Multiparameter Microwave Sensor for Rapid and Nondestructive Grading of In-Shell Nuts

Authors: Trabelsi, Samir; Lewis, Micah

Submitted to: IEEE Instrumentation & Measurement Society
Publication Type: Proceedings
Publication Acceptance Date: 3/25/2024
Publication Date: N/A
Citation: N/A

Interpretive Summary: Multiple quality parameters are needed in the grading process of in-shell peanuts. Among these parameters are kernel moisture content and "meat" content. Kernel moisture content is the one single parameter that determines if a truck load of peanuts is a "sale" or "no sale". In Georgia, kernel moisture content has to be less than 10.49% to be declared a "sale". Currently, during the grading process well-trained inspectors clean the in-shell peanuts samples from foreign materials, inspect them visually, size them, and shell them before determining the kernel moisture content. This is tedious, time consuming, and uses outdated pieces of equipment. The technology proposed allows the simultaneous determination of three important quality attributes of peanuts, namely, bulk density, kernel moisture content, and "meat" content, without having to shell them. By determining these quality attributes at the front end of the grading process instantaneously and nondestructively, the grading time is reduced drastically resulting in significant time and labor savings. Measurements on in-shell peanuts were performed with a low-cost microwave sensor that was assembled, calibrated, and tested in the laboratory and in the field. The sensor measures the dielectric properties which are considered the electric "finger print" of the material at a single microwave frequency (here 5.8 Gigahertz, and correlates them to the quality attributes of interest. In this paper, equations correlating dielectric properties of in-shell peanuts to bulk density, kernel moisture content, and "meat" content are given along with estimates of prediction errors. The sensor and calibration algorithm presented here for in-shell peanuts can be used for other in-shell nuts.

Technical Abstract: A calibration algorithm for simultaneous determination of three quality attributes of in-shell nuts, namely, bulk density, moisture content, and “meat” content with the same low-cost microwave sensor is proposed. The sensor principle is based on free-space measurement of the dielectric properties of in-shell nuts at a single microwave frequency of 5.8 GHz. In this paper, measurements were performed on in-shell peanuts (pods). Calibration equations correlating each quality attribute with in-shell peanuts dielectric properties are given along with the standard error of calibration (SEC).