PM10 concentration forecasting in the metropolitan area of Oviedo (Northern Spain) using models based on SVM, MLP, VARMA and ARIMA: A case study

P J García Nieto; F Sánchez Lasheras; E García-Gonzalo; F J de Cos Juez

doi:10.1016/j.scitotenv.2017.11.291

PM₁₀ concentration forecasting in the metropolitan area of Oviedo (Northern Spain) using models based on SVM, MLP, VARMA and ARIMA: A case study

Sci Total Environ. 2018 Apr 15:621:753-761. doi: 10.1016/j.scitotenv.2017.11.291. Epub 2017 Dec 1.

Authors

P J García Nieto¹, F Sánchez Lasheras², E García-Gonzalo², F J de Cos Juez³

Affiliations

¹ Department of Mathematics, Faculty of Sciences, University of Oviedo, 33007 Oviedo, Spain. Electronic address: lato@orion.ciencias.uniovi.es.
² Department of Mathematics, Faculty of Sciences, University of Oviedo, 33007 Oviedo, Spain.
³ Exploitation and Prospecting Department, University of Oviedo, 33004 Oviedo, Spain.

PMID: 29202286
DOI: 10.1016/j.scitotenv.2017.11.291

Abstract

Atmospheric particulate matter (PM) is one of the pollutants that may have a significant impact on human health. Data collected over seven years in a city of the north of Spain is analyzed using four different mathematical models: vector autoregressive moving-average (VARMA), autoregressive integrated moving-average (ARIMA), multilayer perceptron (MLP) neural networks and support vector machines (SVMs) with regression. Measured monthly average pollutants and PM₁₀ (particles with a diameter less than 10μm) concentration are used as input to forecast the monthly averaged concentration of PM₁₀ from one to seven months ahead. Simulations showed that the SVM model performs better than the other models when forecasting one month ahead and also for the following seven months.

Keywords: Autoregressive integrated moving-average (ARIMA); Multilayer perceptron (MLP); Particulate matter (PM(10)) forecasting; Support vector regression (SVR); Vector autoregressive moving-average (VARMA).