A review of artificial neural network models for ambient air pollution prediction

Research output: Contribution to journalReview article

3 Citations (Scopus)

Abstract

Research activity in the field of air pollution forecasting using artificial neural networks (ANNs) has increased dramatically in recent years. However, the development of ANN models entails levels of uncertainty given the black-box nature of ANNs. In this paper, a protocol by Maier et al. (2010) for ANN model development is presented and applied to assess journal papers dealing with air pollution forecasting using ANN models. The majority of the reviewed works are aimed at the long-term forecasting of outdoor PM10, PM2.5, and oxides of nitrogen, and ozone. The vast majority of the identified works utilised meteorological and source emissions predictors almost exclusively. Furthermore, ad-hoc approaches are found to be predominantly used for determining optimal model predictors, appropriate data subsets and the optimal model structure. Multilayer perceptron and ensemble-type models are predominantly implemented. Overall, the findings highlight the need for developing systematic protocols for developing powerful ANN models.

LanguageEnglish
Pages285-304
Number of pages20
JournalEnvironmental Modelling and Software
Volume119
Early online date30 Jun 2019
DOIs
Publication statusPublished - 1 Sep 2019

Fingerprint

Air pollution
artificial neural network
ambient air
atmospheric pollution
Neural networks
prediction
Multilayer neural networks
Model structures
Ozone
ozone
oxide
Nitrogen
Oxides
nitrogen

Keywords

  • air pollution
  • artificial neural networks
  • backpropagation algorithm
  • forecasting
  • multilayer perceptron

Cite this

@article{f5c3935664214bd2bd2b4c09c4f651b7,
title = "A review of artificial neural network models for ambient air pollution prediction",
abstract = "Research activity in the field of air pollution forecasting using artificial neural networks (ANNs) has increased dramatically in recent years. However, the development of ANN models entails levels of uncertainty given the black-box nature of ANNs. In this paper, a protocol by Maier et al. (2010) for ANN model development is presented and applied to assess journal papers dealing with air pollution forecasting using ANN models. The majority of the reviewed works are aimed at the long-term forecasting of outdoor PM10, PM2.5, and oxides of nitrogen, and ozone. The vast majority of the identified works utilised meteorological and source emissions predictors almost exclusively. Furthermore, ad-hoc approaches are found to be predominantly used for determining optimal model predictors, appropriate data subsets and the optimal model structure. Multilayer perceptron and ensemble-type models are predominantly implemented. Overall, the findings highlight the need for developing systematic protocols for developing powerful ANN models.",
keywords = "air pollution, artificial neural networks, backpropagation algorithm, forecasting, multilayer perceptron",
author = "Cabaneros, {Sheen Mclean} and Calautit, {John Kaiser} and Hughes, {Ben Richard}",
year = "2019",
month = "9",
day = "1",
doi = "10.1016/j.envsoft.2019.06.014",
language = "English",
volume = "119",
pages = "285--304",
journal = "Environmental Modelling and Software",
issn = "1364-8152",

}

A review of artificial neural network models for ambient air pollution prediction. / Cabaneros, Sheen Mclean; Calautit, John Kaiser; Hughes, Ben Richard.

In: Environmental Modelling and Software, Vol. 119, 01.09.2019, p. 285-304.

Research output: Contribution to journalReview article

TY - JOUR

T1 - A review of artificial neural network models for ambient air pollution prediction

AU - Cabaneros, Sheen Mclean

AU - Calautit, John Kaiser

AU - Hughes, Ben Richard

PY - 2019/9/1

Y1 - 2019/9/1

N2 - Research activity in the field of air pollution forecasting using artificial neural networks (ANNs) has increased dramatically in recent years. However, the development of ANN models entails levels of uncertainty given the black-box nature of ANNs. In this paper, a protocol by Maier et al. (2010) for ANN model development is presented and applied to assess journal papers dealing with air pollution forecasting using ANN models. The majority of the reviewed works are aimed at the long-term forecasting of outdoor PM10, PM2.5, and oxides of nitrogen, and ozone. The vast majority of the identified works utilised meteorological and source emissions predictors almost exclusively. Furthermore, ad-hoc approaches are found to be predominantly used for determining optimal model predictors, appropriate data subsets and the optimal model structure. Multilayer perceptron and ensemble-type models are predominantly implemented. Overall, the findings highlight the need for developing systematic protocols for developing powerful ANN models.

AB - Research activity in the field of air pollution forecasting using artificial neural networks (ANNs) has increased dramatically in recent years. However, the development of ANN models entails levels of uncertainty given the black-box nature of ANNs. In this paper, a protocol by Maier et al. (2010) for ANN model development is presented and applied to assess journal papers dealing with air pollution forecasting using ANN models. The majority of the reviewed works are aimed at the long-term forecasting of outdoor PM10, PM2.5, and oxides of nitrogen, and ozone. The vast majority of the identified works utilised meteorological and source emissions predictors almost exclusively. Furthermore, ad-hoc approaches are found to be predominantly used for determining optimal model predictors, appropriate data subsets and the optimal model structure. Multilayer perceptron and ensemble-type models are predominantly implemented. Overall, the findings highlight the need for developing systematic protocols for developing powerful ANN models.

KW - air pollution

KW - artificial neural networks

KW - backpropagation algorithm

KW - forecasting

KW - multilayer perceptron

UR - https://www.sciencedirect.com/journal/environmental-modelling-and-software

U2 - 10.1016/j.envsoft.2019.06.014

DO - 10.1016/j.envsoft.2019.06.014

M3 - Review article

VL - 119

SP - 285

EP - 304

JO - Environmental Modelling and Software

T2 - Environmental Modelling and Software

JF - Environmental Modelling and Software

SN - 1364-8152

ER -