암모니아 가스 농도 예측을 위한 딥러닝 활용 검토

전민규¹ ; 윤성환²^{, *}

1해군사관학교 기계시스템공학과 부교수
2국립한국해양대학교 기관시스템공학부 부교수

A Study on Deep Learning Approaches for Predicting Ammonia Gas Concentration

Min-Gyu Jeon¹ ; Sung Hwan Yoon²^{, *}

1Associate professor, Dept. of Mechanical System Engineering, Republic of Korea Naval Academy
2Associate professor, Division of Marine System Engineering, Korea Maritime and Ocean University

Correspondence to: *Sung Hwan Yoon Division of Marine System Engineering, Korea Maritime and Ocean University 727 Taejong-ro, Yeongdo-Gu, Busan, 49112, Republic of Korea Tel: +82-51-404-3985 E-mail: shy@kmou.ac.kr

초록

최근 산·학·연·군 협력은 국방 혁신을 진전시키고자 하며, 해군은 미래 전장을 위한 신기술을 모색하고 있다. 탈탄소를 위한 대체 연료인 암모니아(NH₃)는 선상에서 독성을 나타내며 발화 위험이 높다. 본 연구에서는 SD-LBC를 이용한 TDLAS를 사용하여 암모니아 농도를 측정한다. 5개의 선택된 흡수 파장에 1D-CNN을 적용하여 시계열 데이터로부터 농도 분류를 평가하였다.

Abstract

Recent industry–academia–military collaborations aim to advance defense innovation, with the Navy exploring emerging technologies for future battlefields. Ammonia (NH₃), a carbon-free alternative fuel, poses toxicity and high ignition risks onboard. This study measures NH₃ concentrations using non-contact TDLAS with SD-LBC to extract target gas absorbance. A 1D-CNN using five selected absorption wavelengths was applied to evaluate real-time concentration classification from time-series data.

Keywords:

Ammonia, Gas Leakage Detection, TDLAS, SD-LBC, 1D-CNN

키워드:

암모니아, 가스 누출 감지, 가변 다이오드 레이저 흡수 분광법, 표준편차-선형기준수정, 1차원 합성곱 신경망

Acknowledgments

이 논문은 2025년도 한화시스템(주)의 재원을 받아 수행된 연구임.

References

J. K. Magnuson, T. N. Anderson, R. P. Lucht, “Application of a Diode-laser-based Ultraviolet Absorption Sensor for in situ Measurements of Atomic Mercury in Coal-combustion Exhaust,” Energy and Fuels, Vol. 22, No. 5, pp. 3029-3036, 2008.
T. Kamimoto, Y. Deguchi, Y. Kiyota, “High Temperature Field Application of Two Dimensional Temperature Measurement Technology Using CT Tunable Diode Laser Absorption Spectroscopy,” Flow Measurement and Instrumentation, Vol. 46, No. A, pp. 51-57, 2015.
D. W. Choi, M. G. Jeon, G. R. Cho, T. Kamimoto, Y. Deguchi, D. H. Doh, “Performance Improvements in Temperature Reconstructions of 2-D Tunable Diode Laser Absorption Spectroscopy (TDLAS),” Journal of Thermal Science, Vol. 25, No. 1, pp. 84-89, 2016.
M. G. Jeon, Y. Deguchi, T. Kamimoto, D. H. Doh, G. R. Cho, “Performances of New Reconstruction Algorithms for CT-TDLAS (Computer Tomography-tunable Diode Laser Absorption Spectroscopy),” Applied Thermal Engineering, Vol. 115, pp. 1148-1160, 2017.
M. G. Jeon, D. H. Doh, Y. Deguchi, “Measurement Enhancement on Two-dimensional Temperature Distribution of Methane-air Premixed Flame Using SMART Algorithm in CT-TDLAS,” Applied Sciences, Vol. 9, No. 22, 4955, 2019.
M. G. Jeon, J. W. Hong, D. H. Doh, Y. Deguchi, “A Study on Two-dimensional Temperature and Concentration Distribution of Propane-air Premixed Flame using CT-TDLAS,” Modern Physics Letters B, Vol. 34, No. 7n9, 2040020, 2020.
M. G. Jeon, G. R. Cho, “A Study on Continuous Temperature Measurement Using Direct Absorption Spectroscopy for Fuel-air Premixed Flames,” Journal of Mechanical Science and Technology, Vol. 37, No. 8, pp. 4109-4116, 2023.
J. W. Hong, S. H. Yoon, M. G. Jeon, “A Study on Baseline Selection for Ammonia Concentration Measurement Using TDLAS,” Journal of Mechanical Science and Technology, Vol. 39, No. 3, pp. 1255-1261, 2025.
M. G. Jeon, “Deep Learning-based Precision Measurement of Gas Leakage Inside Naval Vessels,” Journal of the KNST, Vol. 8, No. 2, pp. 236-240, 2025.
M. A. Z. Chowdhury, M. A. Oehlschlaeger, “Deep Learning for Gas Sensing via Infrared Spectroscopy,” Sensors (Basel), Vol. 24, No. 6, 1873, 2024.
S. Kaur, N. Singh, A. Kumar, W. S. Philanim, V. K. Verma, A. K. Patil, R. Bhardwaj, A. Riar, “Evaluating the Impact of Hyperparameters on the Performance of 1D CNN Model for Nutritional Profiling of Underutilized Crops using NIRS Data,” Smart Agricultural Technology, Vol. 11, 101025, , 2025.