| 摘要: | 石化產業在生產過程中會產生許多污染物質，例如空氣懸浮微粒(Particulate Matter, PM)、硫氧化物(SOx)、氮氧化物(NOx)、揮發性有機物(VOCs)、碳氧化物(CO & CO2)等。尤其在石化工業工安事故發生時，更會在短時間內造成重大空氣污染危害。因此，本子計畫在總計畫「石化廠製程安全評估與風險控制之維護管理系統」架構下，與中油桃園煉油廠實際合作針對石化業產生之空氣污染進行分析與擴散模擬研究。本研究計畫整體包括六大研究步驟，包括：(1)石化產業與異質環境空品巨量資料融合與蒐集整理及文獻探討。(2)分析歷史桃園煉油廠鄰近空氣污染監測資料時空變化特徵。(3)從監測資料時空變化特徵探討石化廠對鄰近空氣品質的影響。(4)分析歷史桃園煉油廠工安事故事件所造成鄰近空氣污染監測資料時空變化，同時與空氣品質模式模擬結果比對。(5)利用空氣品質模式模擬歷史桃園煉油廠工安事故事件，並與實際監測資料進行比對。(6)模擬未來在石化廠工安事故造成氣體洩漏的情境下，其可能的氣體污染物成分分析與擴散情形。本研究透過巨量資料探勘分析，使得原本極為龐大的時空環境監測巨量資料維度縮減到最重要的數個時空特徵，並從中找出其重要參數與物理機制。透過此方法，我們不僅可以分析現有氣象局與環保署觀測資料並連結其他子計畫中的石化廠監測數據，探索影響石化廠空氣污染物排放的大量異質資料中，最具關鍵性的重要參數，探討石化產業對空氣品質的影響機制。同時透過空氣擴散模擬模式進行污染事件模擬與比對，模擬石化廠工安事故造成氣體洩漏的情境下，其可能的氣體污染物分析與擴散情形，並提供回饋給總計畫與其他子計畫進行石化產業風險管理參考依據。 ;The petrochemical industry produces many pollutants in the production process, such as Particulate Matters (PM), sulfur oxides (SOx), nitrogen oxides (NOx), volatile organic compounds (VOCs), and carbon oxides (CO & CO2) and so on. Especially in the petrochemical industry, when the accident occurs, it will cause major air pollution hazards in a short period of time. Therefore, under the framework of the integrated research project “Maintenance & Management System of Process Safety Assessment and Risk Control for Petroleum Plant Facilities”, the subproject plans to analyze the air pollution generated by the petrochemical industry with the actual cooperation of the Taoyuan Refinery, CPC Corporation. The research project includes six research steps, including: (1) Heterogeneous Environmental Big data Fusion and literature review in the petrochemical industry. (2) Analyze the temporal and spatial variation characteristics of the air pollution monitoring data of the historical Taoyuan refinery. (3) The influence of the petrochemical plant on the adjacent air quality will be discussed from the temporal and spatial characteristics of the monitoring data. (4) The temporal and spatial changes of the adjacent air pollution monitoring data caused by the accidents in the historical Taoyuan refinery will be analyzed, and compared with the air quality model simulation results. (5) The air quality simulation model will be used to simulate the accidents of the historical Taoyuan refinery and the actual monitoring data were compared. (6) Simulates the diffusion of possible gas pollutants in the future under the scenario of gas leakage caused by industrial safety accidents in petrochemical plants.Through the big data mining analysis, this study reduced the big data dimension of the extremely large space-time environment monitoring to the most important time-space features, and find out its important parameters and physical mechanisms. Through this method, we can not only analyze the existing CWB and EPA observation data and link the petrochemical plant monitoring data in other sub-projects, but also explore the most important and important parameters among the large amount of heterogeneous data affecting the petrochemical plant's air pollutant emissions. Furthermore, we investigate the impact of the petrochemical industry on air quality. At the same time, through the air diffusion and simulation model, the pollution event simulation and comparison are carried out to simulate the gas leakage of the petrochemical plant, and the possible gas pollutant analysis and diffusion situation, and provide feedback to the other sub-projects. Also, provide results to the petrochemical industry risk management for reference. |
