Development of an artificial neural network model for ambient air pollution prediction | |
| Author | Nonpawit Choksirimongkolchai |
| Call Number | AIT Thesis no.EV-21-12 |
| Subject(s) | Particles--Environmental aspects--Thailand--Bangkok Ozone--Environmental aspects--Thailand--Bangkok Air--Pollution--Thailand--Bangkok Neural networks (Computer science) |
| Note | A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering in Environmental Engineering and Management |
| Publisher | Asian Institute of Technology |
| Abstract | Bangkok faces fine particulate matter problem (PM2.5) over the years, and ozone is a silent threat that has a greater impact due to higher temperature and increased precursor pollutants. Although several models were used to predict pollutant concentration, and the results were used as the waning system in Bangkok, these models were complicated and needed intensive resources. In this study, the aim was to use the artificial neural network (ANN) to predict hourly PM2.5 and ozone concentration in Bangkok. Pollutant concentration and meteorological data were used to train and validate the ANN in each high pollutant months during a three-year period, 2017 – 2019. The inputs of the ANN consisted of pollutants concentration (PM2.5 and ozone) and meteorological parameters (wind speed, wind direction, temperature, relative humidity and pressure). The ANN include six nodes of input layer, two hidden layers (six nodes for the first hidden layer, three nodes for the second hidden layer) and one node of output layer. The data was split into two sets, the training and validating datasets as 80:20. The results of the ANN model showed that the predictions of the next hour pollutant concentration were good for both the training and validating datasets. For the performance of validation, the range performance of the coefficient of determination (R2 ) and the Root Mean Square Error (RMSE) were 0.85 ~ 0.95 and 6.8 ~ 14.4, respectively. Finally, the models were used to predict the next 8-hr Ozone and 24-hr PM2.5 concentration, and then, used for AQI calculation, based on the criteria in Thailand. The spatial distribution of AQI was developed for a warning system of high pollution level. |
| Year | 2021 |
| Type | Thesis |
| School | School of Environment, Resources, and Development (SERD) |
| Department | Department of Energy and Climate Change (Former title: Department of Energy, Environment, and Climate Change (DEECC)) |
| Academic Program/FoS | Environmental Engineering and Management (EV) |
| Chairperson(s) | Ekbordin Winijkul; |
| Examination Committee(s) | Nguyen Thi Kim Oanh;Chutiporn Anutariya; |
| Scholarship Donor(s) | Royal Thai Government Fellowship; |
| Degree | Thesis (M. Eng.) - Asian Institute of Technology, 2021 |