Air Pollution Index Estimation Model Based On Artificial Neural Network
Loading...
Date
2021-06-01
Authors
Mohammed Nasser, Al-Subaie
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
Environmental conservation efforts are always dealing with a complex problem
because it involves a large number of variables. However, choosing a correct model
structure, and optimum training algorithm with minimum complexity is crucial.
Therefore, a dimensional reduction method was implemented based on the multiway
principal component analysis (MPCA) method. Three models were built in first part;
ozone estimation model, particulate matter 10 (PM10) estimation model, and air pollution
index (API) estimation model. Six inputs were used in ozone and PM10 models, which
are nitrogen oxides( NOx), carbon monoxide (CO), sulphur dioxides (SO2), wind speed,
air temperature, and relative humidity. After that, ozone and PM 10 were used as input to
the API estimation model. The result shows that the implementation of the MPCA has
insignificant improvement on the overall correlation factor due to the high nonlinearity
of data.