Multiple hard fault classification in analog circuits using learning vector quantization neural networks

Abstract EN

A new neural network-based fault classification strategy for hard multiple faults in analog circuits is proposed The magnitude of the' harmonics of the Fourier components of the circuit response at different lest nodes due to a sinusoidal input signal are first measured or simulated A selection criterion for determining the best components that describe the circuit behavior under fault-free (nominal) and fault situations is presented An algorithm that estimates the overlap between different faults in the measurement space is also introduced I he learning vector quantization neural network is then effectively named to classify circuit faults Performance measures reveal very high classification accuracy in both naming and testing stages A detailed example which demonstrates the proposed strategy, is described.