Tianhan Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Tianhan Zhang has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
Yiguang Ju
7 shared publications- Zheng Chen
Peking University
5 shared publications - Zhi-Qin John Xu
School of Mathematical Sciences Shanghai Jiao Tong University
5 shared publications - Weiqi Sun
Princeton University
3 shared publications Weizong Wang
School of Astronautics Beihang University 100191 Beijing China
3 shared publicationsYuxiao Yi
Institute of Natural Sciences School of Mathematical Sciences MOE-LSC Shanghai Jiao Tong University 200240 Shanghai China
3 shared publications
Latest Publications
- 29 July 2026
- Published by Elsevier in Proceedings of the Combustion Institute
- Vol. 42
Dynamics of flame initiation, propagation, and transition to detonation
- 29 July 2026
- Published by Elsevier in Proceedings of the Combustion Institute
- Vol. 42
Dynamics of flame initiation, propagation, and transition to detonation
- Open Access
An output scaling layer boosts deep neural networks for multiscale ODE systems
Simulating complex diffusion-reaction systems is often prohibitively expensive due to the high dimensionality and stiffness of the underlying ODEs, where state variables may span tens of orders of magnitude. Deep learning has recently emerged as a powerful tool in scientific computing, achieving remarkable progress in modeling and sampling stiff systems. However, data scaling techniques remain largely underexplored, despite their crucial role in addressing the frequency bias of deep neural networks when handling multi-magnitude or high-frequency data. In this work, we propose the Generalized Box-Cox Transformation (GBCT), a novel nonlinear scaling method designed to mitigate multiscale challenges by rescaling inherent multi-magnitude components toward a more consistent order of magnitude. We integrate GBCT into our previous data-driven framework and evaluate its performance against the original baseline surrogate model across six representative scenarios: a 21-species chemical reaction kinetics, a 13-isotope nuclear reaction model, the well-known Robertson problem coupled with diffusion, and practically relevant simulations of two-dimensional turbulent reaction-diffusion systems as well as one- and two-dimensional nuclear reactive flows. Numerical experiments demonstrate that GBCT reduces prediction errors by up to two orders of magnitude compared with the baseline model - particularly in the long-term evolution of dynamical systems - and achieves comparable performance with only about one-sixth of the training epochs. Frequency analysis further reveals that GBCT rescales high-frequency components of the objective function toward lower frequencies to align with the neural network's natural low frequency bias, thereby boosting training and generalization. The source code to reproduce the results in this paper is available at https://github.com/Seauagain/GBCT.
Open AccessAn output scaling layer boosts deep neural networks for multiscale ODE systems
Simulating complex diffusion-reaction systems is often prohibitively expensive due to the high dimensionality and stiffness of the underlying ODEs, where state variables may span tens of orders of magnitude. Deep learning has recently emerged as a powerful tool in scientific computing, achieving remarkable progress in modeling and sampling stiff systems. However, data scaling techniques remain largely underexplored, despite their crucial role in addressing the frequency bias of deep neural networks when handling multi-magnitude or high-frequency data. In this work, we propose the Generalized Box-Cox Transformation (GBCT), a novel nonlinear scaling method designed to mitigate multiscale challenges by rescaling inherent multi-magnitude components toward a more consistent order of magnitude. We integrate GBCT into our previous data-driven framework and evaluate its performance against the original baseline surrogate model across six representative scenarios: a 21-species chemical reaction kinetics, a 13-isotope nuclear reaction model, the well-known Robertson problem coupled with diffusion, and practically relevant simulations of two-dimensional turbulent reaction-diffusion systems as well as one- and two-dimensional nuclear reactive flows. Numerical experiments demonstrate that GBCT reduces prediction errors by up to two orders of magnitude compared with the baseline model - particularly in the long-term evolution of dynamical systems - and achieves comparable performance with only about one-sixth of the training epochs. Frequency analysis further reveals that GBCT rescales high-frequency components of the objective function toward lower frequencies to align with the neural network's natural low frequency bias, thereby boosting training and generalization. The source code to reproduce the results in this paper is available at https://github.com/Seauagain/GBCT.
- 22 October 2025
- Published by Springer Nature in Advances in Astronautics
- Vol. 8 (4) , 393-406
Comprehensive Machine Learning-Based Time-Series Anomaly Detection for ADN-Based Thruster
- 22 October 2025
- Published by Springer Nature in Advances in Astronautics
- Vol. 8 (4) , 393-406
Comprehensive Machine Learning-Based Time-Series Anomaly Detection for ADN-Based Thruster
- Open Access
- 9 October 2025
- Published by Elsevier in Applications in Energy and Combustion Science
- Vol. 24
DeePMO: An iterative deep learning framework for high-dimensional kinetic parameter optimization
- 9 October 2025
- Published by Elsevier in Applications in Energy and Combustion Science
- Vol. 24
Open AccessDeePMO: An iterative deep learning framework for high-dimensional kinetic parameter optimization
- 1 October 2025
- Published by Elsevier in Computer Physics Communications
- Vol. 318
Fire: An open-source adaptive mesh refinement solver for supersonic reacting flows
- 1 October 2025
- Published by Elsevier in Computer Physics Communications
- Vol. 318
Fire: An open-source adaptive mesh refinement solver for supersonic reacting flows
- Open Access
- 2 September 2025
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 990 (2) , 105
Deep Neural Networks for Modeling Astrophysical Nuclear Reacting Flows
In astrophysical simulations, nuclear reacting flows pose computational challenges due to the stiffness of reaction networks. We introduce neural network-based surrogate models using the DeePODE framework to enhance simulation efficiency while maintaining accuracy and robustness. Our method replaces conventional stiff ordinary differential equation (ODE) solvers with deep learning models trained through evolutionary Monte Carlo sampling from zero-dimensional simulation data, ensuring generalization across varied thermonuclear and hydrodynamic conditions. Tested on 3-species and 13-species reaction networks, the models achieve ≲1% accuracy relative to semi-implicit numerical solutions and deliver a ∼2.6× speedup on CPUs. A temperature-thresholded deployment strategy ensures stability in extreme conditions, sustaining neural network utilization above 75% in multidimensional simulations. These data-driven surrogates effectively mitigate stiffness constraints, offering a scalable approach for high-fidelity modeling of astrophysical nuclear reacting flows.
- 2 September 2025
- Published by American Astronomical Society in The Astrophysical Journal
- Vol. 990 (2) , 105
Open AccessDeep Neural Networks for Modeling Astrophysical Nuclear Reacting Flows
In astrophysical simulations, nuclear reacting flows pose computational challenges due to the stiffness of reaction networks. We introduce neural network-based surrogate models using the DeePODE framework to enhance simulation efficiency while maintaining accuracy and robustness. Our method replaces conventional stiff ordinary differential equation (ODE) solvers with deep learning models trained through evolutionary Monte Carlo sampling from zero-dimensional simulation data, ensuring generalization across varied thermonuclear and hydrodynamic conditions. Tested on 3-species and 13-species reaction networks, the models achieve ≲1% accuracy relative to semi-implicit numerical solutions and deliver a ∼2.6× speedup on CPUs. A temperature-thresholded deployment strategy ensures stability in extreme conditions, sustaining neural network utilization above 75% in multidimensional simulations. These data-driven surrogates effectively mitigate stiffness constraints, offering a scalable approach for high-fidelity modeling of astrophysical nuclear reacting flows.
- Top Cited Papers
- 10 March 2025
- Published by Elsevier in Combustion and Flame
- Vol. 275
Enforcing physical conservation in neural network surrogate models for complex chemical kinetics
- 10 March 2025
- Published by Elsevier in Combustion and Flame
- Vol. 275
Top Cited PapersEnforcing physical conservation in neural network surrogate models for complex chemical kinetics
- 30 December 2024
- Published by Elsevier in Combustion and Flame
- Vol. 273
Numerical investigation on flame dynamic and regime transitions during shock-cool flame interaction
- 30 December 2024
- Published by Elsevier in Combustion and Flame
- Vol. 273
Numerical investigation on flame dynamic and regime transitions during shock-cool flame interaction
- 4 January 2024
- Published by Elsevier in Combustion and Flame
- Vol. 261
Deep mechanism reduction (DeePMR) method for fuel chemical kinetics
- 4 January 2024
- Published by Elsevier in Combustion and Flame
- Vol. 261
Deep mechanism reduction (DeePMR) method for fuel chemical kinetics