PREDICTION OF DIMETHYL ETHER DENSITY USING ARTIFICIAL NEURAL NETWORK

In this work, the ability of Artificial Neural Network based on back-propagation algorithm to predicting of dimethyl ether density has been investigated. Several feed-forward neural network with different architectures were tried to determine the best network configuration. TheLevenberg-Marquardt algorithm is applied as the training rule. Comparisons of results show, a good agreement between experimental data and artificial neural network predictions. Results prove that artificial neural network can be a successful tool to effectively represent complex nonlinearsystems, if developed efficiently. An important feature of the model is its needlessness to anytheoretical knowledge or human experience during the network training process.

Keywords:

Neural network , Dimethyl ether , Density , Thermodynamic properties