نوع مقاله : ترویجی
نویسندگان
1 دکتری، پژوهنده، گروه فناوریهای تبدیل و بهینهسازی، پژوهشکده توسعه فناوریهای پالایش و فرآورش نفت، پژوهشگاه صنعت نفت، شهر تهران، استان تهران، صندوق پستی 137-14665، ایران
2 کارشناسی، مدیر گروه، گروه فناوریهای تبدیل و بهینهسازی، پژوهشکده توسعه فناوریهای پالایش و فرآورش نفت، پژوهشگاه صنعت نفت، شهر تهران، استان تهران، صندوق پستی 137-14665، ایران
چکیده
کلیدواژهها
عنوان مقاله [English]
نویسندگان [English]
Crude oil associated water contains various types of salts causing problems such as corrosion, fouling and plugging of exchangers, furnace pipes and bottom trays of distillation columns in refinery processes. In this study, the performance of refinery desalters is evaluated by calculating the salinity and water cut efficiencies using artificial neural network (ANN) technique. ANN is selected due to its potential for modeling of highly nonlinear phenomena involving in the desalting process. In this study, the performance of the desalting/dehydration process is evaluated by calculating the salinity and water cut efficiencies using data based methods. Five variables namely fresh water flow rate, temperature of fresh water fed to first desalter, oil flow rate and water temperature and flow rate to the second desalter, are introduced to the ANN model as input parameters. The simulation results are compared to the experimental data extracted from Bandar Abbas refineries. The overall agreement between the ANN predictions and experimental data for water cut are acceptable with average error about 0.4% for training and 1.9% for test. Sensitivity analysis has revealed different operating parameters affecting various desalting units. The model is capable of predicting the behavior of desalting process quite well if we access to enough and suitable experimental data.
کلیدواژهها [English]