About

Professor Datta Gaitonde is a faculty member in the Department of Mechanical and Aerospace Engineering at The Ohio State University. He holds the John Glenn Chair for Technology and Space Exploration and is associated with the Aerospace Research Center. His research focuses on high-fidelity computational multi-physics, aerospace flow control, gas turbine engines, hypersonics, uncrewed aircraft systems, and automotive aeroacoustics. He is involved in various laboratories and research groups dedicated to aerodynamics, flow physics, turbulence, thermodynamics, and complex systems, contributing to advancements in aerospace technology and engineering.

Research topics

• Physics
• Mechanics
• Engineering
• Structural engineering
• Classical mechanics
• Mathematical analysis
• Materials science
• Geometry
• Mathematics

Selected publications

• Benchmarking Long Roll-outs of Auto-regressive Neural Operators for the Compressible Navier-Stokes Equations with Conserved Quantity Correction

  Open MIND · 2026-01-30

  preprint

  Deep learning has been proposed as an efficient alternative for the numerical approximation of PDE solutions, offering fast, iterative simulation of PDEs through the approximation of solution operators. However, deep learning solutions have struggle to perform well over long prediction durations due to the accumulation of auto-regressive error, which is compounded by the inability of models to conserve physical quantities. In this work, we present conserved quantity correction, a model-agnostic technique for incorporation physical conservation criteria within deep learning models. Our results demonstrate consistent improvement in the long-term stability of auto-regressive neural operator models, regardless of the model architecture. Furthermore, we analyze the performance of neural operators from the spectral domain, highlighting significant limitations of present architectures. These results highlight the need for future work to consider architectures that place specific emphasis on high frequency components, which are integral to the understanding and modeling of turbulent flows.

  DOI

• Elucidating Three-Dimensional Coherent Structures in a Multi-Stream Jet

  ArXiv.org · 2026-01-22

  articleOpen access

  Nominal two-dimensional (2D) shear layers have been studied extensively, and their principal dynamics are well understood. In practical configurations, however, the behavior of such shear layers is affected by proximal surfaces. In this study, we investigate three-dimensional (3D) coherent structures developing downstream of a relatively thick splitter plate in a realistic nozzle featuring sidewalls, an upper boundary formed by a single expansion ramp, and a lower boundary defined by a protruding deck. As a result, in addition to the primary splitter plate shear layer (SPSL) arising from mixing between the core and bypass streams, the flow contains upper (USL) and lower (LSL) shear layers with the ambient. Large-eddy simulation data are analyzed to characterize the unsteady flow dynamics, while the mean flow provides insight into the underlying amplification mechanisms. Spectral proper orthogonal decomposition reveals a clear separation of broadband and tonal dynamics across frequency bands. The broadband low-frequency modes are highly 3D and originate in the USL and LSL. In contrast, tonal high-frequency content is associated with a 2D instability in the SPSL. Both the broadband and tonal signatures also appear in the nonlinear energy transfer mechanisms. Triglobal resolvent analysis further clarifies the amplification mechanisms within the USL and LSL. Low-frequency response modes are excited by forcing localized near the nozzle geometry and are governed by 3D Kelvin-Helmholtz dynamics. The low-frequency streamwise vortices generated at the nozzle corners drive the axis-switching behavior characteristic of rectangular jets. Wavemaker analysis further demonstrates that these corner vortices are part of self-sustaining low-frequency dynamics.

  PublisherOA PDF

• Effect of Surface Panel Motion on Turbulent Compound Shock/Boundary-Layer Interactions

  2026-01-08

  article

  Fluid–structure interactions (FSI) induced by shock-wave boundary layer interactions (SBLI) play a critical role in shaping the aerodynamic performance and structural integrity of high-speed vehicles. The dynamics of turbulent SBLIs can be substantially altered by surface deformations, depending on the receptivity of the flow to induced perturbations. This study investigates the response of a highly three-dimensional SBLI generated by a canonical double-fin configuration using Large-Eddy simulations, which abstracts multiple complex interacting features akin to realistic flight scenarios. The baseline undisturbed turbulent flowfield is analyzed to identify dominant spatio-temporal features, which are then used to guide the placement and actuation of a compliant panel. Two representative panel mode shapes, Mode (1,1) and Mode (2,1), are prescribed at a forcing frequency aligned with the separated shear-layer dynamics of the baseline flow, in an efficient and appropriate one-way structure-to-fluid analysis framework, wherein the feedback to the structure is muted. The resulting surface motion alters local viscous–inviscid interactions, enhances mean flow separation, and expands the near-wall subsonic region. The flow modifications persist downstream of the panel, leading to discernible changes in surface pressure. The imposed panel forcing intensifies shear-layer flapping and modulates near-wall turbulence by prematurely disrupting the streak cycle and delaying the downstream recovery of boundary-layer anisotropy. The shock unsteadiness is largely dictated by the panel dynamics, with its exact characteristics dependent on the specific mode shape. A Lagrangian modal analysis-based implementation suited for moving meshes applied to Bispectral Mode Decomposition reveals nonlinear triadic interactions facilitating energy transfer across the fundamental forcing frequency and its harmonics, establishing a modal feedback cascade. Additionally, a time-localized phase-based analysis is proposed, showing significant phase-locking between the panel and shock motion, with transient phase “slipping” giving rise to stable preferred phase differences. This highlights the phase synchrony in the complex forced interaction, where the stable phase difference values depend on the forcing mode shape.

  PublisherDOI

• Effect of Fins on the Mean Wake of a Generic Supersonic Projectile

  Journal of Spacecraft and Rockets · 2026-03-29

  articleSenior author

  A numerical study of a generic projectile with a solid base at Mach 2 is used to highlight the impact of fins on the wake across a range of roll and pitch angles from zero to 12 deg. To accommodate this wide range of parameters, with and without fins, the Reynolds-averaged Navier–Stokes equations are solved with a realizable [Formula: see text] turbulence model and validated against available experimental data. The effects of fins on the wake are characterized by a detailed analysis of the emergent shock and vortical structures, as well as the consequent aerodynamic loads. At a zero angle of attack, the fin trailing edge introduces new shocks/expansions that fundamentally alter the shape and strength of the recompression shock. Pitch and roll further modify the shocks and vortices, which can now be distinguished as forebody, wingtip, and trailing-edge components. At certain pitch and roll conditions, wingtip vortices bound and split the recompression shock into multiple sections, changing their inception and subsequent trajectory. Under pitch conditions, fins effectively flip the trailing-edge vortices, which now form on the lower side of the base surface rather than the upper side, with a reversal of streamwise vorticity direction. The results identify the main features of interest for further exploration with scale-resolving simulations.

  PublisherDOI

• Three-Dimensional Unsteadiness of a Shock Train in a Low Aspect Ratio Rectangular Duct

  2026-01-08

  articleSenior author

  The unsteady dynamics of a shock train in a low aspect ratio, rectangular duct are investigated using large eddy simulations. Flow conditions are defined by the incoming Mach number (M∞ = 2.0) and the ratio of mean outlet to inlet pressures (pb/pi = 2.5), whereas geometric conditions are defined by the aspect ratio of duct width to height (w:h = 0.82). Significant differences between the statistics of the lateral (side) and vertical (top/bottom) walls are noted, with the side wall experiencing elevated pressure fluctuations. These differences are explained by examining the band isolated dynamics of side and bottom wall pressure fields and the oscillatory motion of the shock feet interacting with the boundary layers lining each wall. Low-frequency shock motions manifest as an essentially rigid body motion of the entire shock column, whose signature manifests as low-frequency pressure oscillations on all duct walls. Mid-band dynamics are dominated by tonal shock oscillations, driven by a combined hydrodynamic-acoustic feedback loop between adjacent shock-boundary layer interactions within the train. These tones, characterized by a dominant fundamental frequency and its first two harmonics, leave a significantly larger pressure signature on the side wall than the bottom wall. Large differences in pressure variance between the lateral and vertical walls are attributed to this phenomenon. However, high-frequency dynamics, such as those associated with eddying motions of the turbulent boundary layer and associate local shock responses, remain similar between the two walls.

  PublisherDOI

• Effects of Distributed-Roughness on Hypersonic Boundary Layer Instabilities

  2026-01-08

  articleSenior author

  Instabilities in boundary layers over the surfaces of vehicles often drive the transition from laminar to turbulent states, significantly impacting their aerodynamic performance. Surface texture alters boundary layer profiles and thus the instability mechanisms leading to the transition. We examine the effects of two-dimensional (2D, spanwise homogeneous) roughness patches on stability characteristics and transition processes in hypersonic flow by comparing the disturbance evolution with a benchmark flat plate configuration. Two distinct roughness heights are considered, corresponding to 1/6th and 1/3rd of the incoming boundary layer thickness. Comparing the velocity profiles, the steady-state base flow exhibits localized reverse flow over the roughness patch within the boundary layer particularly in cases with taller roughness elements. Examination of the streamwise pressure gradient shows that, with an increase in the surface roughness element height, the magnitude of the compression and expansion shock waves increases. Linear stability analysis of the altered basic state indicates the presence of unstable modes, as evidenced by the eigenspectrum distribution. Larger instabilities are observed over the taller roughness elements, with the growth rate being almost three times higher than that of the shorter roughness elements. Spatial stability characteristics are investigated by simulating the development of a wavepacket disturbance imposed in the wall-normal velocity component. For identical perturbations, the streamwise disturbance spectrum of the baseline case shows an amplification of the high-frequency lobe, while the roughness-imposed case shows a relatively stable boundary layer. Wavelet analysis of surface pressure fluctuations at different streamwise locations further illustrates the modulation in the spectral characteristics of the boundary layer. Finally, dynamic mode decomposition of the pressure fluctuations in the spanwise midplane shows the dual-lobed pressure distribution associated with the high-frequency lobe, reaffirming the second mode instability characteristics. Placement of roughness elements impacts the stability characteristics and thus alters the spatial distribution of the pressure eigenmode.

  PublisherDOI

• Elucidating Three-Dimensional Coherent Structures in a Multi-Stream Jet

  arXiv (Cornell University) · 2026-01-22

  preprintOpen access

  Nominal two-dimensional (2D) shear layers have been studied extensively, and their principal dynamics are well understood. In practical configurations, however, the behavior of such shear layers is affected by proximal surfaces. In this study, we investigate three-dimensional (3D) coherent structures developing downstream of a relatively thick splitter plate in a realistic nozzle featuring sidewalls, an upper boundary formed by a single expansion ramp, and a lower boundary defined by a protruding deck. As a result, in addition to the primary splitter plate shear layer (SPSL) arising from mixing between the core and bypass streams, the flow contains upper (USL) and lower (LSL) shear layers with the ambient. Large-eddy simulation data are analyzed to characterize the unsteady flow dynamics, while the mean flow provides insight into the underlying amplification mechanisms. Spectral proper orthogonal decomposition reveals a clear separation of broadband and tonal dynamics across frequency bands. The broadband low-frequency modes are highly 3D and originate in the USL and LSL. In contrast, tonal high-frequency content is associated with a 2D instability in the SPSL. Both the broadband and tonal signatures also appear in the nonlinear energy transfer mechanisms. Triglobal resolvent analysis further clarifies the amplification mechanisms within the USL and LSL. Low-frequency response modes are excited by forcing localized near the nozzle geometry and are governed by 3D Kelvin-Helmholtz dynamics. The low-frequency streamwise vortices generated at the nozzle corners drive the axis-switching behavior characteristic of rectangular jets. Wavemaker analysis further demonstrates that these corner vortices are part of self-sustaining low-frequency dynamics.

  PublisherDOI

• Effect of Surface Motion on Background Unsteadiness in Turbulent Shock/Boundary-Layer Interactions

  2026-01-08

  articleSenior author

  A sufficient understanding of shock-dominated aeroelastic phenomena is critical to the routine fielding of high-speed flight vehicles. However, the role of the inherently unsteady flow in the aeroelastic response and the impact of the structural dynamics on the turbulence is poorly understood. To this end, implicit large eddy simulations (ILES) are used to explore the effects of wall motion on turbulent shock/boundary-layer interactions and to gain insights into the modification of inherent flow unsteadiness. In this work, the first eigenmode of a clamped-clamped beam is chosen to represent the structural dynamics observed in relevant experimental campaigns. Flow statistics are computed relative to the structural dynamics using phase and spanwise averaging, which is formalized using a triple decomposition of the flow. Comparisons across phase-mean streamwise velocity and surface pressure profiles present insights into the change in the bulk flow topology. In particular, a shift in the shear layer, mean separation shock position, and reattachment point is observed. Modifications to the flow unsteadiness are addressed using phase-local, second-order statistical moments. The surface motion modulates the background wall pressure and streamwise velocity fluctuations near the panel trailing edge and foot of the separated shear layer. Additionally, an analysis of the phase-local energy transfer between the phase-mean flow and background unsteadiness is performed. Changes in energy transfer due to deceleration and phase-mean shear are observed near the separation point and downstream of flow reattachment.

  PublisherDOI

• Benchmarking Long Roll-outs of Auto-regressive Neural Operators for the Compressible Navier-Stokes Equations with Conserved Quantity Correction

  ArXiv.org · 2026-01-30

  articleOpen access

  Deep learning has been proposed as an efficient alternative for the numerical approximation of PDE solutions, offering fast, iterative simulation of PDEs through the approximation of solution operators. However, deep learning solutions have struggle to perform well over long prediction durations due to the accumulation of auto-regressive error, which is compounded by the inability of models to conserve physical quantities. In this work, we present conserved quantity correction, a model-agnostic technique for incorporation physical conservation criteria within deep learning models. Our results demonstrate consistent improvement in the long-term stability of auto-regressive neural operator models, regardless of the model architecture. Furthermore, we analyze the performance of neural operators from the spectral domain, highlighting significant limitations of present architectures. These results highlight the need for future work to consider architectures that place specific emphasis on high frequency components, which are integral to the understanding and modeling of turbulent flows.

  PublisherOA PDF

• Receptivity and evolution of free-stream acoustic disturbances in hypersonic flow over cone–cylinder–flare

  Journal of Fluid Mechanics · 2026-02-06

  articleSenior author

  Direct numerical simulations are performed to investigate the receptivity and subsequent evolution of free-stream acoustic disturbances, including the associated instability mechanisms in a Mach 6 flow over a cone–cylinder–flare configuration. The geometry and flow parameters replicate an experimental study at the Purdue BAMQ6T facility (Benitez et al. , AIAA Aviation 2020 Forum , 2020, p. 3072). The results are analysed to reveal new physical insights into boundary-layer separation, instability growth and nonlinear processes. The effects of changing wall thermal conditions from the experimental cold isothermal ( $T_w = 30\,\text{K}$ ) to adiabatic (hot) are also examined. The basic state exhibits an attached boundary layer over the cone, followed by the formation of a separation bubble over the cylinder and flare, and reattachment over the aft section of the flare. In the case of a hot wall, the separation bubble size increases significantly compared with the isothermal case, leading to altered shear-layer dynamics and delayed reattachment with steeper gradients. Stability investigation reveals first- and second-mode disturbances as distinct spectral bands. Direct numerical simulation spectra and linear analysis indicate enhanced amplification of low-frequency first-mode disturbances for the adiabatic wall compared with the isothermal case. Bispectral analysis over the cone, centred at a second-mode wave, reveals weak subharmonic–fundamental coupling, but strong fundamental–fundamental coupling near the nosetip. The rapidly distorted mean flow within the separation bubble supports amplification of low-frequency disturbances, exhibiting an irregular spatial distribution, making it difficult to distinctly separate mutually exclusive modes (e.g. shear-layer or boundary-layer modes) due to their coexistence and influence on each other. Further downstream, the reattachment zone over the flare exhibits the combined effect of boundary layer and shear-generated waves, where distinct boundary-layer modes are evident at higher frequencies. Bispectral mode decomposition indicates strong phase-locked interaction along the leading-edge shock and within the separated and reattachment zones. These interactions are further amplified with increasing inflow forcing amplitude, leading to the formation of localised hotspots indicative of strong nonlinear amplification.

  PublisherDOI

Frequent coauthors

• Miguel R. Visbal

  United States Air Force Research Laboratory

  59 shared

• S. Unnikrishnan

  42 shared

• J. S. Shang

  Wuxi Institute of Technology

  39 shared

• Joseph Shang

  Wuhan University of Technology

  35 shared

• Jack J. McNamara

  The Ohio State University

  28 shared

• Chitrarth Prasad

  Oklahoma State University Oklahoma City

  26 shared

• Rajesh Ranjan

  Indian Institute of Technology Kanpur

  24 shared

• Faure Malo-Molina

  24 shared

Education

• PhD, Mechanical and Aerospace Engineering

  Rutgers University