Optimal design approach for the plate-fin heat exchangers using neural networks cooperated with genetic algorithms

Abstract

The paper demonstrates the successful application of genetic algorithm (GA) combined with back propagation neural networks (BP) for the optimal design of plate-fin heat exchangers (PFHE). The major objectives in the PFHE design are the minimum total weight and total annual cost for a given constrained conditions. The total weight target ensures an exchanger of the smallest size with minimum capital cost, whereas the annual cost target yields the optimum pressure drops accounting for the tradeoff between power consumption and heat exchanger weight. Total length and width of PFHE core, number of hot side layers, fin height and pitch on each side of PFHE are considered as variables to be optimized by means of GA combined with BP method. This optimization method of PFHE is universal and can be used for various PFHEs.