Review
Green & Sustainable Science & Technology
Adil Masood, Kafeel Ahmad
Summary: Accurate air quality forecasting is crucial for pollution control and public health. Traditional forecasting techniques lack consistent accuracy, leading to increased interest in AI-based methods. This study reviews the most commonly used AI techniques for air pollution forecasting and identifies Deep Neural Networks as the best performing method.
JOURNAL OF CLEANER PRODUCTION
(2021)
Article
Computer Science, Interdisciplinary Applications
Sheen Mclean Cabaneros, Ben Hughes
Summary: The use of data-driven techniques, such as artificial neural network (ANN) models, for outdoor air pollution forecasting has been popular in the past two decades. However, research on the uncertainty surrounding the development of ANN models has been limited. This review outlines the approaches for addressing model uncertainty and reveals that input uncertainty has received the most attention, while structure, parameter, and output uncertainties have been less focused on. Ensemble approaches, particularly neuro-fuzzy networks, have been widely employed, but the direct measurement of uncertainty has received less attention. The study also suggests the need for development and application of approaches that can handle and quantify uncertainty in ANN model development.
ENVIRONMENTAL MODELLING & SOFTWARE
(2022)
Article
Engineering, Environmental
M. I. Rodriguez-Garcia, M. C. Ribeiro Rodrigues, J. Gonzalez-Enrique, J. J. Ruiz-Aguilar, I. J. Turias
Summary: The main aim of this research is to accurately predict pollutant concentrations associated with maritime traffic in the Bay of Algeciras in southern Spain. The study analyzes different scenarios and uses databases of air pollution, meteorological measurements, and maritime traffic to develop prediction models. These models have been compared using various classification indexes and the best models have been selected based on their performance. The results demonstrate the potential of the models to forecast air pollution and support decision-making regarding air quality.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2023)
Article
Environmental Sciences
Stephanie Lima Jorge Galvao, Junia Cristina Ortiz Matos, Yasmin Kaore Lago Kitagawa, Flavio Santos Conterato, Davidson Martins Moreira, Prashant Kumar, Erick Giovani Sperandio Nascimento
Summary: Concerns about air pollution in urban areas have grown worldwide, especially regarding PM2.5, which can have adverse effects on human health, particularly children. Deterministic and stochastic models are key tools for predicting atmospheric behavior and guiding preventive actions against air pollution. However, deterministic models have limitations, leading to an increasing interest in deep learning for its simpler implementation and success. This study aims to develop and evaluate the performance of different deep artificial neural network (DNN) topologies in predicting PM2.5 concentrations, considering feature augmentation through discrete wavelet transform (DWT).
Article
Computer Science, Information Systems
K. U. Jaseena, Binsu C. Kovoor
Summary: Weather forecasting is the practice of predicting the state of the atmosphere based on different weather parameters. Accurate weather forecasts are crucial in various fields. With the advancement of atmospheric observing systems and the increasing volume of weather data, deep learning techniques are being used to improve weather prediction. This paper provides a comprehensive review of weather forecasting approaches and discusses potential future research directions.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Physics, Multidisciplinary
Dimitrios Nikolopoulos, Aftab Alam, Ermioni Petraki, Michail Papoutsidakis, Panayiotis Yannakopoulos, Konstantinos P. Moustris
Summary: This paper investigates a 17-year PM10 time series from five stations in Athens, Greece using statistical and entropy methods to analyze existing stochastic and self-organisation trends. It finds decreasing trends at all stations and explores self-organization through Boltzmann and Tsallis entropy in selected parts. The study identifies non-stochastic areas with fractal, long-memory, and self-organisation patterns, combining multiple fractal and SOC analysis techniques.
Article
Environmental Sciences
Yara S. Tadano, Sanja Potgieter-Vermaak, Yslene R. Kachba, Daiane M. G. Chiroli, Luciana Casacio, Jessica C. Santos-Silva, Camila A. B. Moreira, Vivian Machado, Thiago Antonini Alves, Hugo Siqueira, Ricardo H. M. Godoi
Summary: Studies have shown that global lockdowns due to the COVID-19 outbreak have led to significant reductions in air pollutant levels worldwide. Different lockdown levels are directly related to new COVID-19 cases, air pollution, and economic restrictions. By using Artificial Neural Networks to predict air pollution levels, effective control and prediction can be achieved under flexible lockdown measures.
ENVIRONMENTAL POLLUTION
(2021)
Article
Engineering, Chemical
Gerson Uriel Colorado Cifuentes, Antonio Flores Tlacuahuac
Summary: A deep neural network model was developed for short-term prediction of ozone and particulate matter concentrations in a major northwestern metropolitan area in Mexico. The model, based on data from the local air quality monitoring system, accurately predicts the concentrations of pollutants and discusses the training process and performance metrics.
CANADIAN JOURNAL OF CHEMICAL ENGINEERING
(2021)
Article
Computer Science, Interdisciplinary Applications
Jordan Limperis, Weitian Tong, Felix Hamza-Lup, Lixin Li
Summary: This paper presents a new approach named TPPM25, based on the state-of-the-art Transformer neural network and various data embedding techniques, for forecasting PM2.5. Experimental results demonstrate the effectiveness of the TPPM25 model, outperforming a cutting-edge ensemble deep learning model from Zhang et al. and maintaining high prediction accuracy over longer periods of time.
EARTH SCIENCE INFORMATICS
(2023)
Article
Green & Sustainable Science & Technology
Nur 'atiah Zaini, Ali Najah Ahmed, Lee Woen Ean, Ming Fai Chow, Marlinda Abdul Malek
Summary: Accurate air pollution forecasting is crucial for urban planning and health risk management. This study utilizes hybrid deep learning models to forecast PM2.5 concentration in Kuala Lumpur, Malaysia, and improves prediction accuracy through analyzing neighboring station data and optimizing model parameters.
JOURNAL OF CLEANER PRODUCTION
(2023)
Article
Computer Science, Artificial Intelligence
Baowei Wang, Weiwen Kong, Peng Zhao
Summary: This study presents a more accurate method for predicting PM2.5 mass concentration by using Convnet and Dense-based Bidirectional Gated Recurrent Unit, which combines Convnet, Dense, and Bi-GRU to extract air quality data features and provide more accurate results.
Article
Computer Science, Information Systems
Ola Surakhi, Martha A. Zaidan, Pak Lun Fung, Naser Hossein Motlagh, Sami Serhan, Mohammad AlKhanafseh, Rania M. Ghoniem, Tareq Hussein
Summary: This paper investigates the impact of selecting an appropriate time-lag value on forecasting accuracy in time-series forecasting. The results show that the proposed LSTM model with heuristic algorithm is the best method for determining the optimal time-lag value.
Article
Environmental Sciences
Vahid Nourani, Hossein Karimzadeh, Aida Hosseini Baghanam
Summary: This study developed an efficient model for predicting CO pollutant concentrations using artificial neural network (ANN) and adaptive neural-fuzzy inference system (ANFIS), demonstrating the importance of air quality monitoring and developing effective models for sustainable development goals.
ENVIRONMENTAL EARTH SCIENCES
(2021)
Article
Environmental Sciences
Davood Namdar Khojasteh, Gholamreza Goudarzi, Ruhollah Taghizadeh-Mehrjardi, Akwasi Bonsu Asumadu-Sakyi, Masoud Fehresti-Sani
Summary: The research aimed to explore the long-term effects of air pollution on respiratory morbidity and mortality, and to develop accurate prediction models. The study found that nitrogen monoxide and carbon monoxide had significant effects on respiratory mortality, while other pollutants (NO2, SO2, O-3, PM10) had no significant impact on respiratory morbidity and mortality.
ATMOSPHERIC POLLUTION RESEARCH
(2021)
Article
Computer Science, Artificial Intelligence
Da-Chun Wu, Babak Bahrami Asl, Ali Razban, Jie Chen
Summary: The study used artificial neural networks to predict the electrical load of air compressors and validated two models under different control mechanisms. The results showed that both networks worked well for compressors using variable speed drive, while only the long short-term memory model provided acceptable results for compressors using on/off control. However, the models yielded unsatisfactory results for load/unload type air compressors.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Agriculture, Multidisciplinary
V. K. Chasiotis, D. A. Tzempelikos, A. E. Filios, K. P. Moustris
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2020)
Article
Green & Sustainable Science & Technology
K. Moustris, K. A. Kavadias, D. Zafirakis, J. K. Kaldellis
Article
Meteorology & Atmospheric Sciences
Dimitrios Nikolopoulos, Konstantinos Moustris, Ermioni Petraki, Demetrios Cantzos
Summary: This study investigates the chaos in the concentration dynamics of PM10 in the greater Athens area using DFA and R/S analysis. Findings reveal the presence of critical fractal behavior and long memory patterns in several segments of the PM10 time series, with 12 segments identified by both techniques. The importance of the agreement between the two chaos-analysis techniques and proper threshold selection is discussed.
METEOROLOGY AND ATMOSPHERIC PHYSICS
(2021)
Article
Biophysics
Stelios Maniatis, Panagiotis T. Nastos, Kostas Moustris, Iliana D. Polychroni, Athanasios Kamoutsis
INTERNATIONAL JOURNAL OF BIOMETEOROLOGY
(2020)
Article
Green & Sustainable Science & Technology
Michael Darmanis, Murat Cakan, Konstantinos P. Moustris, Kosmas A. Kavadias, Konstantinos-Stefanos P. Nikas
Article
Physics, Multidisciplinary
Dimitrios Nikolopoulos, Aftab Alam, Ermioni Petraki, Michail Papoutsidakis, Panayiotis Yannakopoulos, Konstantinos P. Moustris
Summary: This paper investigates a 17-year PM10 time series from five stations in Athens, Greece using statistical and entropy methods to analyze existing stochastic and self-organisation trends. It finds decreasing trends at all stations and explores self-organization through Boltzmann and Tsallis entropy in selected parts. The study identifies non-stochastic areas with fractal, long-memory, and self-organisation patterns, combining multiple fractal and SOC analysis techniques.
Article
Environmental Sciences
Georgios C. Spyropoulos, Panagiotis T. Nastos, Konstantinos P. Moustris
Summary: A significant portion of European cities' population is exposed to harmful levels of air pollution, leading to the installation of low-cost electrochemical sensor monitoring systems. The market is developing new air quality monitoring systems to provide forecasting services based on advanced technologies, protocols, and characteristics. This study compares data quality and Air Quality Index between low-cost sensors and fixed monitoring stations, revealing the need for flexible and affordable alternatives for monitoring low-cost gas sensors.
Article
Environmental Sciences
Christos Tsitsis, Dimitrios E. E. Alexakis, Konstantinos Moustris, Dimitra E. E. Gamvroula
Summary: The research aimed to evaluate the surface water system of Lake Vegoritida in Greece using the DPSIR methodological approach. Data analysis from 1983 to 1997 showed an increase in nutrient concentration contributing to eutrophic conditions. Artificial neural network (ANN) was used for predictions based on water quality data. The findings emphasized the need for sustainable land management, reduction in point sources of pollution, and agrochemicals in the study area to improve the water quality status and overall ecological assessment.
Article
Environmental Sciences
Dimitrios Nikolopoulos, Aftab Alam, Ermioni Petraki, Panayiotis Yannakopoulos, Konstantinos Moustris
Summary: This paper explores the multifractal characteristics of PM10 time series in the Greater Athens Area, Greece. The study utilizes a novel method called multifractal detrended fluctuation analysis (MFDFA) to analyze raw and shuffled series in 74 segments. The results show multifractality in all segments, with generalized and classical Hurst exponents falling within specific ranges. A new parameter, FWHM / f(max), is used to identify important multifractal behavior. The findings provide strong evidence of the multifractality of the PM10 time series.
Article
Energy & Fuels
Konstantinos Moustris, Dimitrios Zafirakis
Summary: Grid operators of islands with limited system tolerance often face the challenge of curtailing wind energy to maintain system stability and security of supply. Wind park owners, on the other hand, struggle with rejected wind energy production due to lack of storage facilities and flexibility options. This study focuses on predicting day-ahead wind energy production in island grids using artificial neural networks, aiming to provide accurate forecasts for wind park owners and explore alternative actor schemes in similar systems.
Article
Environmental Sciences
Kleopatra Ntourou, Konstantinos Moustris, Georgios Spyropoulos, Kyriaki-Maria Fameli, Nikolaos Manousakis
Summary: This paper aims to provide quantitative and qualitative data on the impact of long-term air pollution on the health of residents in the Greater Athens Area. The AirQ+ model is used to estimate the prevalence of bronchitis in children and the incidence of chronic bronchitis in adults due to particulate matter exposure. The results show a significant correlation between PM10 concentrations and adverse health effects, with a higher prevalence of bronchitis in children compared to adults. The unhealthiest areas were found to be the center of Athens city and suburban areas. There was a decreasing trend in both PM10 concentrations and the prevalence of chronic bronchitis over the 20-year period, particularly from 2010 to 2020.
Article
Environmental Studies
Georgios C. Spyropoulos, Panagiotis T. Nastos, Konstantinos P. Moustris, Konstantinos J. Chalvatzis
Summary: This study provides a comprehensive review and analysis of the evolution of the Greek vehicle fleet over the past 30 years. It predicts a reduction in emissions by 2030 and reveals that Greece is making positive progress in reducing air pollution from the transportation sector, but further improvements are still needed.
Article
Environmental Sciences
Konstantinos P. Moustris, Ermioni Petraki, Kleopatra Ntourou, Georgios Priniotakis, Dimitrios Nikolopoulos
Proceedings Paper
Green & Sustainable Science & Technology
Eleni Papazoglou, Konstantinos P. Moustris, Konstantinos-Stefanos P. Nikas, Panagiotis T. Nastos, John C. Statharas
TECHNOLOGIES AND MATERIALS FOR RENEWABLE ENERGY, ENVIRONMENT AND SUSTAINABILITY (TMREES)
(2019)
Article
Environmental Sciences
Dimitrios Nikolopoulos, Konstantinos Moustris, Ermioni Petraki, Dionysios Koulougliotis, Demetrios Cantzos