Machine Learning-Driven Permittivity Prediction of Liquids Using a Novel Encircled Coupling Microwave Sensor

To enhance the accuracy and sensitivity of liquid permittivity measurements, a novel microwave sensor developed using an encircled coupling Split Ring Resonator(SRR) structure. The proposed sensor uses a dielectric substrate integrated with a microstrip line and SRR, which significantly improved sensitivity due to increased electric field confinement and coupling. The distinctive resonant characteristics of the proposed sensor structure used for precise permittivity evaluation. Simulation results demonstrates a frequency shift from 2.59 GHz to 2.02 GHz as the liquid permittivity changes from 1 to 100 and has a highest frequency shift of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 7 4 ~ M H z}$</tex>. The proposed sensor used to measure the permittivity of the liquid sample using relationship between resonant frequency and permittivity, achieving a Quality factor of 313.61 and a sensitivity of up to <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1. 1 6 \%}$</tex>.