Heriot-Watt University
ROS: The Research Output Service. Heriot-Watt University EdinburghNot a member yet
4689 research outputs found
Sort by
Mathematical modelling and analysis of soil and plant root interactions
Get PDFThe influence of plants on soil water transport is a relevant factor in a number
of ecological contexts. Examples include: the resistance of crops to drought, the
prevention of floods and the protection of soils from erosion. There exists strong
experimental evidence that interactions between soil and plant roots change a soil’s
hydraulic properties. Nevertheless, it remains a challenge to anticipate the impact
of specific root traits on the infiltration of water through soil.
In an attempt to address the issue above, this thesis presents modifications of
Richards’ equation—the classic model for water transport through soil—to incorporate some effects that root systems are known to have on soil hydraulic properties.
First, a model is developed that incorporates the phenomenon of root-oriented
preferential flow. Using the finite element method and Bayesian optimisation, a
pipeline is developed to calibrate the model against experimental data. Moreover, it
is shown how existing root architectural models can be used in conjunction with our
model to investigate the influence that root system traits have on infiltration and
water uptake. Results suggest that this modification of Richards’ equation leads
to improved agreement of simulations with reference pore water pressure profiles,
which are derived from experimental data regarding the hydraulic conductivity of
vegetated soils.
Following this, the developed model is used to obtain simulations of various
infiltration scenarios. These reveal that, up to a critical point, increasing preferential
flow strength reduces water loss from the rooted zone. Furthermore, evidence is
provided to suggest that root systems with a reduced gravitropic response allow a
greater retention of water in the rooted zone following precipitation and, hence, are
among the most effective at delaying the onset of water deficits.
In another case, an alternative modification is proposed whereby Richards’ equation is coupled with an equation for water transport through roots. This model
accounts for root water uptake and hydraulic lift through a Neumann boundary
condition at the root-soil interface. By using the methods of Rothe and Galerkin,
existence of a solution to this coupled model is then established. Uniqueness is
shown by Kruzkov’s variable doubling method, but applied only in time.UK Engineering and Physical Sciences Research Council (EPSRC) grant
EP/L016508/01Scottish Funding Counci
Radar-based localization and mapping for large-scale environments and adverse weather conditions
Get PDFIn mobile robotics, localization and mapping is one of the fundamental capabilities
towards autonomy. Navigating autonomously in large-scale, unstructured, extreme
and dynamical environments is particularly challenging due to the high variations in
the scene. To deliver a robotic system that can operate 24/7 in outdoor environment,
we need to design a state estimation system that is robust in all weather conditions.
In this thesis, we propose, implement and validate three systems to tackle the
problem of long-term localization and mapping. We focus on using radar-only platform to realize the SLAM and localization systems in probabilistic manners. We
first introduce a radar-based SLAM system that can operate in city-scale environment. Second, we present an improved version of the radar-based SLAM system
with enhanced odometry estimation capability and extensive experiments on extreme weather conditions, proving that our proposed radar SLAM solution is viable
in all weather conditions. We also demonstrate the superiority of radar-based SLAM
system compared to LiDAR and vision based system in snowy and low light conditions respectively. Finally, we show how to combine online public maps and radar
sensor to achieve accurate localization even we do not have a prior sensor map. We
show that our proposed localization system can generalize to different scenarios and
we validate it across three datasets collected in three different continents
Quantum XOR and Rabin oblivious transfer
Get PDFOblivious transfer is a cryptographic primitive involving two non-trusting communicating parties. Since it is a basic building block for any two-party computation,
it is a quite powerful and important cryptographic functionality and thus topic of
various research investigations in the classical as well as in the quantum setting. It
was unfortunately shown that oblivious transfer can in neither setting be done with
information-theoretic security. However, in the quantum case, it is possible to limit
the cheating probabilities of unrestricted dishonest parties.
The most well-known variant is 1-out-of-2 oblivious transfer, where the sender
sends two bits and the receiver receives one of them without the sender learning
which one was received. While this has been the primary focus of investigations,
there exist other variants of the protocol which have been less studied. This thesis
focuses on two such variants, XOR oblivious transfer and Rabin oblivious transfer.
Different quantum protocols for these two variants are presented and analysed
for their security against cheating parties. Calculating the cheating probabilities in
general for non-interactive XOR oblivious transfer with symmetric states, the optimality of the presented XOR oblivious transfer protocol is shown. Non-interactive
means that there is only one state transmission from the sender to the receiver who
applies a measurement, and no further communication between the parties. We
further extend the concept of XOR oblivious transfer to the sender not sending two
but n bits and analyse the effect of an increasing n on the participants’ cheating
probabilities.
The reversal of oblivious transfer is also looked at; that is, implementing oblivious
transfer in both directions even if only one of the two communicating parties can
send a quantum state and the other one can only measure. We determine the
reversed protocol versions of a 1-out-of-2 and an XOR oblivious transfer protocol
and show that the protocols’ cheating probabilities remain unchanged.
For Rabin oblivious transfer, both protocols using pure states and protocols using
mixed states are investigated. Comparing them to each other, we determine under
which circumstances the protocol with the pure states outperforms the protocol with
the mixed states and vice versa
Integrated photonics with semiconductor based quantum light sources
No full textIntegrated nanophotonic circuits have reached a high level of sophistication and
maturity but they do not generate light efficiently. For this reason, hybrid integration of quantum light generating materials is required. Atomically thin Transition
Metal Dichalcogenides (TMDs) are particularly intriguing light sources because
they exhibit strong light-matter interaction, valley and spin-dependent optical
properties and the possibility of deterministically create localised quantum light
sources through mechanical strain and ion etching. In addition, TMDs can be engineered into heterostructures containing arrays of exotic quantum light sources
through low-cost fabrication techniques. However, major challenges still need to
be addressed in the fabrication of complex heterostructures and milestone achievements such as the Purcell effect have still to be demonstrated for exotic emitters
in this type of platform. On the other hand, self-assembled III-V quantum dots
have reached milestone results, including the Purcell enhancement of irradiated
photons, but more has to be done to achieve complete scalability.
This thesis addresses the main challenges that arise during the design, fabrication
and characterisation of integrated quantum photonic devices, with a particular
focus on TMD materials and III-V semiconductors as harvesting platforms for
quantum light sources. Chapter 1 will provide an introduction to the relevant
physics to understand the projects of the thesis. Chapter 2 focuses on the design
of a multi-spot confocal microscope, built to measure photoluminescence from
spatially separated spots on an integrated photonic circuit. Chapter 3 focuses on
the integration of TMD materials with silicon nitride waveguides and nanobeam
cavities. Chapter 4 investigates the measurement of the optical dipole orientation of excitonic complexes in TMDs, essential for determining the strength of
light-matter interaction. Chapter 5 explores strain tuning through laser heating
of GaAs quantum dots in a multiplexed waveguide architecture. Chapter 6 focuses on the design and assembly of a multi-core fibre based confocal microscope
for multiplexed spectroscopy of engineered ion-etched localised emitters in MoS2.
Chapter 7 is a brief account of the study of methods for interconnecting waveguide based chips with optical fibre networks, essential for long-distance quantum
communication
Commercial banking and financial inclusion of Uganda’s middle class - the case of Kampala city
Get PDFStudies have demonstrated that financial inclusion or being part of a formal financial system,
has the potential to improve the quality of life, and is one of the key pillars of economic growth,
given that it promotes social inclusion and reduces poverty levels. Consequently, a number of
jurisdictions including Uganda, have over the years put in place deliberate policies to encourage
the financial inclusion of their citizenry. However, the national census of 2014 revealed that
approximately 57 percent of the urban households did not have bank accounts. Most of the
middle class reside in urban areas and, therefore, it implies that a large number of them appear
to be excluded from the banking sector. It is worth noting that in the context of Uganda and
this research, bank account do not include mobile money accounts, although mobile money
companies are regulated by the Central Bank.
The middle class play a significant role in promoting economic growth especially through their
purchasing power. As such, their continued exclusion from the formal banking sector,
specifically that which is regulated by the Central Bank, could have far-reaching implications
on the pace of the country’s economic growth. A significant proportion of bank deposits in the
formal banking sector regulated by Bank of Uganda sit in Commercial banks, with a small
percentage held in Credit Institutions and Micro finance Deposit Taking Institutions (MDIs).
As such, this study, sought to understand the major factors affecting access, usage and quality
of commercial banking services accessed by the middle class, using a case study of Kampala
which is the capital city of Uganda.
It has been ascertained that financial inclusion is sensitive to context and is affected by the
emotions of people. It was against this background that a literature review on factors affecting
financial inclusion in six (6) countries was conducted. The countries spanned across the
developed and less developed world namely: United States of America, United Kingdom, India,
Nigeria, Kenya and Uganda. The review gave the researcher a broad view of the various factors
affecting financial inclusion or access to formal financial services generally, excluding mobile
money. These were subjected to further research, to ascertain whether they affected the middle class in Uganda. In order to obtain credible research findings, a mixed methods research
approach was adopted. Logistic regression and the Linear Probability Model (LPM) were used
to analyse the primary data which was collected using questionnaires completed by a quasi
random sample of two sub-groups of the urban middle class. Follow-up qualitative interviews
were held with respondents who were willing to share additional information. This enabled the
researcher obtain a better appreciation of the motivations and experiences lying behind the
statistical results. The results of the study revealed that the major factors affecting financial
inclusion amongst the Ugandan middle class were; high bank charges, inadequate handling of
customer complaints, fear of compromised privacy, low trust in the banking sector, long lines
in banking outlets, negative experiences with banks, unstable internet and ATM services,
amongst others.
Arising from these findings, and benchmarking with other countries, the study has highlighted
a number of recommendations for various stakeholders. The Government of Uganda should
consider the need to have a professionally managed and widely marketed Government owned
commercial bank with a large branch network to cater better for the needs of the middle and
lower class Ugandan. The Central Bank should consider engaging with commercial banks to
explore the possibility of opening ‘basic bank accounts’ or ‘no frill accounts’ which do not
attract any fees or charges, for certain segments of the population, as has been done in other
jurisdictions. Additionally, Bank of Uganda should enhance its effort in strengthening the
customer complaints management process by both the commercial banks and within the Central
Bank itself. Alternatively they could consider establishing a Consumer Protection Department
or Unit for financial services. The Government should intensify its drive of providing stable,
faster and affordable internet across the country in order to better support alternative channels
of banking. Finally, the public should be better informed about the role of the Deposit Protection
Fund of Uganda in compensating depositors up to the insured limit in the event of a bank
closure
Machine learning approaches for slice to volume biomedical data integration
Get PDFImaging plays an essential role in modern biomedical sciences and lays the foundation for current research and clinical diagnosis. During the last decade, slice-to-volume registration, a particular type of image registration problem, has received
great attention from the medical imaging community due to the emergence of several
medical applications of slice-to-volume mapping (2D to 3D image mapping) using
biomedical data such as biomedical atlas. The task of integrating new data into a
biomedical atlas is a typical 2D to 3D image registration problem. Images created
in experiments are mostly 2D images, while modern biomedical atlases are mostly
3D models. To transfer the data related to the 2D image (e.g., gene expression
data) to the 3D Atlas, it is necessary to determine the position of the new image
in the 3D model. This is typically done by experts who review the 2D sections and
manually position 2D data into 3D with some tools. Manual positioning 2D data
into 3D is financially expensive, time consuming, and require extensive work by experts. However, finding experts who have domain knowledge is also another crucial
challenge. To resolve this problem, this thesis automate the process of positioning
the 2D image into the 3D model. This study contributes by creating two datasets
that convert the 3D Atlas into a series of 2D slices. Then, we utilize a Convolutional Neural Network (CNN) for registering purposes. The proposed CNN model
is trained to determine the distance and pitch values used to describe the position
of the 2D slice in the atlas coordinate system, and the proposed model obtained
94% accuracy. Furthermore, we tested different variants of CNN architectures and
different transfer learning techniques to build an optimal image base model for image analysis. We employ all the data modalities available in the biomedical Atlases,
such as the images and the textual anatomical data. To test the performance in
real-life situation, the performance of the proposed model is evaluated on the unseen dataset. The results show that the proposed model outperforms the image-only
data and obtain 97% accuracy. A different data set (contained cropped images) is
used to test the performance of the proposed technique for image matching, and
the algorithm achieved 94% accuracy. The study has shown that different data
modalities available within the atlases can train the machine learning to overcome
many of the issues related to the use of image-processing based or ontology-based
techniques
Corruption in Egyptian construction : an anti-corruption framework
Get PDFCorruption is a major issue facing the global construction market, where its covert nature
hinders mitigation efforts. Corrupt actions undermine project management strategies,
costing the industry billions worldwide. NGO reports suggest Egypt has elevated
corruption levels; however, there is a lack of academic research regarding corruption in
Egyptian construction. Construction corruption studies were conducted in other countries,
which informed this research. However, due to corruption’s subjectivity, it is important
to study the issue from the Egyptian perspective.
The research aims to understand corruption in Egyptian construction and to provide
recommendations to address it. To achieve this, firstly, a literature review was conducted
to explore corruption characteristics, corruption research in construction management,
and corruption in Egypt. Secondly, 18 semi-structured interviews were conducted among
Egyptian construction professionals to understand corruption in the industry and explore
efforts of combating it. Based on findings from primary and secondary sources, an anti-corruption framework (the first of its kind in this context) was developed to be
implemented in Egyptian construction. Eight industry professionals evaluated the validity
of the framework and provided suggestions for its enhancement.
This research is the first to present a narrative regarding corruption in Egyptian
construction, indicating that corruption is part of the Egyptian construction culture, and
is caused by need, greed, and occupational alienation. These are enabled by a limited
understanding of corruption, resistance to corrective action, corruption normalcy, and the
difficulty of surviving in the industry without participating in corruption. Construction
organisations make no effort to combat corruption, and laws are unevenly enforced,
which results in extended project durations, increased project costs, poorer quality end
products, an unfair business environment, and further corruption normalisation. The
developed framework provides a holistic approach to addressing corruption in Egyptian
construction and can enable the reduction of corruption provided country level support
can be obtained. The framework can also be adopted in other countries by adjusting its
operational procedures to the requirements and capabilities of said countries
DMAIC 4.0 - innovating the Lean Six Sigma methodology with Industry 4.0 technologies
Get PDFLean Six Sigma (LSS) is a continuous improvement methodology that emerged around
2000 (George and George, 2002; Snee, 2010). It combines the strengths of two
methodologies, Lean and Six Sigma, into an effective process and quality improvement
framework. Although many organisations have successfully applied LSS over the past
two decades, over 60% of Lean and Six Sigma implementations have failed (Albliwi et
al., 2014; Sony et al., 2020c), and, accordingly, a significant number of improvement
projects. Consequently, researchers have investigated the reasons behind these failures
and revealed numerous failure factors, criticisms, impediments, and barriers that
jeopardise the success of LSS initiatives. These reasons, also recognised as LSS
limitations, represent the problem addressed in this research.
On the other hand, the Industry 4.0 (I4.0) era, entailing machine connectivity, big data
technologies and artificial intelligence, offers new opportunities for data-driven quality
improvement strategies such as LSS. Therefore, this study explored how I4.0
technologies can enhance the traditional LSS methodology by following a Design Science
Research (DSR) approach. The aim was to design a solution integrating I4.0 data-driven
tools into the traditional DMAIC framework to enhance the success and effectiveness of
LSS projects. DMAIC stands for Define, Measure, Analyse, Improve, and Control,
representing project phases executed in a prescribed order. The designed solution is a
DMAIC 4.0 framework that should help organisations overcome the limitations of LSS
by exploiting modern technologies and techniques.
This study adopts the DSR process described by Peffers et al. (2007), combined with
qualitative methods suggested by Offermann et al. (2009). There are three main phases:
(1) Problem Identification, (2) Solution Design and (3) Evaluation. Expert interviews
were conducted in phase 1 to confirm the problem and underpin its relevance. The design
built in phase 2 is based on existing knowledge and field experience. In phase 3, the
researcher successfully evaluated the framework’s utility and effectiveness within a
German manufacturing organisation through action research. Additionally, a Delphi
study demonstrated that the design presented is relevant and applicable to various
industries. Upon Delphi panel feedback, a roadmap was created to guide organisations in
implementing the new framework.
To the authors’ knowledge, this is the first DMAIC 4.0 framework presented in the
academic literature thus far. Knowledge and novel contributions were generated through the design and evaluation process. The validated framework includes 42 LSS tasks
enhanced by I4.0 technologies. It incorporates knowledge from extant research related to
LSS, DMAIC and I4.0. Furthermore, it focuses on tools and tasks and is more detailed
than previously presented frameworks integrating I4.0 with LSS. Unlike conceptual
frameworks, it is empirically validated, which should motivate LSS practitioners to
innovate their projects. Clearly, there is still room for expansion as there are many more
tools in both areas, LSS and I4.0. Researchers and practitioners can customise and apply
the framework in various contexts to establish a new standard for DMAIC
Novel quantum emitters for integrated quantum photonics
Get PDFNovel quantum emitters based on quantum dots in 2D materials and defects in
silicon carbide were investigated. To this end, a state-of-the-art low-temperature
setup to study quantum emitters, was created, beginning from an empty lab. The
system allows for 3d spatial mapping of photoluminescence from single photon emission centres, such as quantum dots and defects in wide-band gap semiconductors.
The system is based around a cryostat which allows for cooling to 4K and coplanar
waveguides are used to facilitate magnetic resonance experiments, incorporating a
coplanar waveguide for applying microwave control pulses, enabling magnetic resonance experiments.
The first time intergration of single-quantum emitters in 2-D materials to simple
optical integrated circuits is outlined, demonstrating a key step in the theme of
scalable quantum emitters and showcasing a novel platform for quantum photonics.
This theme of integrated quantum photonics is further expanded, where we showcase
divacancy related defects in silicon carbide. We demonstrate that this platform benefits from long coherence time electron spins which can be addressed with microwave
magnetic fields, as shown in our optically detected magnetic resonance experiments.
We further show that light extraction from the bulk crystal can be enhanced with
simple and scalable nanopillar arrays and outline the future direction of developing
scalable immersion lenses.
I aid of discovery and characterisation of novel emitters, we demonstrate a new tool
for single photon spectroscopy, the TWINS time domain spectrometer, with particular appeal for research in the field of defects in wide-band gap semiconductors.
This system uses a single element detector is extremely beneficial when working
around the highly desirable telecom window, where for previously spectroscopic
measurements where made difficult and costly through the use of InGaAs arrays.
The system can be easily integrated with the recently developed Superconducting
Nanowire Single Photon Detectors, for extremely low dark counts, high quantum
efficiency and low timing jitter. I describe how the technique has the potential to aid
in the discovery and characterisation of new defects as well as develop understanding of underlying energy levels and charge states through the use of time-resolved
spectroscopy
Numerical simulation of WAG and SWAG injection in carbonate rocks at reservoir conditions
Get PDFSimulation of Enhanced Oil Recovery methods such as Water-Alternating-Gas
(WAG) at reservoir-scale requires precise estimation of three-phase relative permeability
(kr) data and the hysteresis phenomenon. Many physical factors affect the relative
permeability data, such as wettability, interfacial tension (IFT), heterogeneity, hysteresis,
and pore size distribution of the porous media. Despite its significant importance, the
study of carbonate rocks has attracted less attention, and most of the research available in
the literature is focused on cyclic experiments performed on sandstone rocks.
This thesis focuses on simulation and numerical analysis of fluid flow in mixed-wet
carbonate rocks in terms of oil recovery, trapped gas saturation, the effect of hysteresis
on relative permeability and changes in relative permeability with interfacial tension. For
this purpose, there are a series of two-phase (gas/oil) coreflood experiments and WAG
and SWAG (Simultaneous Water And Gas) experiments have been conducted on a
mixed-wet homogenous carbonate rock at various conditions (near miscible, immiscible
and miscible). These experiments were performed using real reservoir fluids.
Experiments are performed with different injection scenarios and at high and low IFT
conditions. Then, the results of the coreflood experiments were history matched using
3RPSim to generate two-phase and three-phase relative permeability data.
The first aim in this research, investigate the performance of three-phase relative
permeability models, which are used in oil industry for simulating of WAG and SWAG
injections for homogenous carbonate rock. Then, the hysteresis model suggested by
Heriot-Watt University was used for the estimation of hysteresis in relative permeability
data. The performance of the model was compared with the experimental data from
sandstones to evaluate the impact of heterogeneity on the hysteresis phenomenon. The
results presented in this thesis show that the hysteresis effect is less dominant in
homogeneous porous media. In this research, a recently published hysteresis model
developed by Aghabozorgi et al. (2022) at Heriot-Watt University (referred to as
Aghabozorgi’s Method or HWU model) was used to simulate the WAG experiments
conducted on mixed-wet homogenous carbonate rock. This model was previously tested
in WAG experiments using a model oil (binary C1-C4 system) on a sandstone core.
Overall, the HWU hysteresis model improved the match for fluid saturation, trapped gas
saturation and pressure drop.
The second part of this study, simulate Simultaneous Water And Gas (SWAG)
injection experiments are performed to better understand the fluid flow behaviour in the reservoir. The results and discussions presented here show that SWAG experiments are
different from WAG experiments and ordinary steady-state three-phase experiments.
Therefore, they cannot be fully described using the models developed for these flow
patterns. The SWAG study presented in this research is novel due to the lack of SWAG
experimental data conducted in reservoir carbonate rocks using real reservoir fluids.
Mostly, the rocks used in the experiments are outcrops using synthetic fluid. The study in
this work is important study because none of the researchers discussed the three-phase
relative permeability for SWAG injection and how to predict SWAG injection by
simulation analysis based on experiment data.
Finally, the effect of IFT on the relative permeability data has been discussed. In the
last chapter, a new method is presented which relates the residual oil saturation to the IFT
value. A relationship similar to the Michaelis Menten Kinetics model (widely used in
biology studies) is suggested to estimate the residual oil saturation as a function of
interfacial tension. Also, a new set of correlations was proposed to calculate gas and oil
relative permeability at different IFT. The accuracy of the model is then assessed against
experimental data available in literature and predictions of a default model in commercial
simulators (Coat’s model). Although the model needs fewer input data and it requires
fewer calculations than Coat’s model, it improved predictions. By using the new
methodology, the error in estimating gas permeability and oil relative permeability was
reduced by 50%. This approach can provide an alternative method to the available models
for IFT scaling, such as Coats’ model.
The novelty of this study can be summarised as follows;
Evaluating and finding three-phase relative permeability for fluids at different
WAG experiments by using real fluids and real carbonate rock.
• Examine the effect of heterogeneity on three-phase flow and relative
permeability hysteresis. The study prove that hysteresis is ignored in the
homogenous system.
• In this research, three-phase relative permeability for immiscible SWAG were
found by history match, and that led to understand the SWAG behaviour in real
porous media.
• Develop new correlations to Estimate residual oil saturation and gas/oil two-phase relative permeability curves at different IFT. This correlation is not
restricted to any rock type or fluid type
Access Repository Dashboard
Do you manage ROS: The Research Output Service. Heriot-Watt University Edinburgh ? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!