A Neural Network Classifier with Multi-Valued Neurons for Analog Circuit Fault Diagnosis

• DOI: 10.3390/electronics10030349
• Publication Date: 2021-02-02T10:44:42Z
• Authors: Igor Aizenberg; Riccardo Belardi; Marco Bindi; Francesco Grasso; Stefano Manetti; Antonio Luchetta; Maria Cristina Pi...
• Keywords: Analogue electronics

In this paper, we present a new method designed to recognize single parametric faults in analog circuits. The technique follows rigorous approach constituted by three sequential steps: calculating the testability and extracting ambiguity groups of circuit under test (CUT); localizing failure putting it correct fault class (FC) via multi-frequency measurements or simulations; (optional) estimating value faulty component. fabrication tolerances healthy components are taken into account every step procedure. work combines machine learning techniques, used for classification approximation, with analysis procedures