Presenter

Mr Gokarna Pandey - ETRO, Vrije Universiteit Brussel [Email]

Abstract

Materials are characterized by their properties such as moisture content, ice content, density, and volumetric fraction of other constituents. Accurate determination of these properties of interest (POIs) are important for legal and labelling requirements, quality control, economic reasons and production process monitoring. These POIs can be determined using sensor solutions which immensely benefit the industries by increasing their throughput and energy saving while maintaining a consistent quality. This PhD thesis focusses on the design of compact, cost-effective and industrial millimeter (mm) wave sensor topologies and the development of novel methodologies for extraction of the POIs using reflection and transmission coefficients.
First, the dielectric theory and electromagnetic wave interactions with the materials at mm-wave frequencies are modelled which has been lacking in the literature. Vector network analyzer (VNA) and table-top-based quasi-optical system often used for material characterization in free space. We overcome this conventional and expensive way by designing and developing the cost-effective mm-wave sensors using off-the-shelf components. Thereafter, a free-space filter-based dielectric permittivity sensitivity enhancement technique is invented to enable the detection of very small POI variations. Consequently, inverse problems are solved to extract the POIs using reflection and transmission coefficients. A novel microscopic dielectric spectroscopy (uDS) is also invented to extract multiple POIs at the same time. Finally, lab prototypes are converted into industrial sensors.
The methodology for the level of frozenness (LOF) extraction of food products by real-time inline measurements has been validated in different companies. The developed sensor solution optimizes throughput and energy consumption, and simultaneously allows the real-time quality control of the food products. By implementing the enhancement technique in the sensor assemblies, the sensitivity of the sensor increased by 4 orders of magnitude and resolutions of 0.00045% is obtained for moisture content measurement

Short CV

Master in Electrical Engineering, KU Leuven, 2012