Prediction of carbon dioxide water content using a particle swarm optimization-artificial neural network hybrid model
Koolivand-Salooki, M, Azarmehr, A, Bahadori, A, Kord, Sh, Sharifi, F, Alfkhan, Sh 2015, 'Prediction of carbon dioxide water content using a particle swarm optimization-artificial neural network hybrid model', Petroleum & Coal, vol. 57, no. 5, pp. 573-586.
which includes carbon dioxide, because CO2 exhibits a minimum in the water content which cannot be predicted accurately by existing methods. In the present study, particle swarm optimization (PSO) was coupled to artificial neural network (ANN) to develop a robust hybrid training algorithm with both local and global search capabilities to predict carbon dioxide water content accurately. A database of CO2 water content is used to train and validate the PSO-ANN hybrid model. The proposed model was used for optimization of ANN structure using particle swarm optimization (PSO). To demonstrate the capacity of the proposed model, we provide performance surface of the model. The results show that the proposed hybrid learning algorithm simulates the CO2 water content at various conditions accurately.