Journal of the Korean Society of Propulsion Engineers (한국추진공학회지)

The Korean Society of Propulsion Engineers (한국추진공학회)
권호 표지
DOI QR코드

DOI QR Code

Application of Artificial Neural Network to Flamelet Library for Gaseous Hydrogen/Liquid Oxygen Combustion at Supercritical Pressure

초임계 압력조건에서 기체수소-액체산소 연소해석의 층류화염편 라이브러리에 대한 인공신경망 학습 적용

  • Jeon, Tae Jun (School of Mechanical Engineering, Kyungpook National University) ;
  • Park, Tae Seon (School of Mechanical Engineering, Kyungpook National University)
  • Received : 2021.09.08
  • Accepted : 2021.12.09
  • Published : 2021.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

To develop an efficient procedure related to the flamelet library, the machine learning process based on artificial neural network(ANN) is applied for the gaseous hydrogen/liquid oxygen combustor under a supercritical pressure condition. For hidden layers, 25 combinations based on Rectified Linear Unit(ReLU) and hyperbolic tangent are adopted to find an optimum architecture in terms of the computational efficiency and the training performance. For activation functions, the hyperbolic tangent is proper to get the high learning performance for accurate properties. A transformation learning data is proposed to improve the training performance. When the optimal node is arranged for the 4 hidden layers, it is found to be the most efficient in terms of training performance and computational cost. Compared to the interpolation procedure, the ANN procedure reduces computational time and system memory by 37% and 99.98%, respectively.

층류화염편 라이브러리에 대한 효율적인 계산과정을 개발하기 위하여 초임계 압력조건의 기체수소/액체산소 연소기에 대해 인공신경망을 이용한 기계학습과정이 적용되었다. 학습성능과 계산효율성에 근거한 최적의 계산과정을 찾기 위하여 은닉층에 대한 ReLU와 쌍곡탄젠트 함수의 25가지 조합이 선택되었다. 정확성이 우수한 높은 학습성능을 얻는데 쌍곡탄젠트 활성화함수가 적절하였다. 인공신경망의 학습성능을 개선하기 위해서 학습데이터 변환이 제안되었다. 4개의 은닉층에 최적의 노드를 배치할 때 학습성능 및 계산비용 관점에서 모두 효율적인 것으로 나타났다. 층류화염편 라이브러리의 보간법보다 인공신경망을 사용하는 경우 전체 계산시간은 37%, 시스템 메모리는 99.98% 감소되었다.

Keywords

Acknowledgement

이 성과는 정부의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(NRF-2016R1D1A1B02012446).

References

  1. Waxenegger-Wilfing, G., Dresia, K., Deeken, J. and Oschwald, M., "Machine Learning Methods for the Design and Operation of Liquid Rocket Engines - Research Activities at the DLR Institute of Space Propulsion," Space Propulsion, Virtual, 2021.
  2. Schmitt, T., "Large-Eddy Simulations of the Mascotte Test Cases Operating at Supercritical Pressure," Flow, Turbulence and Combustion, pp. 159-189, 2020.
  3. Zhang, J., Huang, H., Xia, Z., Ma, L., Duan, Y., Feng, Y. and Huang, J., "Artificial Neural Networks for Chemistry Representation in Numerical Simulation of the Flamelet-Based Models for Turbulent Combustion," IEEE, Vol. 8, pp. 80020-80029, 2020.
  4. Milan, P.J., Wang, X., Hickey, J.-P., Li, Y. and Yang, V., "Accelerating Numerical Simulations of Supercritical Fluid Flows using Deep Neural Networks," AIAA Scitech 2020 Forum, Orlando, U.S.A., AIAA 2020-1157, Jan. 2020.
  5. Li, B., Lee, Y., Yao, W., Lu, Y. and Fan, X., "Development and Application of ANN Model for Property Prediction of Supercritical Kerosene," Computers and Fluids, Vol. 209, 104665, 2020. https://doi.org/10.1016/j.compfluid.2020.104665
  6. Kim, W.H. and Park, T.S., "An Evaluation of Numerical Schemes in a RANS-based Simulation for Gaseous Hydrogen/Liquid Oxygen Flames at Supercritical Pressure," J. Kor. Soc. of Propulsion Engineers, pp. 21-29, 2013.
  7. Thomas, J.L. and Zurbach. S., "Test case RCM-3: Supercritical Spray Combustion at 60 bars at Mascotte," Proceedings, 2nd International Workshop on Rocket Combustion Modeling, Lampoldhausen, Germany, pp. 13-23, 2001.
  8. Li, J., Zhao, Z., Kazakov, A. and Dryer, F. L., "An Updated Comprehensive Kinetic Model of Hydrogen Combustion," Int. J. Chemical Kinetics, Vol. 36, pp. 566-575, 2004. https://doi.org/10.1002/kin.20026
  9. Soave, G., "Equilibrium Constants from a Modified Redlich-Kwong Equation of State," Chemical Engineering Science, Vol. 37, pp. 1197-1203, 1972. https://doi.org/10.1016/0009-2509(72)80096-4
  10. Chung, T.-H., Ajlan, M., Lee, L.L. and Starling, K.E., "Generalized Multiparameter Correlation for Nonpolar and Polar Fluid Transport Properties," Ind. Eng. Chemistry Research, Vol. 27, pp. 671-679, 1988. https://doi.org/10.1021/ie00076a024
  11. Kim, S.-K., Joh, M., Choi, H.S. and Park, T.S., "Multidisciplinary Simulation of a Regeneratively Cooled Thrust Chamber of Liquid Rocket Engine: Turbulent Combustion and Nozzle Flow," Int. J. of Heat and Mass Transfer, Vol. 70, pp. 1066-1077, 2014. https://doi.org/10.1016/j.ijheatmasstransfer.2013.10.046
  12. Marquardt, D.W., "An Algorithm for Least-Squares Estimation of Nonlinear Parameters," J. Soc. Ind. and Applied Mathematics, Vol. 11, No. 2, pp. 431-441, 1963. https://doi.org/10.1137/0111030
  13. He, K., Zhang, X., Ren, S. and Sun, J., "Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification," IEEE Int. Conference on Computer Vision, pp. 1026-1034, 2015.
  14. Glorot, X. and Bengio, Y., "Understanding the Difficulty of Training Deep Feedforward Neural Networks," Int. Conference on Artificial Intelligence and Statistics, pp. 249-256, 2010.
  15. Kim, T., Kim, Y. and Kim, S.-K., "Numerical Analysis of Gaseous Hydrogen/Liquid Oxygen Flamelet at Supercritical Pressures," Int. J. of Hydrogen Energy, Vol. 36, pp. 6303-6316, 2011. https://doi.org/10.1016/j.ijhydene.2011.02.043