Smart Determination of Difference Index for Asphaltene Stability Evaluation

Abstract

Precipitation and deposition of asphaltene during different stages of petroleum production is recognized as problematic in oil industry because of the increase in production cost and the inhibition of a consistent flow of crude oil in different medium. Numerous correlations have been developed to determine asphaltene stability in crude oil. In this study, a novel ONN method was used to estimate difference index from SARA fraction data for rapid, accurate, and cost-effective determination of asphaltene stability. Neural networks are highly in danger of trapping in local minima. To eliminate this flaw, a hybrid genetic algorithm-pattern search technique was used instead of common back-propagation algorithm for training the employed neural network. A comparison between neural network and optimized neural network indicated superiority of optimized neural network.

Prediction of crude oil refractive index through optimized support vector regression: a competition between optimization techniques

“…Gholami et al (2014a) presented a new mathematical approach based on fuzzy logic to predict the value of crude oil RI from SARA fraction data. In another report, Gholami et al (2014b) attempted to create a correlation between these two parameters. They published a smart model based on neural network optimized by a genetic algorithm pattern that searches for estimation of crude oil RI as a function of SARA fraction data.…”

An improved support vector regression model for estimation of saturation pressure of crude oils

“…The GA uses three main rules including selection, crossover and mutation rules to converge individuals into global minimum. Finally, the GA terminates when a stop condition such as maximum number of generation is satisfied [53][54][55].…”

Estimation of the longitudinal dispersion coefficient via a fusion of optimized models

“…For longitudinal dispersion coefficient modeling considering interrelated influencing parameters, a proper selection of input parameters is important (Dehghani et al 2020). Since an intelligent model learns the relationship between input variables and the output variable, it can estimate the target value in unseen data (Gholami et al 2014b(Gholami et al , 2018 The performance in computing the longitudinal dispersion coefficient was judged by using two criteria as given by Equations ( 3)-( 6): where Y i obs is the measured value of sample i, Y i pred is the estimated value of sample i, Ŷobs is the average of real values, and n is the number of samples. When the values of MSE, MAE, and PB are close to zero and the value of R 2 is close to 1, a model with superior performance is achieved.…”