Title :
ANN-based error reduction for experimentally modeled sensors
Author :
Arpaia, Pasquale ; Daponte, Pasquale ; Grimaldi, Domenico ; Michaeli, Linus
Author_Institution :
Facolta di Ingegneria, Universita del Sannio, Benevento, Italy
fDate :
2/1/2002 12:00:00 AM
Abstract :
A method for correcting the effects of multiple error sources in differential transducers is proposed. The correction is carried out by a nonlinear multidimensional inverse model of the transducer based on an artificial neural network. The model exploits independent information provided by the difference in actual characteristics of the sensing elements, and by an easily controllable auxiliary quantity (e.g., supply voltage of conditioning circuit). Experimental results of the correction of an eddy-current displacement transducer subject to the combined interference of structural and geometrical parameters highlight the practical effectiveness of the proposed method
Keywords :
displacement measurement; eddy currents; electric sensing devices; error compensation; intelligent sensors; neural nets; ANN-based error reduction; artificial neural network; auxiliary quantity; conditioning circuit; differential transducers; displacement measurements; eddy currents; eddy-current displacement transducer; error compensation; geometrical parameters; intelligent sensors; modeled sensors; multiple error sources; neural network applications; nonlinear multidimensional inverse model; sensing elements; structural parameters; Artificial neural networks; Circuits; Error correction; Intelligent sensors; Interference; Inverse problems; Multidimensional systems; Proposals; Transducers; Voltage control;
Journal_Title :
Instrumentation and Measurement, IEEE Transactions on
