Predicting hydrate stability zones of petroleum fluids using sound velocity data of salt aqueous solutions

Determination of Critical Properties and Acentric Factors of Petroleum Fractions Using Artificial Neural Networks

“…ANNs have large numbers of computational units called neurons, connected in a massively parallel structure, and do not need an explicit formulation of the mathematical or physical relationships of the handled problem. 2, [7][8][9][10][11][12][13][14][15][16] The most commonly used ANNs are the feed-forward neural networks, 10 which are designed with one input layer, one output layer, and hidden layers. 2,[7][8][9][10][11][12][13][14][15][16] The number of neurons in the input and output layers equals to the number of 'inputs' and 'outputs' physical quantities, respectively.…”

“…2, [7][8][9][10][11][12][13][14][15][16] The most commonly used ANNs are the feed-forward neural networks, 10 which are designed with one input layer, one output layer, and hidden layers. 2,[7][8][9][10][11][12][13][14][15][16] The number of neurons in the input and output layers equals to the number of 'inputs' and 'outputs' physical quantities, respectively. The disadvantage of FNNs is the determination of the ideal number of neurons in the hidden layer(s); few neurons produce a network with low precision, and a higher number leads to overfitting and bad quality of interpolation and extrapolation.…”

“…ANNs have large numbers of computational units called neurons, connected in a massively parallel structure, and do not need an explicit formulation of the mathematical or physical relationships of the handled problem. ,− The most commonly used ANNs are the feed-forward neural networks, which are designed with one input layer, one output layer, and hidden layers. ,− The number of neurons in the input and output layers equals to the number of ‘inputs' and ‘outputs' physical quantities, respectively. The disadvantage of FNNs is the determination of the ideal number of neurons in the hidden layer(s); few neurons produce a network with low precision, and a higher number leads to overfitting and bad quality of interpolation and extrapolation.…”

“…The optimization method chosen in this work was the Levenberg−Marquart algorithm 4-6 along with Bayesian regularization technique, as mentioned earlier. This algorithm is specially indicated to optimize ANNs using small learning sample size. − The testing set is used during the adjustment of the network's synaptic weights to evaluate the algorithms performance on the data not used for tuning and stop the tuning if the error on the testing set increases . Finally, the validation set measures the generalization ability of the model after the fitting process …”

“…To our knowledge, such a technique has not already been applied to determine these properties for petroleum fractions. Among the various ANNs reported in the literature, the feed-forward (back-propagation) neural network (FNN) method with a modified Levenberg−Marquardt optimization algorithm − is used, which is known to be effective to represent the nonlinear relationships between variables in complex systems. ,− To develop this model, the data recommended in the literature data 1 on the critical properties and acentric factors of C 1 −C 45 petroleum fractions are used. The developed model is then employed to predict independent data (not used in the training and developing the ANN model).…”