Application of Bayesian Network to Predict Ozone Concentration – A Case Study at Dali Air Quality Station in Taiwan

• Main Authors: Tzu-Yin Chen, 陳姿吟
• Other Authors: Ho-Wen Chen
• Format: Others
• Language: zh-TW
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/wmpp24

碩士 === 東海大學 === 環境科學與工程學系 === 103 === In recent years, urban air quality deteriorates gradually. Air pollution affects human health significantly, and air quality is getting more attention among our societies. Long-term monitoring data from the past has showed that ozone is a major air pollutant in urban area. Alert model of ozone will be establish in order to let the public to know about ozone concentration in advance, and it adopts appropriate policies to reduce the negative impact of ozone on human health. However, due to randomness and complexness of ozone (O3) pollution’s characteristics, the Bayesian theory, Monte Carlo stimulation and Markov Chain is used to create ozone forecasting model in our study. So, Dali monitoring station is chosen as object of this study to verify the feasibility of our model, in which the selection of hourly ozone monitoring data for five years between 2006 and 2011 were integrated into Bayesian theory and Monte Carlo stimulation to establish the time series forecasting model of ozone concentrations and stimulate the trend of changes of urban ozone concentration in the air. The result is shown that the probabilistic model can predict the ozone concentration’s changes trend effectively. Thus, with the help of this model, people or government will be able to take preventive measures to reduce harm to human health.