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.

Modelling of Urban Air Pollutant Concentrations with Artificial Neural Networks Using Novel Input Variables

“…Generally, ANN models can often represent environmental relationships with surprising accuracy, although they are not fully understood by traditional theory. Ultimately, all ANNs are based on the intrinsic principle of the "black-box" [24,33]. Hence, it is not possible to trace and adjust individual steps within the model calculation.…”

“…Hence, it is not possible to trace and adjust individual steps within the model calculation. Connections and weighting adjustments are not visible for the modeler [33].…”

“…First, a strict QA/QC procedure was applied using R Studio [32]. All data were normalised to the range from 0 to 1 (Equation (1)) [18,20,33].…”

Development of mathematical models of gas leakage and its propagation in atmospheric air at an emergency gas well gushing

“…Study [15] provides an algorithm for calculating parameters of releases and gross releases of harmful substances from flaring facilities where hydrocarbon mixtures are burnt but it does not enable determination of the amount of releases in non-burning gushing.…”

Identify the contribution of elevated industrial plume to ground air quality by optical and machine learning methods

“…When the bottom envelope of HISP is close to the ground, chances are that the plume's particles increase ground PM 2.5 concentration since they mainly distribute in fine size [23]. On top of optical methods, more practitioners resort to data-driven methods, such as machine learning, which can predict air pollution and identify the key factors without consider complex physical and chemical processes [24][25][26]. Note that optical and data-driven methods are generally case-specific, produce limited repeatability, which is the irreplaceable advantage of classical physical models [25].…”