Modeling of thermodynamic properties of substances by neural networks

A new method based on neural networks was developed for the modeling of thermodynamic properties of substances. When applied to the mixture of air and H/sub 2/O, the preset accuracy of 1% was obtained at every test point and the neural networks proved to be 5000 times faster than a conventional iterative algorithm. Large tables characteristic of previous interpolation methods are not needed. The neural network models enable new process simulation applications.