Transactions of the Korean hydrogen and new energy society (한국수소및신에너지학회논문집)

The Korean Hydrogen and New Energy Society (한국수소및신에너지학회)
권호 표지
DOI QR코드

DOI QR Code

Performance Prediction Model of Solid Oxide Fuel Cell Stack Using Deep Neural Network Technique

심층 신경망 기법을 이용한 고체 산화물 연료전지 스택의 성능 예측 모델

  • LEE, JAEYOON (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials) ;
  • PINEDA, ISRAEL TORRES (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials) ;
  • GIAP, VAN-TIEN (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials) ;
  • LEE, DONGKEUN (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials) ;
  • KIM, YOUNG SANG (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials) ;
  • AHN, KOOK YOUNG (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials) ;
  • LEE, YOUNG DUK (Department of Clean Fuel and Power Generation, Korea Institute of Machinery and Materials)
  • 이재윤 (한국기계연구원 청정연료발전연구실) ;
  • 이스라엘 또레스 삐네다 (한국기계연구원 청정연료발전연구실) ;
  • 잡 반 티엔 (한국기계연구원 청정연료발전연구실) ;
  • 이동근 (한국기계연구원 청정연료발전연구실) ;
  • 김영상 (한국기계연구원 청정연료발전연구실) ;
  • 안국영 (한국기계연구원 청정연료발전연구실) ;
  • 이영덕 (한국기계연구원 청정연료발전연구실)
  • Received : 2020.08.18
  • Accepted : 2020.10.30
  • Published : 2020.10.30
Download PDF
⟨ Previous Next ⟩

Abstract

The performance prediction model of a solid oxide fuel cell stack has been developed using deep neural network technique, one of the machine learning methods. The machine learning has been received much interest in various fields, including energy system mo- deling. Using machine learning technique can save time and cost requried in developing an energy system model being compared to the conventional method, that is a combination of a mathematical modeling and an experimental validation. Results reveal that the mean average percent error, root mean square error, and coefficient of determination (R2) range 1.7515, 0.1342, 0.8597, repectively, in maximum. To improve the predictability of the model, the pre-processing is effective and interpolative machine learning and application is more accurate than the extrapolative cases.

Keywords

References

  1. T. S. Park, T. H. Nam, D. G. Lee, S. K. Lee, and J. S. Moon, "A study of deduction of evaluation items for design of SOFC stack safety performance evaluation system", J. Korean Inst. Gas., Vol. 21, No. 6, 2017, pp. 81-87, doi: https://doi.org/10.7842/kigas.2017.21.6.81.
  2. F. Ramadhani, M. A. Hussain, H. Mokhlis, and S. Hajimolana, "Optimization strategies for solid oxide fuel cell (SOFC) application: a literature survey", Renewable Sustainable Energy Rev., Vol. 76, 2017, pp. 460-484, doi: https://doi.org/10.1016/j.rser.2017.03.052.
  3. A. Kirubakaran, S. Jain, and R. K. Nema, "A review on fuel cell technologies and power electronic interface", Renewable Sustainable Energy Rev., Vol. 13, No. 9, pp. 2430-2440, doi: https://doi.org/10.1016/j.rser.2009.04.004.
  4. M. C. Williams, S. D. Vora, and G. J. Gary, "Worldwide status of solid oxide fuel cell technology" ECS Transactions, Vol. 96, No.1, 2020, pp. 1-10, doi: https://doi.org/10.1149/09601.0001ecst.
  5. M. Yoda, S. Inoue, Y. Takuwa, K. Yasuhara, and M. Suzuki, "Development and commercialization of new residential SOFC CHP system", ECS transactions, Vol. 78, No.1, 2017, pp. 125-132, doi: https://doi.org/10.1149/07801.0125ecst.
  6. W. M. Chen and H. N. Kim, "Energy, economic, and social impacts of a clean energy economic policy: fuel cells deployment in Delaware", Energy Policy, Vol. 144, 2020, pp. 111617, doi: https://doi.org/10.1016/j.enpol.2020.111617.
  7. M. Noponen, P. Torri, J. Goos, J. Puranen, H. Kaar, S. Pylypko, and E. Ounpuu, "Elcogen-next generation solid oxide cell and stack technology", ECS Transactions, Vol. 91, No. 1, 2019, pp. 91-97, doi: https://doi.org/10.1149/09101.0091ecst.
  8. Y. S. Kim, Y. D. Lee, and K. Y. Ahn, "System integration and proof-of-concept test results of SOFC-engine hybrid power generation system", Applied Energy, Vol. 277, 2020, 115542, doi: https://doi.org/10.1016/j.apenergy.2020.115542.
  9. S. Malek, F. Melgani, and Y. Bazi, "One-dimensional convolutional neural networks for spectroscopic signal regression", J. Chemom., Vol. 32, No. 5, 2018, pp. e2977, doi: https://doi.org/10.1002/cem.2977.
  10. J. M. Jo, "Effectiveness of normalization pre-processing of big data to the machine learning performance", The Journal of the Korea Institute of Electronic Communication Sciences, Vol. 14, No. 3, 2019, pp. 547-552, doi: https://doi.org/10.13067/JKIECS.2019.14.3.547.
  11. R. Ma, T. Yang, E. Breaz, Z. Li, P. Briois, and F. Gao, "Data-driven proton exchange membrane fuel cell degradation predication through deep learning method", Applied Energy, Vol. 231, 2018, pp. 102-115, doi: https://doi.org/10.1016/j.apenergy.2018.09.111.
  12. Y. H. Jang, J. Y. Kim, and S. K. Oh, "A study on information data segmentation method using optimization algorithm", Journal of Korean Institute of Intelligent Systems, Vol. 28, No. 4, 2018, pp. 321-327, doi: https://doi.org/10.5391/JKIIS.2018.28.4.321.