Evolving Spiking Neural Network Model for PM2.5 Hourly Concentration Prediction Based on Seasonal Differences: A Case Study on Data from Beijing and Shanghai

Hengyuan Liu; Guibin Lu; Yangjun Wang; Nikola Kasabov

doi:10.4209/aaqr.2020.05.0247

Evolving Spiking Neural Network Model for PM2.5 Hourly Concentration Prediction Based on Seasonal Differences: A Case Study on Data from Beijing and Shanghai

Hengyuan Liu¹, Guibin Lu This email address is being protected from spambots. You need JavaScript enabled to view it.¹, Yangjun Wang², Nikola Kasabov^3,4

+ Show author affiliations

¹School of Economics, Shanghai University, Shanghai 200444, China
²School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, China
³School of Engineering, Computing and Mathematical Sciences, Auckland University of Technology, Auckland, New Zealand
⁴Intelligent Systems Research Centre, Ulster University, Londonderry, UK

Received: May 19, 2020
Revised: August 11, 2020
Accepted: August 25, 2020

Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.

Download Citation: ||https://doi.org/10.4209/aaqr.2020.05.0247

Download: PDF

Cite this article:

Liu, h., Lu, G., Wang, Y. and Kasabov, N. (2020). Evolving Spiking Neural Network Model for PM_2.5 Hourly Concentration Prediction Based on Seasonal Differences: A Case Study on Data from Beijing and Shanghai. Aerosol Air Qual. Res. https://doi.org/10.4209/aaqr.2020.05.0247

HIGHLIGHTS

A Staging-eSNN model is proposed to predict PM_2.5 hourly concentration.
Seasonal difference in diurnal variation of PM_2.5 have been considered and evaluated.
The available data are processed to capture informative patterns by the Staging-eSNN.

ABSTRACT

In recent years, the harmfulness of air pollutants to human health and the environment have received widespread attention. Accurate prediction of air quality is essential to air quality management and policy development, however, the seasonal differences and the diurnal variations of air pollutants have not been fully investigated by traditional forecast models. Furthermore, the available spatio-temporal data for the problem have not been well processed to capture informative predictive patterns from past data. This paper proposes a Staging evolving spiking neural network model (Staging-eSNN), which uses a time series clustering algorithm to distinguish seasonal differences with diurnal variations of PM_2.5 concentration. Based on this pretreatment, we use the Staging-eSNN model to predict PM_2.5 concentration in the next 1, 3, 6, 12 and 24 hour in Beijing and Shanghai areas. Various evaluation indicators for prediction performance show that the Staging-eSNN model achieves appealing performance than SVR, RF and other eSNN models.

Keywords: Air pollutant prediction; PM2.5 hourly concentration; Seasonality; Evolving spiking neural networks; Time series clustering.

Aerosol Air Qual. Res. 20 :-. https://doi.org/10.4209/aaqr.2020.05.0247