Prediction of Thermal performance nanofluid Al2O3 by Artificial Neural Network and Adaptive Neuro-Fuzzy Inference Systemt

In recent years, the use of modeling methods that directly utilize empirical data is increasing due to the high accuracy in predicting the results of the process, rather than statistical methods. In this paper, the ability of Artificial Neural Network (ANN) and Adaptive Fuzzy-Neural Inference System (ANFIS) models in the prediction of the thermal performance of Al2O3 nanofluid that is measured by thermal resistance is investigated. The laboratory data was extracted from a valid paper that examined the thermal performance of Al2O3 nanoparticles in an oscillating tube. For modeling by ANN, a multi-layered perceptron network was used and for ANFIS a sugeno fuzzy model was used that are both the most accurate and most commonly used modeling methods. Comparison of target values ​​with predicted values ​​by both models was highly satisfactory, and the correlation coefficient for both of them was more than 0.99, which indicates that the accuracy of these two models is high. Finally, the performance of both models was compared with each other, which was very good and close, but in general ANN showed better performance than ANFIS.