Our research focuses mainly on computational modeling of physical systems, with an emphasis on numerical methods, fluid dynamics, and multi-physics simulation.
Current work focuses primarily on aerodynamic systems, combustion dynamics, and emerging AI/ML approaches for physical modelling.
Hydrogen Combustion Stratification in Multi-Jet Combustors
Aviation accounts for 2-3% of global CO2 emissions. Yet, unlike ground transport, aircraft cannot simply switch to electronic components like batteries as the energy density gap is too large. Hydrogen instead offers an alternative, with a gravimetric energy density three times times that of Kerosene and zero emissions at the point of use.
But burning Hydrogen is much harder than it sounds. In non-premixed combustors, the fuel and air enter separately and must mix before they can react. Because hydrogen’s laminar flame speed is so high, combustion is therefore mixing limited, which is when the flame burning is limited purely by how fast fuel and air can mix with one another. Stratification, which is the measure of mixture nonuniformity, therefore helps directly predict the efficiency of such engines.
This study uses a RANS-based CFD approach to run simulations to see how stratification changes along the combustor axis. I run 4 equivalence ratio cases (φ = 0.70–1.45) and use H2O as a proxy to quantity effects on the near-field zone to the far-field.
Read the initial report below!
Simulated analysis of the friction governed dynamics of the looping pendulum (NEW! Published!)
The dynamics of the looping pendulum, a double pendulum wherein a lighter and a heavier mass are connected by an inextensible string looped over a cylindrical rod, are governed by complex frictional interactions between the string and the rod. In this study, we present a comprehensive theoretical and experimental investigation of this system, with particular emphasis on the explicit distinction between static and kinetic friction regimes. This distinction is crucial because it governs the onset and recurrence of halts, thereby shaping the overall dynamical behaviour of the pendulum. Starting from first-principles Newtonian mechanics, we derive coupled ordinary differential equations. These equations are then solved numerically using a Web-VPython environment, enabling real-time visualization of the system’s motion and facilitating direct comparison with experimental data. Model predictions are validated against high-speed video recordings, yielding, at best, a correlation of approximately 98.5%. Our results show that multiple halt–slip cycles naturally emerge at mass ratios M/m ≈ 2 and M/m ≈ 50, a phenomenon not analysed in detail by previous models employing a single friction coefficient. The open-source simulation framework developed in this work provides an accessible platform for both educational and research applications. Our findings demonstrate the necessity of accurately modelling frictional transitions to capture the nonlinear dynamics of the looping pendulum and lay the groundwork for future studies of friction-governed mechanical systems.