Hierarchical approach to diagnosis using ANNs

Feed-forward artificial neural networks (ANNs) have been applied to defects diagnosis in an electronic circuit, and also a hierarchical approach is introduced in diagnosis procedure. The approach is presented on an example of a mixed-mode circuit that can represent every complex circuit. A voting system is created in order to distinguish which ANN´s output is to be accepted as the final diagnostic statement. Three examples illustrate this approach, so validating the effectiveness of this procedure.