Journal of the Korean Society of Mineral and Energy Resources Engineers (한국자원공학회지)

The Korean Society of Mineral and Energy Resources Engineers (한국자원공학회)
권호 표지
DOI QR코드

DOI QR Code

Forecasting Methane Gas Concentration of LFG Power Plant Using Deep Learning

딥러닝 기법을 활용한 매립가스 발전소 포집공의 메탄가스 농도 예측

  • 원승현 ((주)하나티이씨) ;
  • 서대호 (연세대학교 정보대학원) ;
  • 박대원 (서울과학기술대학교 에너지환경대학원 에너지환경공학과)
  • Received : 2018.10.12
  • Accepted : 2018.12.20
  • Published : 2018.12.31
⟨ Previous Next ⟩

Abstract

In this study, after operational data for a landfill gas power plant were collected, the methane gas concentration was predicted using a deep learning method. Concentrations of methane gas, carbon dioxide, hydrogen sulfide, oxygen concentration, as well as data related to the valve opening degree, air temperature and humidity were collected from 23 pipeline bases for 88 matches from January to November 2017. After the deep learning model learned the collected data, methane gas concentration was estimated by applying other data. Our study yielded extremely accurate estimation results for all of the 23 pipeline bases.

본 연구는 매립장 매립가스 발전소를 대상으로 발전소 운영 데이터들을 수집 후, 딥러닝(Deep Learning) 기법을 적용하여 향후 메탄가스 농도를 예측하였다. 2017년 1월부터 11월까지 88일치에 대해서 23개 포집공에서 메탄가스 농도, 이산화탄소 농도, 황화수소 농도, 산소 농도, 밸브 개방정도, 기온, 습도 데이터를 수집하였다. 수집 데이터로 딥러닝 모델을 학습한 후 실제 데이터와 비교하였다. 추정 결과 23개 포집공 모두에서 매우 정확한 추정결과를 보였다.

Keywords

Acknowledgement

Grant : 저발열량 매립가스를 이용한 희박연소 가스 엔진 발전시스템 개발

Supported by : 환경부

References

  1. Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., and Isard, M., 2016. Tensorflow: a system for large-scale machine learning. OSDI, 16, 216-283.
  2. Ministry of Environment., 2006. Investigation of landfill greenhouse gas emissions and construction of statistics, 5530010, Sejong, Korea, p.1-418.
  3. Ministry of Environment., 2014. Status of waste generation and disposal nationwide, 11-B552584-000005-10, Sejong, Korea, p.1-1365.
  4. Fabbricino, M., 2007. Evaluating operational vacuum for landfill biogas extraction. Waste Management, 27(10), 1393-1399. https://doi.org/10.1016/j.wasman.2006.10.006
  5. IPCC, 1997. Revised 1996 IPCC Guidelines for National GHG Inventories, Kyoto, Japan, p.1-88.
  6. Kim, J.Y., 2015. About Google Tensorflow. Korea Computer Information Association, 23(2), 9-15.
  7. Korea Energy Economics Institute, 2016. A comparative analysis of post 2020 GHG reduction targets in major countirist, 16-07, Ulsan, Korea, p.1-28.
  8. LeCun, Y., Bengio, Y., and Hinton, G., 2015. Deep learning. Nature, 521(7553), 436. https://doi.org/10.1038/nature14539
  9. Marvin, M. and Seymour, A. P., 1969. Perceptrons. MIT Press, Cambridge, USA, p.1-37.
  10. McCulloch, W.S. and Pitts, W., 1943. A logical calculus of the ideas immanent in nervous activity. The Bulletin of Mathematical Biophysics, 5(4), 115-133. https://doi.org/10.1007/BF02478259
  11. Christensen, T.H. and Kjeldsen, P., 1989. Basic biochemical processes in landfills. Technology and Environmental Impact, Academic Press, New York, p.29-49.
  12. Rosenblatt, F., 1958. The perceptron: a probabilistic model for information storage and organization in the brain. Psychological Review, 65(6), 386. https://doi.org/10.1037/h0042519
  13. Russell, S. J and Peter, N., 2003. Artificial intelligence: a modern approach (2nd Ed.), Prentice Hall, New Jersey, USA, p.1-1152.
  14. Won, D.K. and Lee., S.P., 2016. Artificial intelligence and implications of the fourth industrial revolution. Ie Magazine, 23(2), 13-22.