Neural networks for chemical engineering unit operations

Application of artificial neural networks in simulation of chemical enginee ring unit operations are studied. On the basis of the present work it is su ggested that the raw data describing the phenomenon should be processed so that the relationships of the input variables are studied and if relationsh ips are found, such as dimensionless numbers, these relationships are set a s inputs in the neural network. Further, if the phenomenon can be described by a simplified mathematical expression, the value calculated by such simp lified solution is set as one of the inputs. The method is demonstrated by three examples, namely, by turbulent fluid flow in a tube, breakthrough cur ve of adsorption, and by suspension crystallization. In all cases the metho d presented in this work results in a more accurate simulation and in easie r training of the neural network.