Publications

  • All
  • Conference
  • Journal
  • Thesis

Blade envelopes Part II: Multiple objectives and inverse design

Journal: Journal of Turbomachinery
C. Y. Wong, P. Seshadri, A. Scillitoe, B. N. Ubald, A. Duncan, G. Parks
Blade envelopes offer a set of data-driven tolerance guidelines for manufactured components based on aerodynamic analysis. In Part I of this two-part paper, a workflow for the formulation of blade envelopes is described and demonstrated. In Part II, this workflow is extended to accommodate multiple objectives. This allows engineers to prescribe manufacturing guidelines that take into account multiple performance criteria. The quality of a manufactured blade can be correlated with …
  • Orthogonal Polynomials
  • Dimension Reduction
  • Sensitivity Analysis
  • RANS
  • 3D Modelling
  • Blender

Density Estimation from Schlieren Images through Machine Learning

Journal: Physics of Fluids (Under Review)
B. N. Ubald, P. Seshadri, A. Duncan
This study proposes a radically alternate approach for extracting quantitative information from schlieren images. The method uses a scaled, derivative enhanced Gaussian process model to obtain true density estimates from two corresponding schlieren images with the knife-edge at horizontal and vertical orientations. We illustrate our approach on schlieren images taken from a wind tunnel sting model, and a supersonic aircraft in flight. …
  • Machine Learning
  • Gaussian Process
  • PyMC
  • Bayesian Inference

Programming with equadratures: an open-source package for uncertainty quantification, dimension reduction, and optimisation

Conference: AIAA Scitech | San Diego, USA
P. Seshadri, C. Y. Wong, A. Scillitoe, B. N. Ubald, B. Hill, I. Virdis, T. Ghisu
This paper presents an overview of the open-source code equadratures. While originally developed to replicate polynomial chaos results seen in literature, it has since evolved to touch upon multiple aspects of computational engineering and machine learning. Today, the code uses orthogonal polynomial approximations to facilitate various parameter-based studies including uncertainty quantification, sensitivity analysis, dimension reduction, and classification. Additionally, it can address well-known limitations of polynomial approximations. These include the ability to …
  • Orthogonal Polynomials
  • Dimension Reduction
  • Sensitivity Analysis
  • RANS

Application of Immersed Boundary Method on Instrumented Turbine Blade With LES

Journal: Journal of Turbomachinery
B. N. Ubald, R. Watson, J. Cui, P. Tucker, S. Shahpar
Leading edge instrumentation used in compressor and turbine blades for jet-engine test rigs can cause significant obstruction and lead to a marked increase in downstream pressure loss. Typical instrumentation used in such a scenario could be a Kiel-shrouded probe with either a thermocouple or pitot-static tube for temperature/pressure measurement. High fidelity analysis of a coupled blade and probe requires the generation of a high quality mesh which can take a significant amount of an engineers time. The application of Immersed Boundary Method (IBM) and Large Eddy Simulation …
  • Dimension Reduction
  • CFD
  • RANS
  • Unsteady Flow

Quantitative Schlieren Using Gaussian Processes

Conference: AIAA Scitech 2021 Forum | Virtual, Online
B. N. Ubald, P. Seshadri, A. Duncan
Schlieren images abound in flow visualisation literature. By and large, these images have been used for the qualitative analysis of flow, such as the study of refraction patterns of shock waves and convection of plumes. Quantitative approaches, while present in literature, are limiting, typically requiring significant knowledge of both the flow and schlieren apparatus. In this paper, we propose a novel quantitative approach for schlieren---one that requires minimal information about the image taken and yet offers principled uncertainty estimates. Our approach …
  • Machine Learning
  • Gaussian Process
  • PyMC
  • Bayesian Inference

Application of Immersed Boundary Method on Instrumented Turbine Blade With LES

Conference: ASME Turbo Expo | Virtual, Online
B. N. Ubald, R. Watson, J. Cui, P. Tucker, S. Shahpar
Leading edge instrumentation used in compressor and turbine blades for jet-engine test rigs can cause significant obstruction and lead to a marked increase in downstream pressure loss. Typical instrumentation used in such a scenario could be a Kiel-shrouded probe with either a thermocouple or pitot-static tube for temperature/pressure measurement. High fidelity analysis of a coupled blade and probe requires the generation of a high quality mesh which can take a significant amount of an engineers time. The application of Immersed Boundary Method (IBM) and Large Eddy Simulation …
  • Dimension Reduction
  • CFD
  • RANS
  • Unsteady Flow

Design Space Exploration of Stagnation Temperature Probes via Dimension Reduction

Conference: ASME Turbo Expo | Virtual, Online
A. Scillitoe, B. N. Ubald, P. Seshadri, S. Shahpar
The measurement of stagnation temperature is important for turbomachinery applications as it is used in the calculation of component efficiency and engine specific fuel consumption. This paper examines the use of polynomial variable projection to identify dimension reducing subspaces for stagnation temperature probes. As an example application we focus on a simplified Kiel probe geometry, but the proposed data-centric approach could be readily applied to new datasets with different geometries, boundary conditions and design objectives.
  • Dimension Reduction
  • CFD
  • RANS
  • Unsteady Flow

Numerical analysis of an instrumented turbine blade cascade

Journal: Journal of Turbomachinery
B. N. Ubald, P. Tucker, J. Cui, R. Watson, S. Shahpar
The measurement accuracy of the temperature/pressure probe mounted at the leading edge of a turbine/compressor blade is crucial for estimating the fuel consumption of a turbo-fan engine. Apart from the measurement error itself, the probe also introduces extra losses. This again would compromise the measurement accuracy of the overall engine efficiency. This paper utilizes high-fidelity numerical analysis to understand the complex flow around the probe and quantify the loss sources due to the interaction between the blade and its instrumentation.
  • CFD
  • RANS
  • LES
  • Paraview
  • Unsteady Flow

Multi-fidelity study of complex multi-scale flows concerning stagnation probes

PhD Thesis: University of Cambridge, Aerospace Engineering | Cambridge, UK
B. N. Ubald
The work in this thesis aims to highlight, quantify the loss sources related to the use of stagnation probes in aero-engines through the use of RANS and high fidelity wall-resolved Large Eddy Simulations (LES), and help develop methods and techniques to aid in improving their design to reduce the impact of these losses. The measurement of stagnation temperature is an important performance metric for the calculation of component efficiency in aero-engines, particularly due to reductions in the margin of error over concurrent improvements in jet engine performance and measurement techniques. Hence, developing an understanding of the flow physics and evaluating any loss sources help aid in reducing calibration errors. (Full abstract available in link)
  • CFD
  • RANS
  • LES
  • Paraview
  • Unsteady Flow

Numerical analysis of an instrumented turbine blade cascade

Conference: ASME Turbo Expo | Oslo, Norway
B. N. Ubald, J. Cui, R. Watson, P. Tucker, S. Shahpar
The measurement accuracy of the temperature/pressure probe mounted at the leading edge of a turbine/compressor blade is crucial for estimating the fuel consumption of a turbo-fan engine. Apart from the measurement error itself, the probe also introduces extra losses. This again would compromise the measurement accuracy of the overall engine efficiency. This paper utilizes high-fidelity numerical analysis to understand the complex flow around the probe and quantify the loss sources due to the interaction between the blade and its instrumentation.
  • CFD
  • RANS
  • LES
  • Paraview
  • Unsteady Flow

Multi-fidelity simulation for a transonic compressor with inflow distortions

Conference: ISROMAC 2017 | Maui, Hawaii
J. Cui, Y. Ma, B. N. Ubald, P. Tucker
The low and high fidelity methods are used to study a transonic fan case. This transonic fan case features a distortion generator upstream of the rotor. The low fidelity methods (immersed boundary method and immersed boundary method with smeared geometry) show promising results against hi-fi simulation (unsteady RANS) and experiments.
  • CFD
  • Paraview
  • Unsteady Flow

Assessment of RANS modelling for a complex stagnation probe case

Conference: 55th AIAA Aerospace Sciences Meeting | Grapevine, Texas
B. N. Ubald, P. Tucker, S. Shahpar
The measurement of stagnation temperature is an important performance metric in various thermal systems as it can be used for the calculation of component efficiency. With concurrent improvements in jet engine performance and measurement techniques, the margin for error is reduced significantly. Heat transfer by convection can affect the accuracy of the sensor when measuring/simulating the stagnation temperature. This paper will quantify the performance of RANS simulations. In addition to the complexity of the problem, the implementation of each RANS model can vary widely for different codes based on in-house testing and application-specific tuning, hence, results from two different codes are presented.
  • CFD
  • RANS
  • Ansys Fluent
  • BoXer
  • Paraview
  • Stagnation Probes

Local Dispersion Behaviour at Marylebone Road Monitoring Site

Masters Thesis: University of Surrey, Aerospace Engineering | Guildford, UK
B. N. Ubald
One method of modelling the dispersion behaviour for the purpose of understanding the physical processes involved in an urban environment, is through the use of a wind tunnel. This present study seeks to model dispersion behaviour in a region surrounding Marylebone Road, London. This area is of particular interest as it is recorded as being one of the most polluted roads in central London and also due to the long standing presence of an air pollution monitoring station. The primary aim of this report is to use wind tunnel modelling to evaluate the impact of the monitoring station on measurements taken at the field site. The secondary aim is to investigate the spatial relationship between the dispersion of air pollutants and several areas of interest, while also gathering good quality data that can be used for validating computational models.
  • Wind Tunnel
  • Environmental Flow
  • Pollutants
  • Marylebone Road
  • Pollutants
  • Sketchup Flow