Konduri Aditya
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Konduri Aditya has a total of 78 co-authors. Here are the top co-authors by number of shared publications:
Diego A. Donzis
4 shared publicationsHemanth Kolla
Sandia National Laboratories Livermore California USA
4 shared publications- Hemanth Kolla
Sandia National Laboratories
3 shared publications Jacqueline H. Chen
3 shared publications- Akash Rodhiya2 shared publications
- Andrea Ciani
Ansaldo Energia Switzerland AG , Baden 5400, Switzerland
2 shared publications
Latest Publications
- 16 June 2026
- Published by Elsevier in Computers & Fluids
- Vol. 317
Accuracy and scalability of asynchronous flow solver for transitional flows
- 16 June 2026
- Published by Elsevier in Computers & Fluids
- Vol. 317
Accuracy and scalability of asynchronous flow solver for transitional flows
- 17 April 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 22 (9) , 4247-4258
ChemXDyn: Dynamics-Informed Species and Reaction Detection Methodology from Atomistic Simulations
Accurate identification of chemical species and reaction pathways from molecular dynamics (MD) trajectories is a prerequisite for deriving predictive chemical kinetic models and for mechanistic discovery in reactive systems. However, state-of-the-art trajectory analysis methods infer bonding from instantaneous distance thresholds, which can misclassify transient, nonreactive encounters as bonds and thereby introduce spurious intermediates, distorted reaction networks, and biased rate estimates. Here, we introduce ChemXDyn, a dynamics-aware computational methodology that leverages time-resolved interatomic distance (IAD) signatures as a core principle to robustly identify chemically consistent bonded interactions and, consequently, extract meaningful reaction pathways. In particular, ChemXDyn propagates molecular connectivity through time while enforcing atomic valence and coordination constraints to distinguish genuine bond-breaking and bond-forming events from transient, nonreactive encounters. We evaluate ChemXDyn on ReaxFF MD simulations of hydrogen and ammonia oxidation and on neural-network potential MD simulations of methane oxidation and benchmark its performance against widely used trajectory analysis methods. Across these cases, ChemXDyn suppresses unphysical species prevalent in static analyses, recovers experimentally consistent reaction pathways, and improves the fidelity of the rate constant estimation. In ammonia oxidation, ChemXDyn removes unphysical intermediates (including N3O, N3O, N4O2, and HN2O2) and resolves key NOx- and N2O-forming and -consuming routes (for example, NH2 + HO2 → H2NO + OH and N2O + H → N2 + OH). In methane oxidation, it reconstructs the canonical progression CH4 → CH3 → CH2 → CH → CHO/CH2O → CO → CO2, which is consistent with established mechanisms yet is often fragmented by threshold-based approaches. By linking atomistic dynamics to chemically consistent reaction identification, ChemXDyn provides a transferable foundation for MD-derived reaction networks and kinetics, with potential utility spanning combustion, heterogeneous catalysis, plasma chemistry, and electrochemical reaction environments.
- 17 April 2026
- Published by American Chemical Society (ACS) in Journal of Chemical Theory and Computation
- Vol. 22 (9) , 4247-4258
ChemXDyn: Dynamics-Informed Species and Reaction Detection Methodology from Atomistic Simulations
Accurate identification of chemical species and reaction pathways from molecular dynamics (MD) trajectories is a prerequisite for deriving predictive chemical kinetic models and for mechanistic discovery in reactive systems. However, state-of-the-art trajectory analysis methods infer bonding from instantaneous distance thresholds, which can misclassify transient, nonreactive encounters as bonds and thereby introduce spurious intermediates, distorted reaction networks, and biased rate estimates. Here, we introduce ChemXDyn, a dynamics-aware computational methodology that leverages time-resolved interatomic distance (IAD) signatures as a core principle to robustly identify chemically consistent bonded interactions and, consequently, extract meaningful reaction pathways. In particular, ChemXDyn propagates molecular connectivity through time while enforcing atomic valence and coordination constraints to distinguish genuine bond-breaking and bond-forming events from transient, nonreactive encounters. We evaluate ChemXDyn on ReaxFF MD simulations of hydrogen and ammonia oxidation and on neural-network potential MD simulations of methane oxidation and benchmark its performance against widely used trajectory analysis methods. Across these cases, ChemXDyn suppresses unphysical species prevalent in static analyses, recovers experimentally consistent reaction pathways, and improves the fidelity of the rate constant estimation. In ammonia oxidation, ChemXDyn removes unphysical intermediates (including N3O, N3O, N4O2, and HN2O2) and resolves key NOx- and N2O-forming and -consuming routes (for example, NH2 + HO2 → H2NO + OH and N2O + H → N2 + OH). In methane oxidation, it reconstructs the canonical progression CH4 → CH3 → CH2 → CH → CHO/CH2O → CO → CO2, which is consistent with established mechanisms yet is often fragmented by threshold-based approaches. By linking atomistic dynamics to chemically consistent reaction identification, ChemXDyn provides a transferable foundation for MD-derived reaction networks and kinetics, with potential utility spanning combustion, heterogeneous catalysis, plasma chemistry, and electrochemical reaction environments.
- Open Access
ChemXDyn: Dynamics-informed species and reaction detection methodology from atomistic simulations
Accurate identification of chemical species and reaction pathways from molecular dynamics (MD) trajectories is a prerequisite for deriving predictive chemical-kinetic models and for mechanistic discovery in reactive systems. However, state-of-the-art trajectory analysis methods infer bonding from instantaneous distance thresholds, which can misclassify transient, nonreactive encounters as bonds and thereby introduce spurious intermediates, distorted reaction networks, and biased rate estimates. Here, we introduce ChemXDyn, a dynamics-aware computational methodology that leverages time-resolved interatomic distance signatures as a core principle to robustly identify chemically consistent bonded interactions and, consequently, extract meaningful reaction pathways. In particular, ChemXDyn propagates molecular connectivity through time while enforcing atomic valence and coordination constraints to distinguish genuine bond-breaking and bond-forming events from transient, nonreactive encounters. We evaluate ChemXDyn on ReaxFF MD simulations of hydrogen and ammonia oxidation and on neural-network potential MD simulations of methane oxidation, and benchmark its performance against widely used trajectory analysis methods. Across these cases, ChemXDyn suppresses unphysical species prevalent in static analyses, recovers experimentally consistent reaction pathways, and improves the fidelity of rate constant estimation. In ammonia oxidation, ChemXDyn removes unphysical intermediates and resolves key NOx- and N2O-forming and -consuming routes. In methane oxidation, it reconstructs the canonical progression from CH4 to CO2. By linking atomistic dynamics to chemically consistent reaction identification, ChemXDyn provides a transferable foundation for MD-derived reaction networks and kinetics, with potential utility spanning combustion, catalysis, plasma chemistry, and electrochemical environments.
Open AccessChemXDyn: Dynamics-informed species and reaction detection methodology from atomistic simulations
Accurate identification of chemical species and reaction pathways from molecular dynamics (MD) trajectories is a prerequisite for deriving predictive chemical-kinetic models and for mechanistic discovery in reactive systems. However, state-of-the-art trajectory analysis methods infer bonding from instantaneous distance thresholds, which can misclassify transient, nonreactive encounters as bonds and thereby introduce spurious intermediates, distorted reaction networks, and biased rate estimates. Here, we introduce ChemXDyn, a dynamics-aware computational methodology that leverages time-resolved interatomic distance signatures as a core principle to robustly identify chemically consistent bonded interactions and, consequently, extract meaningful reaction pathways. In particular, ChemXDyn propagates molecular connectivity through time while enforcing atomic valence and coordination constraints to distinguish genuine bond-breaking and bond-forming events from transient, nonreactive encounters. We evaluate ChemXDyn on ReaxFF MD simulations of hydrogen and ammonia oxidation and on neural-network potential MD simulations of methane oxidation, and benchmark its performance against widely used trajectory analysis methods. Across these cases, ChemXDyn suppresses unphysical species prevalent in static analyses, recovers experimentally consistent reaction pathways, and improves the fidelity of rate constant estimation. In ammonia oxidation, ChemXDyn removes unphysical intermediates and resolves key NOx- and N2O-forming and -consuming routes. In methane oxidation, it reconstructs the canonical progression from CH4 to CO2. By linking atomistic dynamics to chemically consistent reaction identification, ChemXDyn provides a transferable foundation for MD-derived reaction networks and kinetics, with potential utility spanning combustion, catalysis, plasma chemistry, and electrochemical environments.
- 5 May 2025
- Published by American Physical Society (APS) in Physical Review Fluids
- Vol. 10 (5)
Probing quasigeostrophic turbulence via complex networks
- 5 May 2025
- Published by American Physical Society (APS) in Physical Review Fluids
- Vol. 10 (5)
Probing quasigeostrophic turbulence via complex networks
- 20 December 2024
- Published by ASME International in Journal of Engineering for Gas Turbines and Power
- Vol. 147 (6) , 1-12
Flashback Prevention in a Hydrogen-Fueled Reheat Combustor by Water Injection Optimized With Global Sensitivity Analysis
This paper investigates water injection effects in a simplified Ansaldo GT36 reheat system under realistic conditions of 20 atm using large eddy simulation (LES) coupled with thickened flame modeling and adaptive mesh refinement. The water injection conditions are optimized by performing a parametric study based on global sensitivity analysis (GSA) with a surrogate model based on Gaussian process (GP) to reduce computational cost. In particular, the influence of four design parameters, namely, Sauter mean diameter (SMD), water mass flow, and the angles of the spray's hollow cone, is tested to achieve an optimized solution. In the “dry” case, the LES simulations show several flashback events attributed to compressive pressure waves resulting from auto-ignition in the core flow near the crossover temperature. The use of water injection is found to be effective in suppressing the flashback occurrence. In particular, the global sensitivity analysis shows that the external angle of the spray cone and the mass flow of water are the most important design parameters for flashback prevention. NOx emissions are reduced by about 17% with water injection. Once an optimized condition with water injection is found, a recently proposed method to downscale the combustor to lower pressures is applied and tested. Additional LESs are performed for this purpose at the dry, unstable condition and the “wet,” stable condition. Results show that similar dynamics are predicted at 1 atm, validating the method's robustness. This provides avenues for experimentally testing combustion dynamics at simplified conditions which are still representative of high-pressure practical configurations.
- 20 December 2024
- Published by ASME International in Journal of Engineering for Gas Turbines and Power
- Vol. 147 (6) , 1-12
Flashback Prevention in a Hydrogen-Fueled Reheat Combustor by Water Injection Optimized With Global Sensitivity Analysis
This paper investigates water injection effects in a simplified Ansaldo GT36 reheat system under realistic conditions of 20 atm using large eddy simulation (LES) coupled with thickened flame modeling and adaptive mesh refinement. The water injection conditions are optimized by performing a parametric study based on global sensitivity analysis (GSA) with a surrogate model based on Gaussian process (GP) to reduce computational cost. In particular, the influence of four design parameters, namely, Sauter mean diameter (SMD), water mass flow, and the angles of the spray's hollow cone, is tested to achieve an optimized solution. In the “dry” case, the LES simulations show several flashback events attributed to compressive pressure waves resulting from auto-ignition in the core flow near the crossover temperature. The use of water injection is found to be effective in suppressing the flashback occurrence. In particular, the global sensitivity analysis shows that the external angle of the spray cone and the mass flow of water are the most important design parameters for flashback prevention. NOx emissions are reduced by about 17% with water injection. Once an optimized condition with water injection is found, a recently proposed method to downscale the combustor to lower pressures is applied and tested. Additional LESs are performed for this purpose at the dry, unstable condition and the “wet,” stable condition. Results show that similar dynamics are predicted at 1 atm, validating the method's robustness. This provides avenues for experimentally testing combustion dynamics at simplified conditions which are still representative of high-pressure practical configurations.
- 5 September 2024
- Published by ASME International in Journal of Engineering for Gas Turbines and Power
- Vol. 146 (12) , 1-17
A Reactor-Network Framework to Model Performance and Emissions of a Longitudinally Staged Combustion System for Carbon-Free Fuels
Hydrogen and ammonia are considered crucial carbon-free energy carriers optimally suited for seasonal chemical storage and balancing of the energy system. In this context, longitudinally staged combustion systems represent an attractive technology in power generation for their capability of achieving low NOx emissions while conserving high load and, crucially, fuel flexibility at high thermal efficiency. Such two-stage combustion systems have been successfully implemented for natural gas firing of gas turbines and, more recently, have shown significant potential for clean and efficient hydrogen-firing operation. However, optimal operation with ammonia-based fuel mixtures is yet to be established. In recent works, a novel Rich-Quench-Lean (RQL) operational concept was proposed to burn a fuel-rich mixture of partially decomposed ammonia and air (for equivalence ratios ϕ∼1.1−1.2) in the first stage of a longitudinally staged combustion system. Complete oxidation of the remaining (hydrogen) fuel is theoretically ensured by dilution-air addition downstream of the first stage combustor. However, any operational concept based on these near-stoichiometric combustion conditions, while minimizing undesired prompt NOx and N2O formation by ammonia oxidation, can potentially result in significant, and certainly unpractical, thermal load on the first stage combustor liner that needs to be mitigated. In the present study, we exploit a newly developed reactors-network model to efficiently investigate the NOx-emissions performance of a longitudinally staged combustion system fired with natural gas, hydrogen or ammonia. First, the reactors network framework is validated with experimental, computational and other similar reactor network results in the literature. Second, the optimal air distribution within the longitudinally staged combustion system is found for clean (low emissions) and efficient (complete fuel conversion) ammonia-firing operation. Third, the consequences of such “ammonia-optimized” air distribution on flame stabilization and NOx emissions in more conventional natural gas- and hydrogen-firing operation are considered. Finally, an optimal air and fuel distribution is suggested for the longitudinally staged combustion system on the basis that, while still ensuring robust flame stabilization and high turbine inlet temperature, it minimizes NOx emissions for all three fuels considered.
- 5 September 2024
- Published by ASME International in Journal of Engineering for Gas Turbines and Power
- Vol. 146 (12) , 1-17
A Reactor-Network Framework to Model Performance and Emissions of a Longitudinally Staged Combustion System for Carbon-Free Fuels
Hydrogen and ammonia are considered crucial carbon-free energy carriers optimally suited for seasonal chemical storage and balancing of the energy system. In this context, longitudinally staged combustion systems represent an attractive technology in power generation for their capability of achieving low NOx emissions while conserving high load and, crucially, fuel flexibility at high thermal efficiency. Such two-stage combustion systems have been successfully implemented for natural gas firing of gas turbines and, more recently, have shown significant potential for clean and efficient hydrogen-firing operation. However, optimal operation with ammonia-based fuel mixtures is yet to be established. In recent works, a novel Rich-Quench-Lean (RQL) operational concept was proposed to burn a fuel-rich mixture of partially decomposed ammonia and air (for equivalence ratios ϕ∼1.1−1.2) in the first stage of a longitudinally staged combustion system. Complete oxidation of the remaining (hydrogen) fuel is theoretically ensured by dilution-air addition downstream of the first stage combustor. However, any operational concept based on these near-stoichiometric combustion conditions, while minimizing undesired prompt NOx and N2O formation by ammonia oxidation, can potentially result in significant, and certainly unpractical, thermal load on the first stage combustor liner that needs to be mitigated. In the present study, we exploit a newly developed reactors-network model to efficiently investigate the NOx-emissions performance of a longitudinally staged combustion system fired with natural gas, hydrogen or ammonia. First, the reactors network framework is validated with experimental, computational and other similar reactor network results in the literature. Second, the optimal air distribution within the longitudinally staged combustion system is found for clean (low emissions) and efficient (complete fuel conversion) ammonia-firing operation. Third, the consequences of such “ammonia-optimized” air distribution on flame stabilization and NOx emissions in more conventional natural gas- and hydrogen-firing operation are considered. Finally, an optimal air and fuel distribution is suggested for the longitudinally staged combustion system on the basis that, while still ensuring robust flame stabilization and high turbine inlet temperature, it minimizes NOx emissions for all three fuels considered.
- Open Access
- 26 July 2024
- Published by Elsevier in Proceedings of the Combustion Institute
- Vol. 40 (1-4)
Data-driven identification of precursors of flashback in a lean hydrogen reheat combustor
- 26 July 2024
- Published by Elsevier in Proceedings of the Combustion Institute
- Vol. 40 (1-4)
Open AccessData-driven identification of precursors of flashback in a lean hydrogen reheat combustor
- 19 July 2024
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 430
An asynchronous discontinuous Galerkin method for massively parallel PDE solvers
- 19 July 2024
- Published by Elsevier in Computer Methods in Applied Mechanics and Engineering
- Vol. 430
An asynchronous discontinuous Galerkin method for massively parallel PDE solvers
- 1 July 2024
- Published by AIP Publishing in Physics of Fluids
- Vol. 36 (7)
A scale-similarity based workflow for self-supervised reconstruction of small-scales in turbulence using deep learning
Deep learning has been extensively utilized for modeling and analysis of fluid turbulence. One such application is the use of super-resolution (SR) algorithms to reconstruct small-scale structures from their large-scale counterparts for turbulent flows. To date, all SR algorithms have been supervised or require unpaired reference data at a high resolution for training. This renders the model inapplicable to practical fluid flow scenarios, in which the generation of a high-resolution ground truth by resolving all scales down to the Kolmogorov scale becomes prohibitive. Hence, it is imperative to develop physics-guided models that exploit the multiscale nature of turbulence. Considering SR as a state-estimation problem, we present a self-supervised workflow based on deep neural networks to reconstruct small-scale structures that are relevant to homogeneous isotropic turbulence. In addition to visual similarity, we assessed the quality of the obtained reconstruction using spectra, structure functions, and probability density functions of the gradients of velocity and a passive scalar. From the analysis, we infer that the outputs of the workflow are in statistical agreement with the ground truth, for which the training pipeline is agnostic. Insights into learnability, interpretability, and generality of the trained networks have been provided as well. The results of this study can be leveraged to devise techniques for the reconstruction of small-scale structures using large-eddy simulation data.
- 1 July 2024
- Published by AIP Publishing in Physics of Fluids
- Vol. 36 (7)
A scale-similarity based workflow for self-supervised reconstruction of small-scales in turbulence using deep learning
Deep learning has been extensively utilized for modeling and analysis of fluid turbulence. One such application is the use of super-resolution (SR) algorithms to reconstruct small-scale structures from their large-scale counterparts for turbulent flows. To date, all SR algorithms have been supervised or require unpaired reference data at a high resolution for training. This renders the model inapplicable to practical fluid flow scenarios, in which the generation of a high-resolution ground truth by resolving all scales down to the Kolmogorov scale becomes prohibitive. Hence, it is imperative to develop physics-guided models that exploit the multiscale nature of turbulence. Considering SR as a state-estimation problem, we present a self-supervised workflow based on deep neural networks to reconstruct small-scale structures that are relevant to homogeneous isotropic turbulence. In addition to visual similarity, we assessed the quality of the obtained reconstruction using spectra, structure functions, and probability density functions of the gradients of velocity and a passive scalar. From the analysis, we infer that the outputs of the workflow are in statistical agreement with the ground truth, for which the training pipeline is agnostic. Insights into learnability, interpretability, and generality of the trained networks have been provided as well. The results of this study can be leveraged to devise techniques for the reconstruction of small-scale structures using large-eddy simulation data.