ANFIS for analysis friction factor and Nusselt number of pulsating nanofluids flow in the fluted tube under magnetic field

Abstract

Application of adaptive neuro-fuzzy inference system to analyze friction factor and the Nusselt number of pulsating nanofluids in the helically corrugated tube with magnetic field effect is presented. Based on the optimum adaptive neuro-fuzzy inference system (ANFIS) model configuration, it has four input parameters; pulsating flow frequency, nanofluids concentration, mass flow rate and power input. The present experimental data are divided into two subsets for training and testing processes of ANFIS network. ANFIS tunes a fuzzy inference system with back-propagation algorithm and least square estimation approaches to determine the friction factor and the Nusselt number. The predicted results of the proposed ANFIS model are compared with the measured data. There is an excellent agreement between the predicted results and the experimental results and gives average errors of ±2.5%-±5.0%, ±2.5% for the friction factor and Nusselt number, respectively. The ANFIS model is an alternative powerful and reliable method as compared with other methods. It can be used with confidence for predicting thermal performance of the complex thermal systems. © 2020 The Authors.