Title :
A Hierarchical Modeling and Fault Diagnosis Technique for Complex Electronic Devices
Author :
Bing Long ; Zhi-jian Dai ; Shu-Lin Tian ; Hou-Jun Wang
Author_Institution :
Sch. of Autom. Eng., Univ. of Electron. Sci. & Technol. of China, Chengdu
Abstract :
Due to the shortcomings of the diagnosis systems presently used by complex electronic devices system such as failure models hard to build and low fault isolation resolution, a new hierarchical Modeling and diagnosis technique is proposed based on multisignal model and support vector machine (SVM). Multisignal model is used to describe the failure propagation relationship in electronic device system, and the most probable failure printed circuit boards (PCBs) or circuit modules are reasoned by max Bayes postior probability. The exact failure modes/components in the PCB/Module can be identified by SVM. The results show the proposed modeling and diagnosis methods are effective and are suitable for complex electronic devices diagnosis.
Keywords :
Bayes methods; electronic engineering computing; fault diagnosis; printed circuits; support vector machines; PCB; SVM; complex electronic devices; fault diagnosis technique; max Bayes postior probability; multisignal model; printed circuit boards; support vector machine; Aerospace electronics; Aircraft; Circuit faults; Diagnostic expert systems; Dictionaries; Fault diagnosis; Satellites; Spaceborne radar; Support vector machine classification; Support vector machines;
Conference_Titel :
Testing and Diagnosis, 2009. ICTD 2009. IEEE Circuits and Systems International Conference on
Conference_Location :
Chengdu
Print_ISBN :
978-1-4244-2587-7
DOI :
10.1109/CAS-ICTD.2009.4960745