Chipless wireless sensor coupled with nachine learning for oil temperature monitoring.
Chipless wireless sensor.
Machine learning (ML).
Oil temperature.
Permittivity characterization.
Issue Date: 
Estévez, A., Fodop, J., Sancho, J. I., Valderas, D., Ochoa, I., & Pérez, N. (2023). Chipless Wireless Sensor Coupled with Machine Learning for Oil Temperature Monitoring. IEEE Sensors Journal.
Temperature monitoring is essential in several industries driving the need for sensors. Chipless radio frequency identification (RFID) technology has emerged as a cost-effective solution, enabling wireless detection without the need for a power supply or electronics embedded in the sensor tags. However, a significant challenge lies in wirelessly monitoring temperature within liquid environments using chipless RFID tags as resonances vanish due to energy absorption in liquids. This work presents a chipless RFID sensor for wireless detection of oil temperature in a glass container. The temperature monitoring is based on the characterization of the permittivity of oil samples with different concentrations of total polar compounds (TPCs). After evaluating two chipless RFID tag designs, we propose to use a complementary ring resonator (CRR) tag as it exhibits a robust response to oil liquid volume, improving the detection of temperature in low-loss liquids and offering higher sensitivity. When the measurement results are coupled with machine learning (ML), we demonstrate that the response of the proposed tag as a wireless sensor can be used to estimate the temperature of oil samples with different quality (TPC) with an average test RMSE of 4 degrees C (standard deviation < 2 degrees C), in the approximate range 22 degrees C-95 degrees C.

Files in This Item:
3.72 MB
Adobe PDF

Statistics and impact

Items in Dadun are protected by copyright, with all rights reserved, unless otherwise indicated.