Developing a Model Based on a Hybrid Neural Particle Swarm Optimization for Prediction of Dew Point Pressure

Dew point pressure is one of the most important parameters to characterize gas condensate reservoirs. Experimental determination of dew point pressure (DDP) in a window PVT cell is often difficultespecially in case of lean retrograde gas condensate. Therefore, searching for fast and robust algorithms for determination of DPP isusually needed. Despite of the wide range of applications and flexibility of ANNs in petroleum industries, design and structural optimization ofneural networks is still strongly dependent upon the designer's experience. To mitigate this problem, this paper presents a newapproach based on a hybrid neural particle swarm optimization to determine the DPP. Then, equations for DPP prediction by using theoptimized weights of network have been generated. With the obtained correlation, the user may use such results without a running the ANNsoftware. Consequently, this new model is compared with results obtained using other conventional models to make evaluation among different techniques. The results show that the hybrid model can be applied effectively and afford high accuracy and dependability for DPP forecasting for the wide range of gas properties and reservoir temperatures.

Keywords:

Gas condensate reservoir , Dew point pressure , Artificial neural network , Particle swarm optimization