Fouling analysis of a shell and tube heat exchanger using local linear wavelet neural network

A local linear wavelet neural network based model has been developed to predict the temperature differences on both the tube and shell side and the heat exchanger efficiency. This network replaces the straightforward weight by a local linear model. The working process of the proposed network can be viewed as to decompose the complex, nonlinear system into a set of locally active submodels and then smoothly integrate those submodels by their associated wavelet basis functions. For a given approximation or prediction problem with sufficient accuracy, the local linear models provide more power than a constant weight model as the dilation and translation parameters of LLWNN are randomly generated and optimized without predetermination. The closeness of the predicted results with the actual experimental results and higher accuracy with maximum error of 1.25% indicates that LLWNN can be used as a suitable tool for simulation of heat exchangers subjected to fouling. (C) 2014 Published by Elsevier Ltd.