model could be a good option for actualization, which is why this is the case.

show abstract

“…In [ 14 ], the focus was on accurately predicting PM2.5 concentrations. The authors introduced a hybrid model comprising a deterministic prediction module and a Random Fourier Extreme Learning Machine (RF-ELM), combined with an interval prediction module.…”

Section: Introductionmentioning

confidence: 99%

Multivariable Air-Quality Prediction and Modelling via Hybrid Machine Learning: A Case Study for Craiova, Romania

El Mghouchi,

Udristioiu,

Yildizhan

2024

Sensors

View full text Add to dashboard Cite

Inadequate air quality has adverse impacts on human well-being and contributes to the progression of climate change, leading to fluctuations in temperature. Therefore, gaining a localized comprehension of the interplay between climate variations and air pollution holds great significance in alleviating the health repercussions of air pollution. This study uses a holistic approach to make air quality predictions and multivariate modelling. It investigates the associations between meteorological factors, encompassing temperature, relative humidity, air pressure, and three particulate matter concentrations (PM10, PM2.5, and PM1), and the correlation between PM concentrations and noise levels, volatile organic compounds, and carbon dioxide emissions. Five hybrid machine learning models were employed to predict PM concentrations and then the Air Quality Index (AQI). Twelve PM sensors evenly distributed in Craiova City, Romania, provided the dataset for five months (22 September 2021–17 February 2022). The sensors transmitted data each minute. The prediction accuracy of the models was evaluated and the results revealed that, in general, the coefficient of determination (R2) values exceeded 0.96 (interval of confidence is 0.95) and, in most instances, approached 0.99. Relative humidity emerged as the least influential variable on PM concentrations, while the most accurate predictions were achieved by combining pressure with temperature. PM10 (less than 10 µm in diameter) concentrations exhibited a notable correlation with PM2.5 (less than 2.5 µm in diameter) concentrations and a moderate correlation with PM1 (less than 1 µm in diameter). Nevertheless, other findings indicated that PM concentrations were not strongly related to NOISE, CO2, and VOC, and these last variables should be combined with another meteorological variable to enhance the prediction accuracy. Ultimately, this study established novel relationships for predicting PM concentrations and AQI based on the most effective combinations of predictor variables identified.

show abstract

Novel hybrid extreme learning machine and multi-objective optimization algorithm for air pollution prediction

Cited by 26 publications

References 38 publications

Air pollution forecasting and modelling using hybrid machine learning for Craiova City in Romania

Air pollution forecasting and modelling using hybrid machine learning for Craiova City in Romania

Developing the Optimal Hybrid Neural Network for Predicting the Factor of Air Pollutants

Multivariable Air-Quality Prediction and Modelling via Hybrid Machine Learning: A Case Study for Craiova, Romania

Contact Info

Product

Resources

About