| Title |
Air-coupled ultrasound spectroscopy air parameters compensation technique / |
| Authors |
Nakutis, Ž ; Kaškonas, P ; Liaukonis, D ; Svilainis, L |
| DOI |
10.1109/JSEN.2024.3369508 |
| Full Text |
|
| Is Part of |
IEEE Sensors journal.. Piscataway, NJ : IEEE. 2024, vol. 24, iss. 8, p. 12667-12675.. ISSN 1530-437X. eISSN 1558-1748 |
| Keywords [eng] |
air-coupled ultrasound ; error compensation ; inverse problem ; plant sensor ; resonance ultrasound spectroscopy ; thickness and velocity measurement |
| Abstract [eng] |
The air-coupled resonance ultrasound spectroscopy (RUS) of thin plate thickness, density, ultrasound velocity and attenuation measurement are affected by air parameters. If air parameters are left unaccounted errors will occur. Conventional thermometer measurements are not efficient because temperature can vary faster than temperature sensor response. Technique for air parameters estimation and compensation of the RUS inverse solution results is proposed. The ultrasound delay over known distance is used for velocity in air estimation. There is no need for the additional measurement: the propagation time between transducers can be obtained from RUS calibration measurement. Ultrasound velocity in air is then used for temperature estimation. These measurements are augmented by pressure sensor measurement for air density estimation. Evaluation of the attainable measurement errors and analysis of uncertainties under such compensation was carried out using simulated signals. Sensitivity coefficients for every parameter were derived and attainable errors evaluated for temperature range from -5 °C to +40 °C and atmospheric pressure range from 94 kPa to 105 kPa. It was concluded that the relative uncertainty of sample attenuation, ultrasound velocity attenuation, density and thickness could be reduced approximately 22 times compared to case when air parameters are assumed to be equal to those in normal conditions. Experimental verification used 2 mm polycarbonate plate, measured values were compared against reported data. Experiment confirmed the efficiency of the proposed compensation: thickness estimation bias errors were reduced 17 times, bias errors for density were reduced 15 times and velocity estimation bias errors were reduced 5 times. |
| Published |
Piscataway, NJ : IEEE |
| Type |
Journal article |
| Language |
English |
| Publication date |
2024 |
| CC license |
|
