Development of Artificial neural network and neuro-fuzzy for Taguchi based erosion behavior of GF-reinforced polyester composites

Effects of various implemented angle for different fiber loading on erosion behavior of Polyester composites reinforced with three different weight fractions of woven E-glass fiber reinforcement are studied. Multiple inputs single output (MISO) models were developed to predict mechanical properties using Taguchi based Adaptive Neuro-Fuzzy Inference System (TANFIS) and artificial neural network (ANN ). An individual ANFIS model was applied for each mechanical property using various angles (15o, 30o, 45o ,60o ,75o and90o) and fiber loadings (glass fiber weight fractions (40 , 50 and 60 wt%)) as input parameters and erosion behavior as output parameter. To this end, an attempt has been made in this work to in a simplified manner and develop empirical model. The effect of various test parameters and their interactions have been studied using Taguchi method to find out optimal parameter setting for minimum errors. Models were evaluated by the R2 of prediction and root mean squared error (RMSE) for training and testing. It has been observed that the results are in a very good agreement with the experimental ones.

کلیدواژه ها:

composites ، Taguchi method ، ANN ، ANFIS