سال انتشار: ۱۳۹۱

محل انتشار: کنفرانس بین المللی مدل سازی غیر خطی و بهینه سازی

تعداد صفحات: ۵

نویسنده(ها):

Majid Taghizadeh – Department of Chemical Engineering, Babol University of Technology, 4714871167 Babol, Iran
Mohammad Soleimani Lashkenari –

چکیده:

Viscosity is a parameter that plays a pivotal role in reservoir fluid estimations. In this study a robust artificial neural network (ANN) code is developed in MATLAB software topredict the viscosity of Irananian crude oils. The results obtained by the ANN are compared with the experimental data. The prediction procedure is carried out at three different regimes that are bubble point, bellow the bubble point, and above the bubble point using the PVT data of 57 bottom hole samples collected from Iranian Oil Fields. It is confirmed that the ANN yields superior results and has the low deviation from the experimental data