RT journal article
T1 Artificial Neural Networks, Sequence-to-Sequence LSTMs, and Exogenous Variables as Analytical Tools for NO2 (Air Pollution) Forecasting: A Case Study in the Bay of Algeciras (Spain)
A1 González Enrique, Francisco Javier
A1 Ruiz Águilar, Juan Jesús
A1 Moscoso López, José Antonio
A1 Urda, Daniel
A1 Deka, Lipika
A1 Turias Domínguez, Ignacio José
A2 Ingeniería Industrial e Ingeniería Civil
A2 Ingeniería Informática
K1 forecasting
K1 feature selection
K1 air pollution
K1 nitrogen dioxide
K1 artificial neural networks
K1 LSTMs
K1 exogenous variables
K1 deep learning
K1 time series
AB This study aims to produce accurate predictions of the NO2 concentrations at a specific station of a monitoring network located in the Bay of Algeciras (Spain). Artificial neural networks (ANNs) and sequence-to-sequence long short-term memory networks (LSTMs) were used to create the forecasting models. Additionally, a new prediction method was proposed combining LSTMs using a rolling window scheme with a cross-validation procedure for time series (LSTM-CVT). Two different strategies were followed regarding the input variables: using NO2 from the station or employing NO2 and other pollutants data from any station of the network plus meteorological variables. The ANN and LSTM-CVT exogenous models used lagged datasets of different window sizes. Several feature ranking methods were used to select the top lagged variables and include them in the final exogenous datasets. Prediction horizons of t + 1, t + 4 and t + 8 were employed. The exogenous variables inclusion enhanced the model's performance, especially for t + 4 (rho approximate to 0.68 to rho approximate to 0.74) and t + 8 (rho approximate to 0.59 to rho approximate to 0.66). The proposed LSTM-CVT method delivered promising results as the best performing models per prediction horizon employed this new methodology. Additionally, per each parameter combination, it obtained lower error values than ANNs in 85% of the cases.
PB MDPI
SN 1424-8220
YR 2021
FD 2021-03
LK http://hdl.handle.net/10498/24775
UL http://hdl.handle.net/10498/24775
LA eng
DS Repositorio Institucional de la Universidad de Cádiz
RD 07-nov-2026