New Approaches for Prediction of Optimum Operating Pressures of Multistage Oil and Gas Separators

The aim of this study is to present a simple methodology for optimizing separators pressures in the crude oil production unit with methodological HYSYS simulation and also an Artificial Neura l Network (ANN) approach. In ANN approach each stage has trained distinctly. The results show, error between real output and network output is very low. So the network can respond for other input data. The new proposed HYSYS methodology determines the optimum pressures of separators in different stages of separation and consequently optimizes the operating conditions using HYSYS simulation. In this study, the optimum pressures of separators in a 3896 Kgmole/hr oil production unit were determined. As an obtained result, the oil recovery was increased about 6 Kgmole/hr, and the quality of total produced oil was improved by 0.4 °API.

Keywords:

oil recovery , multi stage separator , ANN