A New Method for Analog Circuit Fault Diagnosis Based on Testability and Chaos Particle Swarm Optimization

Provided by: AICIT
Topic: Hardware
Format: PDF
A new method of low-testability fault diagnosis for analog circuit is proposed. Its main content includes: the transfer relation between test nodes and inputs of a circuit is computed based on the topological structure and the information of finite test points of the circuit. The diagnosability of each component in the circuit is analyzed qualitatively to classify diagnosable component groups of the test circuit, effectively avoiding any invalid diagnosis of inseparable components. The identification of fault classification is carried out by Support Vector Machine (SVM) established for parameter optimization so as to accurately implement components ' fault diagnosis based on the determined diagnosable groups, using Support Vector Machine (SVM) that adopts chaos particle swarm optimization to effectively improve the accuracy of fault diagnosis.

Find By Topic