arXiv subjects

Jonathan F. Morrison

Publications and source records attributed to Jonathan F. Morrison.

3 recordsLinked to original sources

arXiv2026

Flow behind the Imperial Front Wing: comparison of results from volumetric PTV experiment and Nektar++ simulations

High-fidelity simulations are increasingly adopted, due to advances in computational power and methods such as Direct Numerical Simulation (DNS) and hybrid Large-Eddy Simulation (LES). These approaches are particularly valuable for unsteady flows around complex geometries at high Reynolds numbers; however they still require careful experimental validation. Planar and stereo Particle Image Velocimetry (PIV) are widely used for measurements but limited by measurement-plane selection and their ability to capture vortices shapes and trajectories. This motivates the growing interest in volumetric techniques, historically difficult to implement in industrial settings. Recent advances in Particle Tracking Velocimetry (PTV) for measuring flows over large volumes make this approach suitable for validating numerical simulations of complex flows.This study compares volumetric PTV measurements against high-fidelity LES to assess the capabilities and limitations for industrial flows. The aim is to establish a benchmark PTV dataset for motorsport aerodynamics using the Shake-The-Box algorithm. The experiment was carried out in the 10x5 wind tunnel at Imperial College London equipped with a rolling road for ground effect simulation and capable of testing up to 50% scale F1 model. Volumetric PTV measurements were performed downstream of the open-source Imperial Front Wing (IFW) at Re=74896. Results are compared with planar PIV studies and implicit LES simulation using spectral h/p elements in Nektar++. This work addresses open questions in the literature concerning the wake of the IFW. Good quantitative agreement is observed in the wake topology. A previously unreported vortex is identified which has the key role of preventing the merging of other dominant structures. These results demonstrate the suitability of PTV and STB for industrial applications while providing a benchmark dataset for the IFW.

physics.flu-dyn↗

arXiv2019

Intermediate Scaling and Logarithmic Invariance in Turbulent Pipe Flow

A three-layer asymptotic structure for turbulent pipe flow is proposed, revealing in terms of intermediate variables, the existence of a Reynolds-number invariant logarithmic region. It provides a theoretical foundation for addressing important questions in the scaling of the streamwise mean velocity and variance. The key insight emerging from the analysis is that the scale separation between two adjacent layers is proportional to $\sqrt{Re_τ}$, rather than $Re_τ$. This suggests that, in order to realise Reynolds-number asymptotic invariance, much higher Reynolds numbers may be necessary to achieve sufficient scale separation. The formulation provides a theoretical basis for explaining the presence of a power law for the mean velocity in pipe flow at low Reynolds numbers and the co-existence of power and log laws at higher Reynolds numbers. Furthermore, the Townsend-Perry `constant' for the variance is shown to exhibit a systematic Reynolds-number dependence.

physics.flu-dyn↗

arXiv2015

Modelling for Robust Feedback Control of Fluid Flows

This paper addresses the problem of obtaining low-order models of fluid flows for the purpose of designing robust feedback controllers. This is challenging since whilst many flows are governed by a set of nonlinear, partial differential-algebraic equations (the Navier-Stokes equations), the majority of established control theory assumes models of much greater simplicity, in that they are firstly: linear, secondly: described by ordinary differential equations, and thirdly: finite-dimensional. Linearisation, where appropriate, overcomes the first disparity, but attempts to reconcile the remaining two have proved difficult. This paper addresses these two problems as follows. Firstly, a numerical approach is used to project the governing equations onto a divergence-free basis, thus converting a system of differential-algebraic equations into one of ordinary differential equations. This dispenses with the need for analytical velocity-vorticity transformations, and thus simplifies the modelling of boundary sensing and actuation. Secondly, this paper presents a novel and straightforward approach for obtaining suitable low-order models of fluid flows, from which robust feedback controllers can be synthesised that provide a priori guarantees of robust performance when connected to the (infinite-dimensional) linearised flow system. This approach overcomes many of the problems inherent in approaches that rely upon model-reduction. To illustrate these methods, a perturbation shear stress controller is designed and applied to plane channel flow, assuming arrays of wall mounted shear-stress sensors and transpiration actuators. DNS results demonstrate robust attenuation of the perturbation shear-stresses across a wide range of Reynolds numbers with a single, linear controller.

physics.flu-dyn↗