Title :
Calibration of fiber-optic sensors using universal approximators
Author :
Porada, E. ; Bock, W.J.
Author_Institution :
Lab. d´´Optoelectron., Quebec Univ., Hull, Que., Canada
Abstract :
A calibration-through-learning method for fiber-optic sensors in measurement systems is presented. The method involves exact integral representations of the measurand value, initially using a continuum of hidden processing units. In the next step, discrete subsets of the continuum uniformly approximate the measurand. This allows for a design of connectionist networks capable of the same approximation capability
Keywords :
calibration; computer architecture; computerised instrumentation; fibre optic sensors; image processing equipment; neural nets; approximation capability; calibration; connectionist networks; discrete subsets; fiber-optic sensors; hidden processing units; integral representations; learning; universal approximators; voltaic models; Calibration; Current measurement; Diodes; Distortion measurement; Inverse problems; Optical fiber sensors; Pressure measurement; Sensor systems; Strain measurement; Voltage;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 1994. IMTC/94. Conference Proceedings. 10th Anniversary. Advanced Technologies in I & M., 1994 IEEE
Conference_Location :
Hamamatsu
Print_ISBN :
0-7803-1880-3
DOI :
10.1109/IMTC.1994.352076
