PM2.5 Concentration Forecasting using Neural Networks for Hotspots of Delhi
Autor: | Anubha Mandal, Maninder Kaur |
---|---|
Rok vydání: | 2020 |
Předmět: |
Pollution
Pollutant Nonlinear autoregressive exogenous model 010504 meteorology & atmospheric sciences Meteorology Computer science media_common.quotation_subject Air pollution 010501 environmental sciences Particulates medicine.disease_cause 01 natural sciences Urbanization medicine Early warning system Air quality index 0105 earth and related environmental sciences media_common |
Zdroj: | 2020 International Conference on Contemporary Computing and Applications (IC3A). |
Popis: | Delhi is ranked as the sixth most polluted city in India for air pollution. Industrialization and urbanization has exacerbated the air quality over recent years. Particularly, particulate matter (PM 2.5 ) concentration has found to be in exceedance than its prescribed limits throughout the year in hotspot regions. Such alarming pollution level has put public on verge of adverse health effects. It is, therefore necessary to establish more accurate air monitoring and early warning system to evaluate the forecasting of air pollutant concentration. Thus, in this paper an attempt has been made to forecast step ahead PM 2.5 concentration for hotspots of Delhi using various artificial neural networks. A comparative assessment has been established to evaluate the forecasting capabilities of these networks. The forecasting results indicate that the non-linear autoregressive network with exogenous input (NARX) is superior to other neural network based models on account of its accuracy. |
Databáze: | OpenAIRE |
Externí odkaz: |