Applying GMDH artificial neural network to predict dynamic viscosity of an antimicrobial nanofluid

Objective (s): Artificial Neural Networks (ANN) are widely used for predicting systems behavior. Group Method of Data Handling (GMDH) is a type of ANNs which has remarkable ability in pattern recognition. The aim the current study was to propose a model to predict dynamic viscosity of silver/water nanofluid which could be used as antimicrobial fluid for several medical purposes. Materials and Methods: In order to have precise model, it is necessary to consider all influential factors. Temperature, concentration and size of nanoparticles are used as input variables of the model. In addition, GMDH artificial neural network is applied to design a proper model. Data for modeling are extracted from conducted experimental studies published in valuable journals. Results: The dynamic viscosity of Ag/water nanofluid is precisely modeled by using GMDH. The obtained values for R-squared is equal to 0.9996 which indicates perfect precision of the proposed model. In addition, the highest relative deviation for the model is 2.2%. Based on the values of these statistical criteria, the model is acceptable and very accurate.