Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model
碩士 === 東海大學 === 高階經營管理碩士在職專班 === 107 === Taiwan's air quality has been an important issue in recent years, and the most important factor affecting air quality is fine particulate matters(PM2.5). Air pollution, including aerosols(PM10), sulfur dioxide(SO2), nitrogen dioxide (NO2) and carbon mono...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2019
|
Online Access: | http://ndltd.ncl.edu.tw/handle/r8899e |
id |
ndltd-TW-107THU01026049 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-107THU010260492019-10-24T05:19:58Z http://ndltd.ncl.edu.tw/handle/r8899e Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model 馬祖地區空氣污染監測與減量模擬 ─ARIMAX模型之應用 LIN, CHUN-TENG 林浚騰 碩士 東海大學 高階經營管理碩士在職專班 107 Taiwan's air quality has been an important issue in recent years, and the most important factor affecting air quality is fine particulate matters(PM2.5). Air pollution, including aerosols(PM10), sulfur dioxide(SO2), nitrogen dioxide (NO2) and carbon monoxide (CO)and Secondary Aerosol. In addition, meteorological factors such as wind speed, temperature and relative humidity are related to the level of primary aerosol, which will also affect the concentration of PM2.5. The purpose of this paper is to estimate and predict the PM2.5 concentration at the Mazu monitoring station. There are many missing values when using the EPA air quality hourly monitoring data. First, use the Back Propagation Neural Network to fill in the missing values; Secondly, the Autoregressive Integrated Moving Average with Explanatory Variable Model(ARIMAX)is used to estimate the regression parameters and the out-of-sample prediction. Finally, the pollution reduction simulation was carried out with an average reduction target of 18 ug/m3. The empirical results show that: (1) the predicted performance of the ARIMAX model, Better than the Autoregressive Integrated Moving Average Model (ARIMA). (2) PM2.5, PM10, NO2, and CO in the first phase will deteriorate PM2.5. (3) During the northeast monsoon period from October to December and March of each year, the average value of PM2.5 reached 23.824 ug/m3; it was significantly higher than the average of other months by 17.839 ug/m3, and there was an additional 2.098 ug/m3 during the day. If the sample average is 20.823 ug/m3, it is found that the foreign pollution in Mazu area is between 20% and 29%. (4) According to the estimation results, we find, on average, the reduction of PM2.5 from 20.823 to 18 per hour mainly depends on the improvement of primary aerosol (about 98%), and followed by CO, NO2 and SO2. LIN, JWU-RONG 林灼榮 2019 學位論文 ; thesis 40 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 東海大學 === 高階經營管理碩士在職專班 === 107 === Taiwan's air quality has been an important issue in recent years, and the most important factor affecting air quality is fine particulate matters(PM2.5). Air pollution, including aerosols(PM10), sulfur dioxide(SO2), nitrogen dioxide (NO2) and carbon monoxide (CO)and Secondary Aerosol. In addition, meteorological factors such as wind speed, temperature and relative humidity are related to the level of primary aerosol, which will also affect the concentration of PM2.5. The purpose of this paper is to estimate and predict the PM2.5 concentration at the Mazu monitoring station. There are many missing values when using the EPA air quality hourly monitoring data. First, use the Back Propagation Neural Network to fill in the missing values; Secondly, the Autoregressive Integrated Moving Average with Explanatory Variable Model(ARIMAX)is used to estimate the regression parameters and the out-of-sample prediction. Finally, the pollution reduction simulation was carried out with an average reduction target of 18 ug/m3. The empirical results show that: (1) the predicted performance of the ARIMAX model, Better than the Autoregressive Integrated Moving Average Model (ARIMA). (2) PM2.5, PM10, NO2, and CO in the first phase will deteriorate PM2.5. (3) During the northeast monsoon period from October to December and March of each year, the average value of PM2.5 reached 23.824 ug/m3; it was significantly higher than the average of other months by 17.839 ug/m3, and there was an additional 2.098 ug/m3 during the day. If the sample average is 20.823 ug/m3, it is found that the foreign pollution in Mazu area is between 20% and 29%. (4) According to the estimation results, we find, on average, the reduction of PM2.5 from 20.823 to 18 per hour mainly depends on the improvement of primary aerosol (about 98%), and followed by CO, NO2 and SO2.
|
author2 |
LIN, JWU-RONG |
author_facet |
LIN, JWU-RONG LIN, CHUN-TENG 林浚騰 |
author |
LIN, CHUN-TENG 林浚騰 |
spellingShingle |
LIN, CHUN-TENG 林浚騰 Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model |
author_sort |
LIN, CHUN-TENG |
title |
Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model |
title_short |
Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model |
title_full |
Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model |
title_fullStr |
Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model |
title_full_unstemmed |
Air Polliution Monitoring and Abatement Simulation in Mazu Area - An Application of ARIMAX Model |
title_sort |
air polliution monitoring and abatement simulation in mazu area - an application of arimax model |
publishDate |
2019 |
url |
http://ndltd.ncl.edu.tw/handle/r8899e |
work_keys_str_mv |
AT linchunteng airpolliutionmonitoringandabatementsimulationinmazuareaanapplicationofarimaxmodel AT línjùnténg airpolliutionmonitoringandabatementsimulationinmazuareaanapplicationofarimaxmodel AT linchunteng mǎzǔdeqūkōngqìwūrǎnjiāncèyǔjiǎnliàngmónǐarimaxmóxíngzhīyīngyòng AT línjùnténg mǎzǔdeqūkōngqìwūrǎnjiāncèyǔjiǎnliàngmónǐarimaxmóxíngzhīyīngyòng |
_version_ |
1719277189098110976 |