1 March 2007 Improved method of hysteresis compensation for a piezoelectric fiber optic voltage sensor
G. Fusiek, Pawel Niewczas, James R. McDonald
Author Affiliations +
Abstract
We report on an improved numerical method for hysteresis compensation within a hybrid piezoelectric fiber optic voltage sensor. The previous technique relied on laborious and relatively inaccurate individual curve-fitting procedures to create multiple sets of lookup functions used for the final derivation of the measured voltage. The new technique uses an aggregate 3-D arrangement of the captured hysteresis loops and applies a surface-fitting algorithm based on the Levenberg-Marquardt method to create two 3-D lookup functions, which are then used to derive the instantaneous value of the measured voltage. Furthermore, a more advanced algorithm for selecting top and bottom hysteresis-loop parts has been applied to eliminate errors associated with the incorrect selection at low voltage levels by the previous algorithm. The proposed enhancements greatly simplify the calibration process and significantly reduce measurement errors. The technique, implemented using a real-time signal-processing system, was tested and its effectiveness evaluated experimentally. The new algorithm provided complete phase error compensation, from approximately 7 to 0 deg, and magnitude error compensation down to 0.15% (full-scale output)—an improvement of more than 3 times over the previous technique.
©(2007) Society of Photo-Optical Instrumentation Engineers (SPIE)
G. Fusiek, Pawel Niewczas, and James R. McDonald "Improved method of hysteresis compensation for a piezoelectric fiber optic voltage sensor," Optical Engineering 46(3), 034401 (1 March 2007). https://doi.org/10.1117/1.2714931
Published: 1 March 2007
Lens.org Logo
CITATIONS
Cited by 7 scholarly publications.
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Sensors

Fiber optics sensors

Fiber Bragg gratings

Optical engineering

Fiber optics

Signal processing

Algorithm development

Back to Top