近年來環保意識的抬頭,對於工廠的設置必須有著更進一步周詳的計畫與環評,期使獲得當地居民的認同,緊急排放處理設備及安全閥設計為保護石化廠安全的最重要防線,因排放設計不當引發之工業災害屢見不鮮。緊急排放之潛在危害實不容忽視。國外近年來若干因緊急排放所引起之災害已漸漸引起國內廠商 之關注,因此落實技術,將對於工廠在排放安全之控制上有其重要性。 本論文以電腦流體動力計算方法,建立一套模擬緊急排放時大氣環境氣體流動變化的流程,可以對於不同緊急排放事件時利用同樣的流程來加以分析。研究過程中以實驗設計法設計整個實驗,再以類神經網路分析分析結果找出緊急排放時可能遭遇情形,以避免工安危害事件發生。
Due to the headline of environmental protection’s consciousness in recent years, the establishment of the factory must have been with careful and advanced plan, yet identify themselves with local inhabitant. Emergency blow-down’s equipment and safety relieving valve are designed to be the most important line of defense in order to protect factory's safety of petrochemical industry. It is common occurrence due to improper relieving design of industrial calamity. It can't be ignored that the potential harmer would be occurred for emergency blow-down. It had already concerning by domestic manufacturer gradually that several abroad blow-down calamities caused in recent years. Therefore, it is very important to the blow-down technology of control and safety in plants. In this study, computational fluid dynamics (CFD) accompanied by mass transport and fluid dynamic models were applied to simulate the emergency blow-down at various environmental parameters. These simulated results were analyzed by the Analysis of Variance (ANOVA) technique to evaluate the weighting and the correlations of the variables studied. Subsequently, the artificial neural network (ANN) method was employed to search for predicting dangerous situation from variables-response correlations using the learning rule and training algorithms