Evolving smart approach for determination dew point pressure through condensate gas reservoirs

Abstract

To design gas condensate production planes with low uncertainty along with robust reservoir simulation, precise estimation or monitoring of dew point pressure play a crucial role. To handle successfully the addressed issue of condensate gas reservoirs, massive attentions have been performed previously but unfortunately fail to develop accurate approach for estimation dew point pressure. Dedicated to this fact, in current study enormous attempts have been put forth to proposed revolutionary method for determining dew point pressure in gas condensate reservoirs. To gain this end the new type of support vector machine method which evolved by Suykens and Vandewalle was utilized to generate robust approach to figure dew point pressure in condensate gas reservoir out. Also, lucrative and high precise dew point pressures reported in previous attentions were carried out to test and validate support vector machine approach. To serve better understanding of the proposed support vector machine approach, the conventional feed-forward artificial neural network and couple of genetic algorithm (GA) and fuzzy logic applied to the referred data banks and the gained solutions were contrasted with each other. According to the root mean square error (RMSE), correlation coefficient and average absolute relative deviation, the suggested support vector machine approach has acceptable reliability, integrity and robustness draw an analogy with the artificial neural network model and conventional methods. Thus, the proposed intelligent based way can be considered as an alternative model to monitor the dew point pressure of condensate gas reservoirs when the required real data are not accessible.