Short-Term Prediction Models Of Pm10 Concentrations In Peninsular Malaysia Using Multivariate Time Series And Machine Learning Methods
| dc.contributor.author | Ramli, Norazrin | |
| dc.date.accessioned | 2022-04-06T08:04:33Z | |
| dc.date.available | 2022-04-06T08:04:33Z | |
| dc.date.issued | 2021-05 | |
| dc.description.abstract | The particulate matter with an aerodynamic diameter less than 10 μm (PM10) is identified as one of the dangerous air pollutants to human health and the concentrations of PM10 in Asian and Pacific cities remain as the most problematic local air pollution issues. The objectives of the research are to determine the characteristics and trend of PM10 concentrations in Malaysia from 1999 to 2015, to propose a Multivariate Time Series (MTS) analysis using Vector Autoregressive (VAR) to predict the short-term PM10 concentrations and interpret the relationship between PM10 concentrations and meteorological parameters using the graphical view of causality. Three models for short-term prediction of PM10 using Multiple Linear Regression (MLR), Bayesian Model Averaging (BMA) and Boosted Regression Tree (BRT) model. The performance indicators (R2, IA, MAE, RMSE, and MAPE) are applied to obtain the best model. A study using seventeen years of air quality monitoring data from the Department of Environment Malaysia (DOE) was used with eight parameters (PM10, NO2, SO2, CO, O3, wind speed, temperature, and relative humidity) and nine monitoring stations were selected which included Kangar, Perai, Shah Alam, Nilai, Larkin, Pasir Gudang, Kertih, Kota Bharu and Jerantut to represent the Northern, Central, Southern and East of Peninsular Malaysia. The trend analysis used the Mann-Kendall test for trend detection and Sen’s slope estimator for trend estimation using monthly average and maximum monthly of PM10 concentrations. The monthly average of air monitoring data was also used for the VAR model, which includes four parameters: PM10, temperature, relative humidity, and wind speed. The MLR, BMA and BRT used daily average of air monitoring data for eight parameters (PM10, NO2, SO2, CO, O3, wind speed, temperature, and relative humidity). Eighty percent of the air monitoring data was used for training and twenty percent was used for validation of the models. The results showed that the daily mean PM10 concentrations at all the stations were in the range between 37.34 μg/m3 to 59.19 μg/m3 for all stations. The trend analysis showed that the monotonic trend was significant at Nilai, Larkin, Kota Bharu and Jerantut which indicated that the mean monthly PM10 concentrations at these stations had a monotonic trend where the PM10 concentrations consistently increased through time. The VAR (2) model is the most suitable model for predicting PM10 concentrations at Kangar and Perai monitoring station. The causality relationship shows that the significant causal relationship at most monitoring stations is between PM10 concentrations, wind speed and relative humidity. The BRT model gave good results to predict the PM10 concentrations for each station. Overall, the finding of the research provides an insight for a new prediction model using BRT especially in relation to the air quality studies and the results support the idea that this models can be adopted and applied in various fields such as air pollution control and monitoring studies. The model will be relevant for Department of Environment Malaysia to cooperatively develop and implement air quality forecasting and monitoring, assessment and early warning system to prevent, monitor and mitigate the air pollution in Malaysia. | en_US |
| dc.identifier.uri | http://hdl.handle.net/123456789/15061 | |
| dc.language.iso | en | en_US |
| dc.publisher | Universiti Sains Malaysia | en_US |
| dc.subject | Education | en_US |
| dc.title | Short-Term Prediction Models Of Pm10 Concentrations In Peninsular Malaysia Using Multivariate Time Series And Machine Learning Methods | en_US |
| dc.type | Thesis | en_US |
Files
License bundle
1 - 1 of 1
Loading...
- Name:
- license.txt
- Size:
- 1.71 KB
- Format:
- Item-specific license agreed upon to submission
- Description: