Prediction of pm10 using multiple linear regression and nonlinear regression Model in industrial areas
Loading...
Date
2019-07
Authors
Mohamad Eizlan Haqimi Bin Mat Zain
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Air pollution is a major problem that occurs in Malaysia. Particulate Matter is one of
the pollutants that contributed to air pollution that can cause adverse effect to living
organisms. Particulate Matter (PM) is the general term used for a mixture of solid particles
and liquid droplets found in the air. PM10 refer to particles of aerodynamic diameter less than
10 µm. The objective is to predict particulate matter concentration (PM10) by using multiple
linear regression and nonlinear regression in industrial areas. There are three industrial
monitoring stations that are Pasir Gudang, Nilai and Perai and one background station which
is Jerantut. The parameters were divided into two which are meteorological parameters and
gaseous parameters. Meteorological parameters consist of temperature (ºC), wind speed
(m/s) and relative humidity (%) while for gaseous parameters are Ozone (O3), Carbon
Monoxide (CO), Sulphur Dioxide (SO2) and Nitrogen Dioxide (NO2). The daily mean data
was used and divided into training data (70%) and validation data (30%) from 2015 until
2017. Highest maximum value of PM10 concentration was recorded at Perai in 2015 377.56
µg/m3) due to massive land and forest fire in Sumatra and Kalimantan, Indonesia. The result
shows that Nonlinear Regression is the best model to predict the PM10 concentration for next
day compared to Multiple Linear Regression.