A fault diagnosis algorithm for nonlinear analog circuits and systems



Journal Title

Journal ISSN

Volume Title


Texas Tech University


A fault detection algorithm for nonlinear analog circuits is developed on the basis of the unique characteristics of the solid stage nonlinear device. A large-scale system can be decomposed into subsystems by driving the nonlinear devices into their cut-off regions. The reversed bias applied to the nonlinear devices wi11 reveal the failures caused by the shorted or leaky devices. The resultant decoupled linear subsystems are of much smaller simension. Test signals can be applied to each subsystem as long as it is decoupled from the other. A diagnosis scheme is developed for testing linear subsystems. The test results are used to construct the test signature table to indicate the failed linear components in each subsystem. The overall system which, is free of shorted or leaky devices and/or failed linear components, is then put back to the normal operating condition. The static operating point and the dynamic performance of each nonlinear device will give the indication of failures caused by opened or deteriorated devices. The fault diagnosis for the large-scale system is thus completed. Several numerical examples are given to illustrate the test scheme and the possible implementation of the automatic test system.



Electric networks, Nonlinear theories, Electric fault location, Nonlinear