A knowledge-based approach for fault detection and isolation in analog circuits

A new knowledge-based analog fault detection and isolation technique is proposed. It is based on utilizing the domain knowledge in order to design a training set which characterizes the behavior of the circuit under test in both fault-free and fault situations. The training set expressed as a set of rules is then mapped into a rule-based connectionist neural network. This network is trained to perform the desired fault isolation. The effectiveness of the technique is demonstrated through a testing example.