Theses & Dissertations

Permanent URI for this collection

https://hdl.handle.net/20.500.14170/2002

Browse

Browsing Theses & Dissertations by Subject "Air quality indexes"
  • Publication
    Air quality assessment in Perlis
    ( 2017)
    Nurul Faizliyana Kamil
    The aim of this research is to determine the factor involved in air pollution in Perlis and to define the Multiple Linear Regression model for ozone concentration based on the variables involved in air pollution and also determine the characteristics of the variables. The observable data refer to the recorded hourly air quality data for 2003 until 2014 for Perlis that have been received from Air Quality Division of the Department of Environment, Ministry of Natural Resources and Environment of Malaysia. Data refer to hourly recorded data that involved five major pollutants and four meteorological variables. The variables are O3, CO, NO2, SO2, PM10, temperature, relative humidity, wind speed and wind direction. Since it involve with large data set, factor analysis is applied. Factor analysis classify the 9 variables into three factors and name the factor as human activities factor, wind factor and atmospheric factor. Based on the analysis, concentration of O3 show significant differences during daytime and night time. So that the data are separated into daytime and night time and then be applied with the multiple linear regression in order to model O3 concentration based on the air pollutants (CO, NO2, PM10, SO2) and meteorological variables(temperature, humidity, wind speed and wind direction). R2 values showed that 45.6% variation of O3 formation is explained by 7 independent variables during day time and 34.5% variation of O3 formation is explained by 7 independent variables during night time. The Multiple Linear Regression model was tested for goodness of fit using Mean Absolute Error Value and Normalized Absolute Error. The test showed that MLR model is best fit model for O3 formation during daytime compared to night time.
      3  14