Multi-resonant Non-intrusive Microwave Sensor for Liquid Identification with High Accuracy

Jinglei Yong,Xingli He,Lingfeng Li,Peng Li,Yunjing Zhang

DOI: https://doi.org/10.1109/piers62282.2024.10618437

2024-01-01

Abstract:Microwave sensors composed of split-ring resonators (SRRs) have proven to be effective in liquid characterization owing to the high sensitivity. Improving the structure of the SRR will enhance the sensitivity and resolution of the sensor. In this paper, an interdigital structure for the SRR is proposed to expand the sensing area and augment the sensitivity. In addition, The sensor in this work integrates multiple SRRs with different resonances to measure corresponding responses in the range of 2–6 GHz. Consequently, when measuring the liquid, a wealth of data spanning a broad frequency band, that is, dispersion relationship can be obtained. By the polynomial regression, the sensor can thoroughly characterize the complex permittivity of the measured liquid, ensuring accurate and reliable identification. Five liquids, water, ethanol, methanol, acetone, and ether, were used to assess the performance of the sensor. Initially, a commercial dielectric probe is employed to measure the liquid’s complex permittivity as the reference. Then the sensor is utilized to measure S 21 of the liquids, thus the resonant frequency f and quality factor Q can be obtained. A prediction model of the complex permittivity is proposed, showing an accurate correlation between the permittivity and the resonant frequency shift Δf and quality factor shift ΔQ. Comparing the measured permittivity with the reference, we find the maximum relative error is 6.79 × 10 −5 %, which is extremely small. Therefore, the sensor enables to identify different liquids with significant accuracy. Because of multiple resonances, the sensor can measure the complex permittivity at four resonant frequencies across 2–6 GHz. This presents the dispersion of the liquid to be measured in the frequency band, eliminating the possible false identification associated with single-resonance measurement. An experiment on identifying the proposed five liquids, plus a comparison with state-of-the-art sensors, demonstrate the superior performance of the multiple resonant sensor. In addition, the sensor has a simple structure and fast response time, enabling on-site and real-time measurements which are demanded in safety inspection, bio-monitoring, etc.