Optimized Multi Frequency Approach to Analog Fault Diagnosis Using Monte Carlo Analysis

摘要

The proliferation of large analog circuits and systems of ever increasing complexity has stirred great interest in different methods for fault diagnosis of analog circuits. In Analog circuit fault diagnosis, the measurements of the circuit are used in the construction of the fault dictionary. The fault dictionary represents readings of the circuit under different fault conditions and one nominal condition at different test frequencies. Based on the fault dictionary readings, sets of faults which have almost the same fault signature are identified and are called ambiguity sets. In this paper the multi frequency approach to fault diagnosis has been optimized using the concept of sub-ambiguity tables. It has been shown here that the number of test frequencies required to diagnose faults has been significantly reduced and also the simulation time taken has been reduced.