DocumentCode
1746663
Title
Fault identification in analog PWL circuits based on homotopy
Author
Robotycki, A. ; Zielonko, R.
Author_Institution
Fac. of Electron., Telecommun. & Inf., Tech. Univ. Gdansk, Poland
Volume
1
fYear
2001
fDate
21-23 May 2001
Firstpage
59
Abstract
The paper deals with a verification method of diagnosis of analog piecewise linear circuits based on homotopy approach. The homotopy maps the passing way of one function into another under influence of homotopic parameter t∈[0,1]. The homotopy can be used for description of voltages in measurement accessible nodes of circuit under test in relation to circuit element parameter changes. In the presented method, homotopy paths associated with particular elements have been applied for verification of fault hypotheses. The assumed hypothesis is true if different homotopy paths concerning measurement nodes have their ends in one point. This gives the possibility of fault localization and also identification on the basis of the path end point determination, The method is illustrated on the example of the single fault diagnosis of one-stage transistor amplifier
Keywords
analogue circuits; circuit analysis computing; fault diagnosis; graph theory; network topology; nonlinear network analysis; piecewise linear techniques; analog piecewise linear circuits; bilinear transformation; circuit element parameter changes; fault hypotheses verification; fault identification; fault localization; homotopic parameter; homotopy approach; homotopy paths; measurement accessible nodes; one-stage transistor amplifier; single fault diagnosis; voltage description; Circuit faults; Circuit testing; Equations; Fault diagnosis; Informatics; Linear circuits; Piecewise linear techniques; Robots; Transistors; Voltage;
fLanguage
English
Publisher
ieee
Conference_Titel
Instrumentation and Measurement Technology Conference, 2001. IMTC 2001. Proceedings of the 18th IEEE
Conference_Location
Budapest
ISSN
1091-5281
Print_ISBN
0-7803-6646-8
Type
conf
DOI
10.1109/IMTC.2001.928788
Filename
928788
Link To Document