A NEURAL NETWORK BASED SYSTEM FOR SOFT FAULT DIAGNOSIS IN ELECTRONIC CIRCUITS
Journal Title: Metrology and Measurement Systems - Year 2005, Vol 12, Issue 4
Abstract
The paper considers the architecture and the main steps of development of a neural network based system for diagnosis of soft faults in analog electronic circuits. The definition of faults of interest, selection of an optimal set of measurements, feature extraction, the construction of the artificial neural network, training and testing the network, are considered. A fault dictionary method was implemented in the system. Experimental results are presented on an example of diagnosis of a 6-th order bandpass filter. The measuring part of the system performs input-output measurements in the frequency domain with the aid of a HP 4192 Transmitance Analyzer.
Authors and Affiliations
WOJCIECH TOCZEK, MICHAŁ KOWALEWSKI
NOVELTIES IN ELECTRICAL MEASUREMENTS AND INSTRUMENTATION<br /> IN THE WAKE OF IMEKO TC4 & TC 19 SYMPOSIA AND ADC WORKSHOP<br /> IN IASI, ROMANIA, SEPTEMBER 2007
From 18 to 22 September 2007, at the Technical University “Gh. Asachi” of Iasi, Romania, there took place undoubtedly important scientific events, held under the auspices of the International Measurement Confederation....
ANALYTICAL MODELLING OF KARMAN VORTEX STREET
A new approach to the Karman vortex street modelling is presented. The model is based on observations of the Karman vortex street. The idea of the model is based on fundamental mechanical formulas. According to the model...
PROBE CAPACITANCE-DEPENDENT SYSTEMATIC ERROR IN I-V MEASUREMENTS OF NANOWIRES: ANALYSIS AND CORRECTION<br />
We propose a method of eliminating the systematic error due to the capacitance of digital oscilloscope probes in the experimental setup used for tracing current-voltage (I-V) curves of nanowires with quantum point contac...
LIFTING BASED COMPRESSION ALGORITHM FOR POWER SYSTEM SIGNALS
The paper concerns the problem of power system signals compression with possible application in system monitoring and control. The proposed compression algorithm for power system signals ensures efficient use of availabl...
DEVELOPMENT AND VALIDATION OF A SELECTIVE METHOD FOR THE DETERMINATION OF CHRYSENE USING SILVER-ENHANCED ROOM-TEMPERATURE PHOSPHORIMETRY<br />
A factorial design was applied for the development of a solid surface room-temperature phosphorimetric method aiming the selective determination of chrysene. Data analysis was made using statistical experimental analysis...