31 December 2008 Modeling the sensor calibration based on RNN with feature extraction
Author Affiliations +
Proceedings Volume 7130, Fourth International Symposium on Precision Mechanical Measurements; 713019 (2008) https://doi.org/10.1117/12.819583
Event: Fourth International Symposium on Precision Mechanical Measurements, 2008, Anhui, China
Abstract
In this work a novel method is proposed for the sensor calibration, by combining Recurrent Neural Network (RNN) with feature extraction. RNN is used as the neural network model trained to calibrate the measuring error, while Kernel Independent Component Analysis (KICA) and Independent Component Analysis (ICA) are introduced in as the feature extraction as comparison. And by examining the data of an example of temperature sensor calibration of Agilent 34970, it is shown that the proposed methods can both perform good calibration comparing with single RBF method. And the KICA method performs better than the ICA method.
© (2008) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Hui Zhang, Yong Ni, Yue Chang, "Modeling the sensor calibration based on RNN with feature extraction", Proc. SPIE 7130, Fourth International Symposium on Precision Mechanical Measurements, 713019 (31 December 2008); doi: 10.1117/12.819583; https://doi.org/10.1117/12.819583
PROCEEDINGS
6 PAGES


SHARE
Back to Top