مقاله Experimental Study of Heavy Liquid Hydrocarbon Pyrolysis: Application of Neural Network to P
مقاله Experimental Study of Heavy Liquid Hydrocarbon Pyrolysis: Application of Neural Network to Predict the Main Product Yields فایل ورد (word) دارای 9 صفحه می باشد و دارای تنظیمات در microsoft word می باشد و آماده پرینت یا چاپ است
فایل ورد مقاله Experimental Study of Heavy Liquid Hydrocarbon Pyrolysis: Application of Neural Network to Predict the Main Product Yields فایل ورد (word) کاملا فرمت بندی و تنظیم شده در استاندارد دانشگاه و مراکز دولتی می باشد.
این پروژه توسط مرکز مرکز پروژه های دانشجویی آماده و تنظیم شده است
توجه : در صورت مشاهده بهم ریختگی احتمالی در متون زیر ،دلیل ان کپی کردن این مطالب از داخل فایل ورد می باشد و در فایل اصلی مقاله Experimental Study of Heavy Liquid Hydrocarbon Pyrolysis: Application of Neural Network to Predict the Main Product Yields فایل ورد (word) ،به هیچ وجه بهم ریختگی وجود ندارد
بخشی از متن مقاله Experimental Study of Heavy Liquid Hydrocarbon Pyrolysis: Application of Neural Network to Predict the Main Product Yields فایل ورد (word) :
سال انتشار: 1389
محل انتشار: چهاردهمین همایش بین المللی نفت، گاز و پتروشیمی
تعداد صفحات: 9
چکیده:
products distribution of thermal cracking of heavy liquid hydrocarbon feedstock were investigated in a tubular reactor at atmospheric pressure. Central composite design CCD was used as an experimental design. The design variables were coli outlet temperature COT steam ratio and feed flow rate. maximum yields of ethylene were 30.37 wt.% at COT,steam ratio and residence time of 869 C , 1.22 gr/gr and 0.208 sec respectively. maximum yields of propylene was 15.37 wt.% at COT , steam ratio and residence time of 825 C , 0.95gr/gr and 0.147 sec respectively. maximum yield of olefin ethylene +propylene was obtained at 842.5 C, steam ratio of 1.24 gr/gr and residence time of 0.17 sec. in these conditions , the yields of ethylene and propylene are 27.2 and 13.6 weight percent respectively. finally , a three layer perceptron neural network , with back propagation BP training algorithm , was developed for modeling of thermal cracking of heay feedstock. the optimum structure of neural network was determined by a trial and error method and different structures were tried.
