Skip to main content
ARS Home » Research » Publications at this Location » Publication #100637
Title: MULTIVARIATE CALIBRATION FOR OPTIMIZATION OF A MICROWAVE MOISTURE CONTENT METHOD FOR GRAIN VARYING IN TEMPERATURE AND BULK-DENSITY

Author
item Archibald, Douglas
item TRABELSI, SAMIR - OICD
item Kraszewski, Andrzej
item Nelson, Stuart

Submitted to: Electromagnetic Wave Interaction with Water and Moist Substances Proceeding
Publication Type: Proceedings
Publication Acceptance Date: 4/11/1999
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: Multivariate calibration methods are adapted to multifrequency microwave data for a set of wheat samples that were treated to introduce substantial variations in moisture-content (M), temperature (K) and bulk-density (D). Three analysis issues are outlined: 1) linearization of input variables, 2) scaling and weighting of variables, and 3) estimation of model error. The results of the analysis make it possible to evaluate a wide variety of instrumental configurations that might be proposed to suit particular engineering criteria such as measurement accuracy, range of operating conditions, and hardware complexity. Moreover, additional insight can be obtained by projecting frequency-dependent trends in model error beyond the instrumental parameters employed in the original study.