Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction
Propagating uncertainties in kinetic models through combustion simulations can provide important metrics on the reliability and accuracy of a model, but remains a challenging and numerically expensive problem especially for large kinetic mechanisms and expensive turbulent combustion simulations. Var...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Published: |
Massachusetts Institute of Technology
2023
|
| Online Access: | https://hdl.handle.net/1721.1/151863 https://orcid.org/0000-0002-5733-0807 |
| _version_ | 1826206105801326592 |
|---|---|
| author | Koenig, Benjamin C. |
| author2 | Deng, Sili |
| author_facet | Deng, Sili Koenig, Benjamin C. |
| author_sort | Koenig, Benjamin C. |
| collection | MIT |
| description | Propagating uncertainties in kinetic models through combustion simulations can provide important metrics on the reliability and accuracy of a model, but remains a challenging and numerically expensive problem especially for large kinetic mechanisms and expensive turbulent combustion simulations. Various surrogate model and dimension reduction techniques have previously been applied in order to reduce the cost of forward uncertainty propagation in combustion simulations, but these are often limited to low-dimensional, simple combustion cases with scalar solution targets. In the current work, a neural network-accelerated framework for identifying a low-dimensional active kinetic subspace was developed that applies to the entire temperature solution space of a flamelet table and can capture the mixture fraction and strain rate dependent effects of the kinetic uncertainty. The computational savings enabled by this novel framework were demonstrated through a proof-of-concept, flamelet-based application in a Reynolds-averaged Sandia Flame D simulation using a chemical mechanism for methane combustion with 217 reactions. By leveraging the large dimensional compression and low-cost scaling of the active subspace method, offloading the initial dimension reduction gradient sampling onto the laminar flamelet simulations, and accelerating the gradient sampling process with a specifically designed neural network, it was possible to estimate the temperature uncertainty profiles across the solution space of the turbulent flame with strong accuracy of 70-85% using just seven perturbed solutions. Additionally, as it occurs entirely within the flamelet table, the cost of identifying the reduced subspace does not scale with the cost of the turbulent combustion model, which is a promising feature of this framework for future application to larger-scale and more complex turbulent combustion applications. |
| first_indexed | 2024-09-23T13:24:11Z |
| format | Thesis |
| id | mit-1721.1/151863 |
| institution | Massachusetts Institute of Technology |
| last_indexed | 2024-09-23T13:24:11Z |
| publishDate | 2023 |
| publisher | Massachusetts Institute of Technology |
| record_format | dspace |
| spelling | mit-1721.1/1518632023-08-24T03:25:53Z Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction Koenig, Benjamin C. Deng, Sili Massachusetts Institute of Technology. Department of Mechanical Engineering Propagating uncertainties in kinetic models through combustion simulations can provide important metrics on the reliability and accuracy of a model, but remains a challenging and numerically expensive problem especially for large kinetic mechanisms and expensive turbulent combustion simulations. Various surrogate model and dimension reduction techniques have previously been applied in order to reduce the cost of forward uncertainty propagation in combustion simulations, but these are often limited to low-dimensional, simple combustion cases with scalar solution targets. In the current work, a neural network-accelerated framework for identifying a low-dimensional active kinetic subspace was developed that applies to the entire temperature solution space of a flamelet table and can capture the mixture fraction and strain rate dependent effects of the kinetic uncertainty. The computational savings enabled by this novel framework were demonstrated through a proof-of-concept, flamelet-based application in a Reynolds-averaged Sandia Flame D simulation using a chemical mechanism for methane combustion with 217 reactions. By leveraging the large dimensional compression and low-cost scaling of the active subspace method, offloading the initial dimension reduction gradient sampling onto the laminar flamelet simulations, and accelerating the gradient sampling process with a specifically designed neural network, it was possible to estimate the temperature uncertainty profiles across the solution space of the turbulent flame with strong accuracy of 70-85% using just seven perturbed solutions. Additionally, as it occurs entirely within the flamelet table, the cost of identifying the reduced subspace does not scale with the cost of the turbulent combustion model, which is a promising feature of this framework for future application to larger-scale and more complex turbulent combustion applications. S.M. 2023-08-23T16:14:23Z 2023-08-23T16:14:23Z 2023-06 2023-07-19T18:45:21.338Z Thesis https://hdl.handle.net/1721.1/151863 https://orcid.org/0000-0002-5733-0807 In Copyright - Educational Use Permitted Copyright retained by author(s) https://rightsstatements.org/page/InC-EDU/1.0/ application/pdf Massachusetts Institute of Technology |
| spellingShingle | Koenig, Benjamin C. Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction |
| title | Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction |
| title_full | Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction |
| title_fullStr | Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction |
| title_full_unstemmed | Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction |
| title_short | Enabling Efficient Uncertainty Quantification of Turbulent Combustion Simulations via Kinetic Dimension Reduction |
| title_sort | enabling efficient uncertainty quantification of turbulent combustion simulations via kinetic dimension reduction |
| url | https://hdl.handle.net/1721.1/151863 https://orcid.org/0000-0002-5733-0807 |
| work_keys_str_mv | AT koenigbenjaminc enablingefficientuncertaintyquantificationofturbulentcombustionsimulationsviakineticdimensionreduction |