Mojtaba Ghaedi - Chemical and Petroleum Engineering Department, Sharif University of Technology, Tehran, Iran
Ali Nejad Ebrahimi - Chemical and Petroleum Engineering Department, Sharif University of Technology, Tehran, Iran
Mahmoud Reza Pishvaie - Chemical and Petroleum Engineering Department, Sharif University of Technology, Tehran, Iran

The produced reservoir fluids are first processed at the surface to get a stabilized liquid phase. This is done by two, three or more separators consecutively at lower pressures. When separator pressures are optimized, then more oil will be produced due to less hydrocarbon losses to the gasphase; so proper selection of separator pressures has a great effect on the both quantity and quality of the produced stock tank liquid. In this paper, genetic algorithm was used to optimize the separator pressures in a multistage crude oil production unit with four separators and in amultistage gas condensate production unit with three separators. Regarding the quality of remained oil in the stock tank, genetic algorithm predicts better separators pressures in comparison with theoperating pressures for both crude oil and gas condensate production unit. Predicted pressures forcrude oil, will result in 33.97 and 32.25 API gravity of stock tank liquid for summer and winter respectively while these values in summer and winter for operation conditions are 33.16 and 31.56 API gravity of stock tank liquid. Also in the case of predicted pressures by genetic algorithm forgas condensate multistage separation, it can be said that API gravity of stock tank liquid will be 64.06 in summer and 63.23 in winter, while thosegiven by operating pressures are 62.42 and 61.65 in summer and winter respectively.

Multistage separation, Crude oil, Gas condensate, Optimization, Genetic Algorithm