Prediction of Mixing Uniformity of Hydrogen Injection inNatural Gas Pipeline Based on a Deep Learning Model
Abstract
:1. Introduction
2. Physical and Mathematical Models
2.1. Physical Model
2.2. Mathematical Model
3. Numerical Methods
3.1. Mesh Generation
3.2. CFD Solution Methods
4. Analysis of Different Factors
4.1. Influence of Gas Flow Velocity
4.2. Influence of Gas Temperature
4.3. Influence of Hydrogen Blending Ratio
4.4. Influence of Pipeline Diameter Ratio
5. COV Prediction Based on a Deep Neural Network Model
5.1. Establishment of DNN Model
- (1)
- Determination of hidden layer number and node number
- (2)
- Determination of activation function and learning rate
5.2. Results and Discussion
- (1)
- Prediction accuracy
- (2)
- Comparison of computational efficiency
6. Conclusions
- (1)
- When hydrogen is injected into the natural gas pipeline, the distribution of the mixed gases gradually becomes uniform with the increase in flow distance. The gas velocity is negatively correlated with the mixing uniformity, while the gas temperature, HBR, pipeline diameter ratio, and flow distance are positively correlated with the mixing uniformity. The hydrogen injection under low flow velocity, high gas temperature, and large branch pipeline diameter are recommended in engineering practice, and a sufficient flow distance should be ensured to fully mix the gas.
- (2)
- The input layer of the established DNN model contained five variables, namely flow velocity, HBR, gas temperature, pipeline diameter ratio, and flow distance. The output layer is the COV of the hydrogen concentration. The optimal hidden layer number, node number, activation function, and learning rate of the DNN model were determined by trial calculations. The accuracy of the DNN model was verified, and the average error for predicting the COV was 4.53%, which can meet the requirements of engineering practice well.
- (3)
- The computational efficiency for predicting the COV by the established DNN model and CFD simulations was compared in detail. The results indicate that the CPU time cost of the CFD simulations was far higher than that of the DNN model whether it was an online prediction or an offline calculation. The computational efficiency of the DNN model was at least two orders of magnitude faster than that of the CFD simulations for predicting the COV.
Author Contributions
Funding
Conflicts of Interest
References
- Zhang, H.; Tian, Z.G. Failure analysis of corroded high-strength pipeline subject to hydrogen damage based on FEM and GA-BP neural network. Int. J. Hydrogen Energ. 2022, 44, 4741–4758. [Google Scholar] [CrossRef]
- Mohtadi-Bonab, M.A. Effect of different parameters on hydrogen affected fatigue failure in pipeline steels. Eng. Fail. Anal. 2022, 137, 106262. [Google Scholar] [CrossRef]
- Chen, Z.-F.; Chu, W.-P.; Wang, H.-J.; Li, Y.; Wang, W.; Meng, W.-M.; Li, Y.-X. Structural integrity assessment of hydrogen-mixed natural gas pipelines based on a new multi-parameter failure criterion. Ocean Eng. 2022, 247, 110731. [Google Scholar] [CrossRef]
- Li, J.F.; Su, Y.; Zhang, H. Research progresses on pipeline transportation of hydrogen-blended natural gas. Nat. Gas Ind. 2021, 41, 137–152. [Google Scholar] [CrossRef]
- Li, J.; Su, Y.; Yu, B.; Wang, P.; Sun, D. Influences of Hydrogen Blending on the Joule–Thomson Coefficient of Natural Gas. ACS Omega 2021, 6, 1672216735. [Google Scholar] [CrossRef] [PubMed]
- Zhao, Y.; McDonell, V.; Samuelsen, S. Influence of hydrogen addition to pipeline natural gas on the combustion performance of a cooktop burner. Int. J. Hydrogen Energ. 2019, 44, 12239–12253. [Google Scholar] [CrossRef]
- Briottet, L.; Moro, I.; Lemoine, P. Quantifying the hydrogen embrittlement of pipeline steels for safety considerations. Int. J. Hydrogen Energ. 2012, 37, 17616–17623. [Google Scholar] [CrossRef]
- Bouledroua, O.; Zahreddine, H.; Milos, B.D. The synergistic effects of hydrogen embrittlement and transient gas flow conditions on integrity assessment of a precracked steel pipeline. Int. J. Hydrogen Energ. 2020, 45, 18010–18020. [Google Scholar] [CrossRef]
- Wang, G.P.; Guo, K.H.; Pan, G.J. Research on natural gas pipeline mixing-flow properties based on computational fluid dynamics. J. Therm. Sci. Technol. 2015, 14, 484–491. [Google Scholar]
- Yan, W.C.; Pei, Q.B.; Xia, B.D.; Shen, C.; Jiang, C.; Li, R. A method for determining the installation location of an on-line gas chromatograph in the mixed transportation pipelines with multiple sources. Natural Gas Ind. 2017, 37, 87–92. [Google Scholar] [CrossRef]
- Peng, S.N.; Jiang, X.C. Security analysis of natural gas mixed transmission system with multiple gas sources. Gas Heat 2014, 34, 1–3. [Google Scholar]
- Zhou, M.; Wiltschko, F.; Kulenovic, R.; Laurien, E. Large-eddy simulation on thermal-mixing experiment at horizontal T-junction with varied flow temperature. Nucl. Eng. Des. 2022, 388, 111644. [Google Scholar] [CrossRef]
- Su, B.; Zhu, Z.; Wang, X.; Ke, H.; Lin, M.; Wang, Q. Effect of temperature difference on the thermal mixing phenomenon in a T-junction under inflow pulsation. Nucl. Eng. Des. 2020, 363, 110611. [Google Scholar] [CrossRef]
- Zhang, Y.; Lu, T. Study of the quantitative assessment method for high-cycle thermal fatigue of a T-pipe under turbulent fluid mixing based on the coupled CFD-FEM method and the rain flow counting method. Nucl. Eng. Des. 2016, 309, 175–196. [Google Scholar] [CrossRef]
- Chen, M.-S.; Hsieh, H.-E.; Ferng, Y.-M.; Pei, B.-S. Experimental observations of thermal mixing characteristics in T-junction piping. Nucl. Eng. Des. 2014, 276, 107–114. [Google Scholar] [CrossRef]
- Eames, I.; Austin, M.; Wojcik, A. Injection of gaseous hydrogen into a natural gas pipeline. Int. J. Hydrogen Energy 2022, 47, 25745–25754. [Google Scholar] [CrossRef]
- Jegatheeswaran, S.; Ein-Mozaffari, F.; Wu, J.N. Process intensification in a chaotic SMX static mixer to achieve an energy-efficient mixing operation of non-newtonian fluids. Chem. Eng. Process 2018, 124, 1–10. [Google Scholar] [CrossRef]
- Liu, J.; Teng, L.; Liu, B.; Han, P.; Li, W. Analysis of hydrogen gas injection at various compositions in an existing natural gas pipeline. Front. Energy Res. 2021, 16, 685079. [Google Scholar] [CrossRef]
- Wahl, J.; Kallo, J. Quantitative valuation of hydrogen blending in European gas grids and its impact on the combustion process of large-bore gas engines. Int. J. Hydrogen Energy 2020, 45, 32534–32546. [Google Scholar] [CrossRef]
- Su, Y.; Li, J.; Yu, B.; Zhao, Y.; Yao, J. Fast and accurate prediction of failure pressure of oil and gas defective pipelines using the deep learning model. Reliab. Eng. Syst. Saf. 2021, 216, 108016. [Google Scholar] [CrossRef]
- Zheng, J.; Wang, C.; Liang, Y.; Liao, Q.; Li, Z.; Wang, B. Deeppipe: A deep-learning method for anomaly detection of multi-product pipelines. Energy 2022, 259, 125025. [Google Scholar] [CrossRef]
- Ahmad, S.; Ahmad, Z.; Kim, C.-H.; Kim, J.-M. A method for pipeline leak detection based on acoustic imaging and deep learning. Sensors 2022, 22, 1562. [Google Scholar] [CrossRef] [PubMed]
- Kim, J.; Chae, M.; Han, J.; Park, S.; Lee, Y. The development of leak detection model in subsea gas pipeline using machine learning. J. Nat. Gas. Sci. Eng. 2021, 94, 104134. [Google Scholar] [CrossRef]
- Fan, L.; Su, H.; Wang, W.; Zio, E.; Zhang, L.; Yang, Z.; Peng, S.; Yu, W.; Zuo, L.; Zhang, J. A systematic method for the optimization of gas supply reliability in natural gas pipeline network based on Bayesian networks and deep reinforcement learning. Reliab. Eng. Syst. Safe. 2022, 225, 108613. [Google Scholar] [CrossRef]
- Su, H.; Zio, E.; Zhang, J.J. A systematic hybrid method for real-time prediction of system conditions in natural gas pipeline networks. J. Nat. Gas. Sci. Eng. 2018, 57, 31–44. [Google Scholar] [CrossRef] [Green Version]
- Su, Y.; Li, J.F.; Yu, B. Numerical investigation on the leakage and diffusion characteristics of hydrogen-blended natural gas in a domestic kitchen. Renew. Energy 2022, 189, 899–916. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Bartzis, J.G. Turbulence modeling in the atmospheric boundary layer: A review and some recent developments. WIT Trans. Ecol. Environ. 2006, 86, 3–12. [Google Scholar]
- Choi, S.K.; Kim, S.O. Turbulence modeling of natural convection in enclosures: A review. Sci. Technol. 2012, 26, 283–297. [Google Scholar] [CrossRef]
- Zhuang, Z.; Yan, J.; Sun, C.; Wang, H.; Wang, Y.; Wu, Z. The numerical simulation of a new double swirl static mixer for gas reactants mixing. Chin. J. Chem. Eng. 2020, 28, 2438–2446. [Google Scholar] [CrossRef]
- Canziani, A.; Paszke, A.; Culurciello, E. An Analysis of Deep Neural Network Models for Practical Applications. arXiv 2018, arXiv:1605.07678. Chapter3, 37–73, 37–73. [Google Scholar] [CrossRef]
- Montavon, G.; Samek, W.; Muller, K.R. Methods for interpreting and understanding deep neural networks. Digit. Signal. Process. 2018, 73, 1–15. [Google Scholar] [CrossRef]
- Tanaka, M. Weighted sigmoid gate unit for an activation function of deep neural network. Pattern Recognit. Lett. 2020, 135, 354–359. [Google Scholar] [CrossRef]
- Wang, X.; Qin, Y. ReLTanh: An activation function with vanishing gradient resistance for SAE-based DNNs and its application to rotating machinery fault diagnosis. Neurocomputing 2019, 363, 88–98. [Google Scholar] [CrossRef]
- Grimstad, B.; Andersson, H. ReLU networks as surrogate models in mixed-integer linear programs. Comput. Chem. Eng. 2019, 131, 106580. [Google Scholar] [CrossRef] [Green Version]
- Mutasa, S.; Sun, S. Understanding artificial intelligence based radiology studies: What is overfitting? Clin. Imaging 2020, 65, 96–99. [Google Scholar] [CrossRef]
- Xue, Y. An overview of overfitting and its solutions. J. Phys. Conf. Ser. 2019, 1168, 022022. [Google Scholar]
Item | Settings in ANSYS Fluent |
---|---|
The turbulence model | Standard k-ε; full buoyancy effects |
Near-wall treatment | Standard wall functions |
Pressure–velocity coupling | SIMPLE |
The spatial discretization of gradient | Least squares cell-based |
The spatial discretization of pressure | Second order |
The spatial discretization of momentum | Second-order upwind |
The spatial discretization of energy | Second-order upwind |
The spatial discretization of turbulent dissipation rate | First-order upwind |
The spatial discretization of turbulent kinetic energy | First-order upwind |
No. | Grid Number | CPU Time (s) |
---|---|---|
1 | 2,403,598 | 18,654 |
2 | 2,403,598 | 18,570 |
3 | 2,403,598 | 19,245 |
4 | 2,205,874 | 16,357 |
5 | 2,804,592 | 20,548 |
6 | 2,804,592 | 21,576 |
7 | 2,403,598 | 19,354 |
8 | 2,584,256 | 18,946 |
Total | 153,250 (1.73 day) |
Method | CPU Time (s) | Acceleration |
---|---|---|
CFD | 153,250 | 1 |
DNN (offline + online) | 351 + 0.01 | 436 |
DNN (online) | =0.01 | =1.5 × 107 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Su, Y.; Li, J.; Guo, W.; Zhao, Y.; Li, J.; Zhao, J.; Wang, Y. Prediction of Mixing Uniformity of Hydrogen Injection inNatural Gas Pipeline Based on a Deep Learning Model. Energies 2022, 15, 8694. https://doi.org/10.3390/en15228694
Su Y, Li J, Guo W, Zhao Y, Li J, Zhao J, Wang Y. Prediction of Mixing Uniformity of Hydrogen Injection inNatural Gas Pipeline Based on a Deep Learning Model. Energies. 2022; 15(22):8694. https://doi.org/10.3390/en15228694
Chicago/Turabian StyleSu, Yue, Jingfa Li, Wangyi Guo, Yanlin Zhao, Jianli Li, Jie Zhao, and Yusheng Wang. 2022. "Prediction of Mixing Uniformity of Hydrogen Injection inNatural Gas Pipeline Based on a Deep Learning Model" Energies 15, no. 22: 8694. https://doi.org/10.3390/en15228694
APA StyleSu, Y., Li, J., Guo, W., Zhao, Y., Li, J., Zhao, J., & Wang, Y. (2022). Prediction of Mixing Uniformity of Hydrogen Injection inNatural Gas Pipeline Based on a Deep Learning Model. Energies, 15(22), 8694. https://doi.org/10.3390/en15228694