In the present work, the methodologies adopted towards the optimal design process of the discrete neural network classifier blocks of discrete ‘ANN’ simulation model, which are designed on the basis of ‘multilayer perceptron’ (MLP) and ‘radial basis function’ (RBF) type of neural network architecture are presented. The design of discrete ‘ANN’ simulation model is done for the classification and qualitative assessment of the state of degradation of insulation present in the respective phases of three-phase ac induction motor. The optimal design specifications realized for the various neural network classifier blocks, which are designed on the basis of ‘multilayer perceptron’ (MLP) and ‘radial basis function’ (RBF) types of neural network architectures are compared. On the basis of comparative analysis it is possible to suggest the appropriate design strategy, which is suitable for the successful hardware implementation of the mentioned application.

Full text: IJEIR_1694_Final.pdf