Improvement of accuracy of dew and frost point measurement and estimation for residual water in gas.

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K04329
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 19020:Thermal engineering-related
• Research Institution: Saga University
• Principal Investigator: Ishida Kenji 佐賀大学, 理工学部, 講師 (20304876)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 微量水分 / 露点測定 / 可視化解析 / 近赤外光 / 偏光 / サーモグラフィ / 機械学習

Outline of Final Research Achievements

To realize highly accurate concentration monitoring of residual water in gases, which is important in semiconductor manufacturing and hydrogen filling stations, a visualization analysis system using near-infrared light for detailed analysis of phenomena occurring on the mirror surface of cooled mirror dew point sensor during measurement has been significantly improved. In addition, a visualization optical system using polarized light was designed and newly integrated into the system. Furthermore, the author proposed the application of thermography to dew point measurement. As a result of examining the effectiveness of these three methods and the possibility of applying them to prototype sensors, the author judged that the method combining the visualization of the sensor's mirror surface using a polarized image sensor and image analysis incorporating machine learning was the most superior method, and fabricated a prototype sensor based on this method to confirm its effectiveness.

Free Research Field

熱工学

Academic Significance and Societal Importance of the Research Achievements

本研究の「偏光イメージセンサによるセンサ鏡面の可視化と機械学習を導入した画像解析を組み合わせた手法」に基づく試作センサは，従来型センサと比べて，偏光イメージセンサにより霜を極めて高感度に検出可能，従来型の鏡面冷却式露点センサで必要であったレーザー光学系を取り除く事ができるため高圧化に適する，画像解析に機械学習を導入したことにより光学系設定の厳密さを抑えられる，照明光の選択の自由度が非常に高い等の優位性がある．本手法は，従来型の鏡面冷却式露点センサの課題を克服して高精度化と高圧対応を実現できる可能性があり，今後の改良を経て，気体中の微量水分の濃度モニターにおいて強力なツールとなり得ると考える．