HETP (Height Equivalent of a Theoretical Plate) and pressure drop are parameters used by engineers to design distillation columns. These parameters are usually estimated by using empirical factors and models. The purpose of this study is to determine if a neural network model for HETP and pressure drop can be used satisfactorily. Data similar to data used to develp the traditional models is used to train and test a neural network. The neural network model is very accurate when compared to the experimental HETP and pressure drop values and more accurate than the predicted values of a traditional model. The sensitivity of the network output values to changes in the network inputs was also studied.
展开▼
