Development of an hourly forecast application of PM2.5 concentration in Taiwan

• Main Authors: Jhong, Che-Hao, 鍾哲豪
• Other Authors: Wu, Ji-cheng
• Format: Others
• Language: zh-TW
• Published: 2016
• Online Access: http://ndltd.ncl.edu.tw/handle/55056075473160804245

碩士 === 國立屏東科技大學 === 工業管理系所 === 104 === The purpose of the thesis is to establish an hourly forecast system of PM2.5 concentrations in Taiwan by using and modifying SARIMA model. At first, we introduce a Longitudinal SARIMA model which uses to simulate PM¬2.5 concentration collected by an observing station in Pingtung. Shows that compare to the typical SARIMA model, using Longitudinal SARIMA model results that residuals tend to fulfil the condition of white noise. Therefore, by referring to this result, we developed a PM2.5 hourly forecast client App based on Android with a three day forward forecasting ability which could provide user to inquire the prediction of hourly PM2.5 concentrations in Taiwan areas by mobiles.