Processing calibration data of low-temperature thermometer based on clustering algorithm

Area of Science:

• Cryogenics
• Metrology
• Sensor Technology

Background:

• High cost and long calibration lead times limit cryogenic thermometer availability.
• In-lab calibration systems are needed for rapid deployment of uncalibrated or custom sensors.

Purpose of the Study:

• To develop a straightforward, high-accuracy in-lab calibration system for cryogenic thermometers.
• To implement an automated data processing model for efficient calibration.

Main Methods:

• Utilized copper screws as thermal links to minimize temperature oscillations on the sensor platform (to millikelvin levels).
• Developed and optimized a data processing model employing clustering algorithms for automated data analysis.
• Integrated a cryogen-free refrigerator for temperature control.

Main Results:

• Cryogenic temperature sensors calibrated showed high accuracy with <1% relative error against standard thermometers.
• Automated data processing for two uncalibrated thermometers completed in just over 10 minutes.
• Temperature oscillation on the sensor platform was suppressed to a few millikelvins.

Conclusions:

• The developed system offers a cost-effective and efficient solution for in-lab cryogenic thermometer calibration.
• The automated data processing model significantly reduces calibration time and enhances accuracy.
• This approach enables faster utilization of uncalibrated and self-made cryogenic sensors.