Development of a CFD-Suitable Deep Neural Network Model for Laminar Burning Velocity
Source abstract
Abstract
Hydrogen is a valued resource for today’s industry. As a fuel, it produces large amounts of energy and creates water during the process, unlike most other polluting energy sources. However, the safe use of hydrogen requires reliable tools able to accurately predict combustion. This study presents the implementation of a deep neural network of laminar burning velocity of hydrogen into an open-source CFD solver flameFoam. DNN was developed based on a previously created larger DNN, which was too large for CFD applications since the calculations took around 40 times longer compared to the Malet correlation. Therefore, based on the original model, a faster, but still accurate, DNN was developed and implemented into flameFoam starting with version 0.10. The paper presents the adaptation of the original DNN into a CFD-applicable version and the initial test results of the CFD–DNN simulation.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Combustion and Detonation Processes · Advanced Combustion Engine Technologies · Combustion and flame dynamics
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Laboratory of Nuclear Installation Safety, Lithuanian Energy Institute, Breslaujos g. 3, LT-44403 Kaunas, Lithuania
References (54)
Show source-provided references
1. A logical calculus of the ideas immanent in nervous activity10.1007/BF02478259 ↗
2. Phillipson, F. (2020, January 11–12). Quantum Machine Learning: Benefits and Practical Examples. Proceedings of the International Workshop on QuANtum SoftWare Engineering & PRogramming (QANSWER), Talavera, Spain.
3. Applications of Machine Learning in Fatty Live Disease Prediction
4. Zhou, J., Zhou, Z., Zhao, Q., Han, Z., Wang, P., Xu, J., and Dian, Y. (2020). Evaluation of Different Algorithms for Estimating the Growing Stock Volume of Pinus massoniana Plantations Using Spectral and Spatial Information from a SPOT6 Image. Forests, 11.10.3390/f11050540 ↗
5. An implementation of artificial neural-network potentials for atomistic materials simulations: Performance for TiO210.1016/j.commatsci.2015.11.047 ↗
6. Efficient and Accurate Machine-Learning Interpolation of Atomic Energies in Compositions with Many Species10.1103/PhysRevB.96.014112 ↗
7. Laminar burning velocity predictions of single-fuel mixtures of C1-C7 normal hydrocarbon and air
8. Laminar burning velocity of hydrogen and carbon-monoxide enriched natural gas (HyCONG): An experimental and artificial neural network study10.1016/j.fuel.2019.03.003 ↗
9. A machine learning methodology for improving the accuracy of laminar flame simulations with reduced chemical kinetics mechanisms10.1016/j.combustflame.2020.02.021 ↗
10. Malik, K., Żbikowski, M., and Teodorczyk, A. (2020). Laminar Burning Velocity Model Based on Deep Neural Network for Hydrogen and Propane with Air. Energies, 13.10.3390/en13133381 ↗
11. Machine learning model to predict the laminar burning velocities of H2/CO/CH4/CO2/N2/air mixtures at high pressure and temperature conditions10.1016/j.ijhydene.2019.11.134 ↗
12. Application and comparison of multiple machine learning techniques for the calculation of laminar burning velocity for hydrogen-methane mixtures10.1016/j.tsep.2022.101306 ↗
13. Prediction of Gasoline Blend Ignition Characteristics Using Machine Learning Models10.1021/acs.energyfuels.1c00749 ↗
14. Exploring the fuel structure dependence of laminar burning velocity: A machine learning based group contribution approach10.1016/j.combustflame.2021.111525 ↗
15. Development of machine learning models for the prediction of laminar flame speeds of hydrocarbon and oxygenated fuels10.1016/j.jfueco.2022.100071 ↗
16. Integrated machine learning-quantitative structure-property relationship (ML-QSPR) and chemical kinetics for high throughput fuel screening toward internal combustion engine10.1016/j.fuel.2021.121908 ↗
17. Ambrutis, A., and Povilaitis, M. (2021, January 24–28). Laminar burning velocity estimation using deep neural network. Proceedings of the 17th International Conference of Young Scientists on Energy and Natural Sciences Issues, Kaunas, Lithuania.
18. Malet, F. (2005). Numerical and Experimental Study of Premixed Turbulent Hydrogen Flame Propagation in Lean and Wet Atmosphere. [Ph.D. Thesis, Orleans University].
19. Heaton, J. (2022). Applications of Deep Neural Networks with Keras. arXiv.
20. On Mean Absolute Error for Deep Neural Network Based Vector-to-Vector Regression10.1109/LSP.2020.3016837 ↗
21. Root mean square error (RMSE) or mean absolute error (MAE)?10.5194/gmd-7-1247-2014 ↗
22. Laminar burning velocities of lean hydrogen-air mixtures at pressures up to 1.0 MPa10.1016/j.combustflame.2006.12.002 ↗
23. Effect of cylindrical confinement on the determination of laminar flame speeds using outwardly propagating flames10.1016/j.combustflame.2009.01.013 ↗
24. Laminar burning velocities of hydrogen-air mixtures from closed vessel gas explosions10.1016/j.jlp.2005.03.007 ↗
25. Laminar flame speeds of moist syngas mixtures10.1016/j.combustflame.2010.09.004 ↗
26. New insights into the peculiar behavior of laminar burning velocities of hydrogen-air flames according to pressure and equivalence ratio10.1016/j.combustflame.2014.02.009 ↗
27. An experimental and computational study of the burning rates of ultra-lean to moderately-rich H2/O2/N2 laminar flames with pressure variations10.1016/S0082-0784(06)80276-6 ↗
28. Experimental and numerical study on laminar burning velocities and flame instabilities of hydrogen-air mixtures at elevated pressures and temperatures10.1016/j.ijhydene.2009.08.044 ↗
29. Finding the Markstein number using the measurements of expanding spherical laminar flames10.1016/S0010-2180(96)00166-6 ↗
30. Burning velocities of hydrogen-air mixtures10.1016/0010-2180(93)90078-H ↗
31. Laminar Flame Speed and Ignition Delay Time Data for the Kinetic Modeling of Hydrogen and Syngas Fuel Blends10.1115/1.4007737 ↗
32. Flammability limits and laminar flame speed of hydrogen-air mixtures at sub-atmospheric pressures10.1016/j.ijhydene.2012.05.049 ↗
33. Flame/stretch interactions of premixed hydrogen-fueled flames: Measurements and predictions10.1016/S0010-2180(00)00229-7 ↗
34. Laminar flame velocity determination for H2-air-He-CO2 mixtures using the spherical bomb method10.1016/S0894-1777(02)00243-1 ↗
35. Alekseev, V. (2015). Laminar Burning Velocity of Hydrogen and Flame Structure of Related Fuels for Detailed Kinetic Model Validation. [Ph.D. Thesis, Lund University].
36. Measurements of the laminar burning velocity of hydrogen-air premixed flames10.1016/j.ijhydene.2009.12.031 ↗
37. Studies of premixed and non-premixed hydrogen flames10.1016/j.combustflame.2014.09.027 ↗
38. Hydrogen explosion in ITER: Effect of oxygen content on flame propagation of H2/O2/N2 mixtures10.1016/j.fusengdes.2013.02.128 ↗
39. Propagation characteristics of laminar spherical flames within homogeneous hydrogen-air mixtures10.1016/j.energy.2016.09.103 ↗
40. Taylor, S.C. (1991). Burning Velocity and the Influence of Flame Stretch. [Ph.D. Thesis, University of Leeds].
41. Morphology and burning rates of expanding spherical flames in H2/O2/inert mixtures up to 60 atmospheres10.1016/S0082-0784(00)80581-0 ↗
42. Further considerations on the determination of laminar flame speeds with the counterflow twin-flame technique10.1016/S0082-0784(06)80776-9 ↗
43. Determination of burning velocities from spherically expanding H2/air flames10.1016/j.proci.2014.05.137 ↗
44. Laminar and unstable burning velocities and Markstein lengths of hydrogen-air mixtures at engine-like conditions10.1016/j.proci.2004.07.042 ↗
45. On the determination of laminar flame speeds from stretched flames10.1016/S0082-0784(85)80693-7 ↗
46. Flame stretch interactions of laminar premixed hydrogen/air flames at normal temperature and pressure10.1016/S0010-2180(96)00151-4 ↗
47. Kuznetsov, M., Czerniak, M., Grune, J., and Jordan, T. (2013, January 9–11). Effect of Temperature on Laminar Flame Velocity for Hydrogen-Air Mixtures at Reduced Pressures. Proceedings of the International Conference on Hydrogen Safety, Progress in Safety of Hydrogen Technologies and Infrastructure: Enabling the Transition to Zero Carbon Energy, Brussels, Belgium.
48. (2022, June 29). RMSprop. Available online: Keras.io/api/optimizers/rmsprop/.
49. Taqi, A.M., Awad, A., Al-Azzo, F., and Milanova, M. (2018, January 10–12). The Impact of Multi-Optimizers and Data Augmentation on TensorFlow Convolutional Neural Network Performance. Proceedings of the IEEE Conference on Multimedia Information Processing and Retrieval (MIPR), Miami, FL, USA.10.1109/MIPR.2018.00032 ↗
50. Some sharp performance bounds for least squares regression with L1 regularization10.1214/08-AOS659 ↗
51. Panda, B., Sharma, S., and Roy, N. (2018). A Comparative Analysis of Various Regularization Techniques to Solve Overfitting Problem in Artificial Neural Network. Data Science and Analytics, Proceedings of the International Conference on Recent Developments in Science, Engineering and Technology, Gurgaon, India, 15–16 November 2017, Springer.
52. FlameFoam: An open-source CFD solver for turbulent premixed combustion10.1016/j.nucengdes.2021.111361 ↗
53. Bentaib, A., Chaumeix, N., Grosseuvres, R., Alexandre, B., Gastaldo, L., Ludovic, M., Jallais, S., Vyazmina, E., Kudriakov, S., and Studer, E. (2018, January 14–18). ETSON-MITHYGENE benchmark on simulations of upward flame propagation experiment in the ENACCEF2 experimental facility. Proceedings of the 12th International Topical Meeting on Nuclear Reactor Thermal-Hydraulics, Operation and Safety (NUTHOS-12), Qingdao, China.
54. Numerical studies of premixed hydrogen/air flames in a small-scale combustion chamber with varied area blockage ratio10.1016/j.ijhydene.2020.03.204 ↗