Broken rice detection based on microwave measurement technique using microstrip wide-ring sensor and microstrip coupled-line sensor

This paper proposes a microstrip wide-ring sensor and a microstrip coupled-line sensor for measuring the percentage of broken rice (BR) based on microwave measurement technique. Sensors with low insertion loss were developed to operate within a frequency range from 1 GHz to 3 GHz. The variations in...

Full description

Bibliographic Details
Main Authors: Mun, Hou Kit (Author), You, Kok Yeow (Author), Dimon, Mohamad Ngasri (Author)
Format: Article
Language:English
Published: 2013.
Subjects:
Online Access:Get fulltext
Description
Summary:This paper proposes a microstrip wide-ring sensor and a microstrip coupled-line sensor for measuring the percentage of broken rice (BR) based on microwave measurement technique. Sensors with low insertion loss were developed to operate within a frequency range from 1 GHz to 3 GHz. The variations in percentage of BR with magnitude and phase of transmission coefficient were investigated using regression analysis at selected frequencies. Calibration equations for the measurement of the percentage of BR have been obtained and validated with white rice for percentage of BR ranging from 0% to 100%. The coupled-line sensor exhibits higher sensitivity to BR detection as compared with the wide-ring sensor. For a 1% change in BR, the change of phase for a coupled-line sensor is 0.00471 rad, around 20% of BR at 2.28 GHz. However, the change of phase for a wide-ring sensor with 1% change in BR is 0.00154 rad, around 20% of BR at 2.50 GHz. Also, the wide-ring sensor has greater accuracy in BR prediction than the coupled-line sensor. The wide-ring sensor has been found to be best suited for BR detection via the phase of transmission coefficient measurement at 2.50 GHz with 2.32% average error within 0% to 20% of BR. The coupled-line sensor is best suited for BR prediction from 20% to 60% by means of the measured phase of transmission coefficient at 2.28 GHz with an average error of less than 9.7%