Predicting heat transfer rate of a ribbed triple-tube heat exchanger working with nanofluid using neural network enhanced by advanced optimization algorithms

Abstract

Eight optimizer methods are combined with a perceptron neural network to achieve an optimal network and minimize errors for predicting the heat transfer rate of a ribbed triple-tube heat exchanger operating with the graphene nanoplatelets-based nanofluid. The optimization techniques consist of Harris Hawks Optimizer (HHO), Grey Wolf Optimizer (GWO), Whale Optimization Algorithm (WOA), Artificial Bee Colony (ABC), Ant Colony Optimization (ACO), Ant Lion Optimizer (ALO), Biogeography-Based Optimization (BBO), and Dragonfly Algorithm (DA). The required data are provided with the aid of numerical simulations. The structural parameters are considered using different rib pitches and heights. The ALO algorithm is the best method for estimating the output. The best performance of this algorithm is gained by population size of 350. By this method, the heat transfer rate is estimated with the Root Mean Square Error (RMSE) values of about 0.0310 and 0.0385 for the training and testing data samples, respectively.