Predicting residual bagging bend height of knitted fabric using fuzzy modelling and neural networks

In this research, fuzzy modelling and neural network methods were used and compared to predict the residual bagging bend height of knitting fabric samples. Studies undertaken to minimize the bagging phenomenon vary significantly with the test conditions including the experimental field of interest, the input parameters and the applied method. Hence, we attempt to formulate a theoretical model of predicting bagging behaviour in our experimental design of interest. By analysing the bend height of overall bagging samples, this paper provides an effective neural network model to evaluate and predict the residual bagging bend height of the knitting specimens after test. It also provides the impact of each input parameter in our experimental field of interest to simulate this phenomenon after use. Moreover, the contribution of these influential input parameters was analysed and discussed. Nevertheless, our results show that residual bagging height decreases when yarn contains elastane filament, Spandex©. This finding is in agreement with Mirostawa et al. [11] that with an increase of the elastane content in fabric, permanent bagging decreases, whereas elastic bagging increases. According to the analytical results obtained, the neural network model gives a more accurate prediction than the fuzzy one.