85000 research outputs found
Sort by
Mapping cross-national conceptions of essential energy to advance housing energy justice - READ ME
No full textThe project entailed qualitative research from focus groups and urban tours. The study leveraged the cross-national collaboration of the WELLBASED project. WELLBASED was a European Union Horizon 2020 project that was trialling energy poverty interventions between 2021 and 2025 across six cities: Leeds (UK), Valencia (Spain), Heerlen (Netherlands), Edirne (Turkey), Obuda (Hungary) and Jelgava (Latvia). The study captured and compared the WELLBASED professionals’ perceptions of essential energy.This dataset contains transcripts of focus groups, photos of urban tours in five of the six cities, and varied versions of a Comic that was produced as part of this project.</p
Simulation and Characterisation of Morphology of Meltpool in Electron Beam Powder Bed Fusion of Copper
No full textAdditive Manufacturing (AM) is an emerging manufacturing process that has been widely used in different high-value industries such as aerospace, defence, automotive and medical. Among all different AM processes, powder bed fusion, directed energy deposition and binder jetting are common to process the metal components. In this research, the electron beam powder bed (EB-PBF) fusion process of pure copper for single tracks component is investigated. The effect of input parameters comprising of beam power, beam speed, and layer thickness on manufacturability including dimensional accuracy and surface quality are discussed. The next phase of this project is to examine the effect of EB-PBF process parameters on the thermal history, morphology of meltpool in the context of the fundamental phenomena that drive the EB-PBF response. To achieve this, a numerical multiphysics simulation of the EB-PBF process will be conducted using computational software such as Flow3D. The numerical simulation is based on powder interaction with the electron beam and the fundamental rheological phenomena during the process. This can then be related to the effect of process parameters on both temperature (both qualitatively and quantitatively) and meltpool features and morphology (transient and steady-state width, depth) as well as fusion mode (for example transition from conduction mode to keyhole mode). Numerical simulation has the potential to provide comprehensive information on both morphology and temperature as well as transient and steady-state phenomena, it is associated with a high computational cost. As an alternate prediction capability, an analytical model was developed to estimate meltpool temperature. Although this analytic method cannot provide insight onto meltpool morphology, or transient behavior, it has the potential to provide computationally rapid insights into the effect of process parameters and material properties on steady state average melt-pool temperature. The final stage of this study involves experimentally assessing the impact of EB-PBF copper process parameters of beam power and scanning velocity on the morphology of the deposited melt-track temperature. This data can be used to provide corroborating insights into the predictions of the numerical simulation and analytical model of the melt pool. By providing previously unavailable predictive models and experimental data on associated meltpool phenomena, this research contributes valuable insights for the optimization and understanding of the manufacturing process into EB-PBF of pure copper, with potential applications in various industries.</p
Modelling Heterogeneous Time Series from Irregular Data Streams
No full textRapid technological advances continuously generate vast amounts of real-world temporal data, playing a vital role in IoT and smart city applications. Despite the promising capabilities of time series analysis, it faces significant challenges when applied to complex real-world systems like healthcare or environmental monitoring. These systems produce irregular, high-dimensional, heterogeneous, and non-stationary data, posing difficulties for traditional machine learning models. Traditional models struggle to capture intricate relationships of this data without extensive preprocessing that consumes time and resources and distorts temporal dependencies, ultimately affecting prediction accuracy and timely access to information. Our research tackles this challenge by developing efficient modelling processes to seamlessly handle irregular, sporadic, highly dimensional, and heterogeneous time series data. Our goal is to advance temporal learning for modelling multi-source inconsistent time series in real-world applications, eliminating unnecessary preprocessing steps.
This thesis makes significant contributions towards effectively learning and modelling complex data. The primary objective is to bridge the gap between time series models and real-world big data, characterized by high levels of variety, volume, and velocity. These attributes present various challenges in data modelling, particularly since real-world data is typically collected in irregular environments from diverse sources. Consequently, this work focuses on three primary areas to enhance the modelling process for real-world datasets. First, we focus on modelling heterogeneous multi-source time series, introducing Parallelised Irregularity Encoders for Forecasting with Heterogeneous Time Series (PIETS) and Parallelised Irregularity Encoders for Multi-step Forecasting with Heterogeneous Time Series (PIETS+). These novel frameworks are specifically designed to tackle the complexities of multi-source time series analysis, which pose a significant challenge in real-world applications. Traditionally, the fusion of multi-source time series has been approached either through ensemble learning models that overlook temporal patterns and correlations within features or by defining a fixed-size window to select specific parts of the datasets. Our proposed models, PIETS and PIETS+, demonstrate enhanced predictive capabilities for both one and multi-step forecasting while effectively modelling heterogeneous time series. The proposed work addresses key challenges of multi-source time series data, including (1) heterogeneity and irregularity, (2) information inconsistency, and (3) highly variable dimensions. By leveraging information from diverse data sources, our models not only outperform in capturing the complexity of temporal data but also accelerate the convergence of the training process.
Next, we delve deeper into the challenge of irregularity, focusing on highly sporadic time series with consecutive unobserved values. Irregular time series, characterised by undefined intervals between observations leading to sporadic sequences, have recently been addressed using neural ordinary differential equations (ODEs). In our research, we investigate the performance of ODE models on time series data with varying levels of sparsity. Following this examination, we introduce SeqLink, a robust neural-ODE architecture for modelling partially observed time series. SeqLink is an innovative neural architecture designed to enhance the robustness of sequence representation. Unlike traditional approaches that rely solely on the hidden state generated from the last observed value, SeqLink leverages ODE latent representations derived from multiple data points, enabling it to generate robust data representations regardless of sequence length or data sparsity. Our model demonstrates its ability to maintain robust continuous representations even over long timescales.
Finally, we expand our contribution to focus on irregular streaming time series and continual learning. As real-world applications involve a continuous flow of data, utilising this wealth of information in real-time is crucial. Existing studies on continuous learning often propose models that necessitate buffering of lengthy sequences, potentially impeding the responsiveness of the inference system. Furthermore, these models are typically tailored for regularly sampled sequences, an assumption that is often unrealistic in real-world scenarios. To address these challenges, we introduce ODEStream: a Buffer-Free Online Learning Framework with an ODE-based Adaptor for Streaming Time Series Forecasting. This novel approach utilises the power of neural ODEs. ODEStream exhibits the capability to adapt to data irregularity and concept drift issues without reliance on complex frameworks. Concept drift refers to the phenomenon where the statistical properties of the target variable, which the model aims to predict, change over time due to various factors, such as changes in the underlying data distribution or evolving relationships between input and output variables. Our method mitigates performance degradation over time by learning how the dynamics of the sequence change, providing a streamlined solution for real-time analysis.
The proposed methods were evaluated on various important tasks and real-world benchmark datasets and have been compared against recent state-of-the-art studies in the field.</p
Microdiamond-embedded Silk: A Hybrid Platform for Temperature Monitoring in Wounds
No full textTimely determination of infection within wounds is crucial for avoiding spread of infectious bacteria, which can lead to further complications including sepsis. Current clinical methods of determining infection, while accurate, are intrusive and painful. Alternative visual identification methods, though less invasive, may lack precision. Both methods require the removal of the wound dressing, which is painful for the patient and time consuming for the clinician. However, early indications of infection can be inferred through monitoring of local area temperatures within a wound, which may show an increase of 3 to 5 deg C when an infection is present.
This study proposes a novel approach that integrates temperature sensors within a biocompatible scaffold, creating a hybrid wound dressing for non-contact thermometry of a wound site, providing a method of infection inference without the need for dressing removal. Unlike reported wound dressing sensing methods in the literature, the method for temperature inference does not require complex electronics, and maintains characteristics of a modern wound dressing to ensure ideal healing conditions. This differentiates it from other reported wound dressing sensing methods in the literature which frequently utilise flexible PCB methods with onboard communication systems to interface with, and do not provide fluid absorption or permeability capabilities that promotes an ideal wound healing environment.
The reported hybrid dressings consist of two main components; optical sensors which are embedded within a biocompatible and transparent silk fibroin scaffold. The utilised optical sensors are micron-sized fluorescent diamond which have been irradiated and processed to contain highly photostable and temperature sensitive nitrogen-vacancy centres.
Local temperature is measured through the fluorescence-based method of Optically Detected Magnetic Resonance (ODMR), enabling rapid measurement of temperature through the transparent silk dressing to infer infection without the need for dressing removal. These sensors are embedded within a biocompatible and biodegradable silk scaffold, that is surface conformable, low cost to manufacture, and can be modified to match the surface properties of commercial wound dressings.
The mechanical and surface properties of the fabricated dressings were compared with commercial dressings,indicating suitability for low-exudate environments akin to clinical ‘second-skin’ products.
The temperature sensing capabilities of the wound dressing were studied in an artificially heated environment as well as an in-vivo context to determine the feasibility of such a sensing platform. These studies determined that temperature sensing utilising this method on a biological system can be completed, however methods for optimising the measurement system for inaccuracies due to inherent optical heating are required. Methods to mitigate this were explored and implemented.
In summary, this work has presented a wound dressing that is capable of non-invasive temperature monitoring while maintaining ideal healing conditions. This protocol for fabricating such a dressing has been developed, and its sensing capability benchmarked in an in-vivo context. However, improvements of the optical measurement system to record temperature measurements spatially are still to be completed.
Future work on the sensing platform will focus on improving the optical measurement system for spacial measurement of temperature, and integrate a pH sensitive material into the dressing for multimodal sensing.</p
The Effects of Toxicants on the Health of Victorian Waterbirds
No full textAnthropogenic disturbances are increasingly threatening the health of wildlife populations globally. One such threat is chemical pollution. With the advent of the Anthropocene, negative impacts from pollution are escalating in all ecosystems. Indeed, the United Nations has recently declared pollution to be the third global catastrophe (along with climate change and biodiversity loss). However, the urgency of this threat has not always been widely appreciated in Australia.
This knowledge gap was addressed in this thesis by providing a contemporary account of exposure to pollutants in Australian waterbirds. This was achieved by opportunistically sampling birds from across the state of Victoria and focusing on three species: Pacific black ducks (Anas superciliosa), grey teal (Anas gracilis), and black swans (Cygnus atratus). I examined two contaminant types in these species: heavy metals (lead, copper, chromium, iron, manganese, mercury, and zinc) and persistent organic pollutants (POPs), and five tissue types: bone, feather, muscle, blood and adipose. The physiological impacts of these pollutants on birds were explored by characterising associations with metabolites in muscle and biochemistry parameters in blood. One key finding of this thesis is that exposure levels to critical contaminants are currently not as high in Australian ducks as in equivalent species inhabiting heavily industrialised areas of other continents. Additionally, there was little evidence that the quantified toxicants had obvious deleterious effects on the health of Victorian waterbirds.
This research project has provided new information relevant to waterbird ecotoxicology in an under-studied global region. It has also allowed me to develop several key skills in field biology, laboratory methods, statistical analysis, and science communication. The crucial recommendation for future research is to conduct further investigations on other environmental contaminants in other Australian ecosystems and taxa.</p
Patterns of Belief: Examining the Epistemologies of International Development Workers in Timor-Leste
No full textPatterns of Belief: Examining the Epistemologies of International Development Workers in Timor-Leste
This thesis contributes to our understanding of the processes of development and the socio-political relationships that are established through its practice. Based on research engagement via interviews and other supporting methods, this thesis examines the epistemological assumptions of international development workers operating in Timor-Leste. It argues that the modern rationality that dominates their epistemological assumptions is evident in what they believe knowledge to be, including its origins and the ways in which it can be acquired and assessed during the process of development work.
In addition to the central claim, three subsidiary arguments are integrated across the work of this thesis. The first subsidiary argument is that development work is understood as a ‘knowledge act’ dominated by a form of modern rationality – that is, a pattern of belief that gives authority to analytic derivation and the projection of verifiable, constructive, linear, and universalised claims. The second subsidiary argument is that the patterns of belief associated with modern rationality tend to produce knowledge hierarchies in which analytically derived claims are perceived as being more reliable and thus more universally applicable than other forms. This argument takes the idea of a ‘knowledge act’ and begins to demonstrate the social significance of the dominance that is given to certain epistemological forms over others. The third subsidiary argument draws on the first and second subsidiary claims to demonstrate how the beliefs associated with modern rationality tend to support the mobility of the worker because these understandings enable them to move across a wide variety of settings while abstractly generating knowledge about, as well as applying knowledge to, them.
While concentrating on the epistemologies of international development workers, this thesis is situated within a broader set of debates that continue to play out in Timor-Leste with respect to the epistemic framing of interventionist activities such as development and their capacity to generate the traction needed to meet many of its pre-set targets. It also reflects upon the broader shifts that have occurred within Critical Development Studies since the 1980s together with the Anthropology of Development literature and the so-called Actor-Oriented or Interactionist Frameworks. In so doing, the thesis draws attention to the ways in which international workers tend to construct a mode of practice that replicates much of their own epistemological assumptions in terms of both their day-to-day practices and their approach to scenarios where customary patterns of belief remain important to how people ‘know’ and understand the world. Critically, this thesis does not focus on the entanglement between various ways of knowing; instead, it focuses on those spaces that have been significantly bereft of analysis to date, namely the epistemological beliefs held by ‘malae’ (foreigners) in Timor-Leste, and how they shape and inform their attempts at social change.</p
Influenza-Induced Remodelling Accelerates the Development of Non-Small Cell Lung Cancer
No full textLung cancer, particularly non-small cell lung cancer (NSCLC), remains a leading cause of
global mortality. While smoking is an undeniable risk factor, the underlying mechanism of
NSCLC development remains unknown, leaving us with limited therapeutic options. The
prevailing model of tumourigenesis assumes a stepwise accumulation of genetic alterations,
each of which confers a selective growth advantage by activating a specific signalling pathway
that drives cancer progression. However, the abundance of genetic alterations observed in both
healthy individuals and non-tumour cells in cancer patients suggests a more nuanced picture,
where additional factors like the tumour microenvironment (TME) play a role. We propose that
injury-induced remodelling promotes changes in the lung microenvironment that promotes the
expansion of pre-existing mutated clones. Notably, influenza infection, which affects roughly
one billion people annually and is known to cause lung remodelling, has shown potential links
to increased lung cancer risk. However, the precise impact of influenza on cancer progression
remains largely unknown.
The TME is a complex network of cells surrounding a tumour, including stromal cell types that
can significantly influence tumour cell growth. Despite this, our understanding of the specific
cell types within the TME and their role in NSCLC development remains limited. This lack of
detail extends to the term "cancer-associated fibroblasts" (CAFs), which encompasses a diverse
population of stromal cells within the TME with functions that can either promote or suppress
tumour growth. By investigating how acute influenza A virus (IAV) infection alters the
composition of the TME, we aim to gain insights into its potential influence on NSCLC
progression, particularly the possibility of an injury-induced tumour promotion stage in lung
tumour formation. To achieve this, we utilised mice with genetically engineered floxed alleles
of the Lkb1 and Pten genes, which are frequently mutated in NSCLC. This model allows us to
investigate whether influenza infection can accelerate NSCLC development in cells with pre
existing mutations and explore the interplay between influenza-induced lung injury, TME
development, and tumour progression.
The first study used a mild strain of influenza to induce aberrant epithelial remodelling in a
mouse model of acute lung injury. Our goal was to identify the role of stromal cell subtypes in
driving this phenomenon. We discovered that a mild strain of influenza led to aberrant
epithelial remodelling, which was accompanied by significant alterations in the stromal
compartment of the lung, persisting for at least 21 days post infection (dpi). As previously
described, the epithelial remodelling was characterised by the bronchiolisation of the alveolar
epithelium with dysplastic expansion of basal cells and differentiation of ciliated cells and
goblet cells in the lung parenchyma. This was also associated with extensive collagen 1
deposition and fibrotic remodelling indicative of maladaptive stromal cell activation. We
developed a flow cytometry strategy to subset stromal cell populations from the mouse lung
and identified unique epithelial supportive stromal subsets, as demonstrated by their ability to
support epithelial growth in vitro using a 3D lung organoid culture system. These include a
FAPα+ resident mesenchymal stem cell population (rMSCs) and a highly supportive IL11+
population we termed auxiliary fibroblasts. In situ, we observed FAPα+ and IL11+ cells
localised to areas of aberrant epithelial remodelling and fibrosis 21 dpi following influenza
infection. Of note, only animals that had prolonged epithelial remodelling displayed fibrosis
with co-localisation of high numbers of FAPα+ and IL11+ cells. This study suggests that stromal
cell subsets may be key players in promoting stem cell-mediated regeneration and remodelling
after influenza-induce lung injury. Interestingly, the stromal cell subsets that we identified in
this study share striking similarities with CAF subtypes that are involved in chronic lung
diseases and NSCLC.
We next looked at the impact of influenza infection on tumour promotion using our transgenic
mouse model of NSCLC, which involved deletion of floxed Lkb1 and Pten genes in the lung
by intranasal delivery of adenovirus expressing Cre recombinase. This study showed that
influenza infection can accelerate NSCLC development from lung epithelial cells with pre
existing mutations in Pten and Lkb1 (genes known to be associated with human NSCLC) as
well as tumour progression (increased size). Non-small cell lung cancer (NSCLC) can be sub
divided into squamous cell carcinoma (SCC) and adenocarcinoma (ADC). In this model we
observed heterogeneity in tumour phenotypes, with smaller tumours being predominately a
mucinous ADC phenotype and larger tumours being a mixed or SCC phenotype. Additionally,
in our long-term study control group, we discovered that mild influenza infection can lead to
chronic aberrant epithelial remodelling persisting up to 40 weeks post infection (wpi), a
previously unknown finding that supports the concept that early life events can have significant
impact on lung health later in life.
The final study examined the relationship between influenza infection, FAPα+ and IL11+
stromal cells and the establishment of a tumour promoting TME. Immunostaining of lung tissue
sections from the Lkb1/Pten-deficient NSCLC mouse model revealed that FAPα+ and IL11+
stromal cells localised to the leading and developing edges of tumours at all stages of
development. Notably, their location differed from pro-fibrotic myofibroblast populations
marked by α-smooth muscle actin (ASMA). These results suggest that FAPα+ and IL11+
supportive fibroblasts (rather than ASMA+ cells) are key players in establishing a tumour
promoting TME.
This research has identified novel stromal cell populations in the lung that, in response to
influenza induced injury, play a critical role in establishing both a regenerative
microenvironment and a tumour promoting TME. These findings represent a significant
breakthrough in understanding the link between influenza infection and lung cancer
development. By focusing on the TME and specific stromal cell populations involved in
epithelial regeneration, we have shed light on the underlying mechanisms at play in lung tissue
regeneration, remodelling and tumour formation. Furthermore, this work underscores the
potential long-term consequences of lung infections, including promotion of lung cancer.</p
Stochastic Geometric Information Fusion for Cooperative Driving
No full textThe integration between sensors and microprocessors in vehicles has become increasingly tighter with technological advancements. However, each sensor's field of view (FoV) still poses a challenge. Multiple vehicles running local multi-object tracking (MOT) filters can exchange information and fuse it to overcome this. While Bayesian approaches have been the traditional method, newer methods such as random finite set (RFS) approaches, including the probability hypothesis density (PHD) filter and labelled multi-Bernoulli (LMB), demonstrate increased tracking accuracy in complex and dynamic scenarios with multiple targets. This thesis comprehensively explores using RFS-based multi-object tracking filters within an intelligent transport systems (ITS) application. It presents solutions to the significant issues preventing such a system from seeing future adoption. Due to its fundamental advantages, this project advocates using the LMB filter for all ITS scenarios. As centralized methods are optimal but not scalable for large vehicular networks, the usage of distributed networks is explored to create a scalable solution to cooperative information fusion within the ITS framework using multiple LMB filter nodes. The solution to a mathematically correct fusion method is investigated by modifying the generalized covariance intersection (GCI) rule to enable cooperative fusion with limited FoVs. Finally, we present a method to incorporate object class into the LMB filter, allowing superior tracking performance and extending it for use in fusion schemes.</p
Effects of Biochar on Methane Production and the Microbial Community in the Anaerobic Digestion of Chicken Manure
No full textThe production of organic waste necessitates proper management to mitigate and minimise potential environmental impacts. Anaerobic digestion has been recognised as a sustainable alternative to organic waste management for volume reduction, while achieving substantial greenhouse gas (GHG) emissions savings. The methane produced during this process also serves as a valuable source for generating electricity. Recognising the necessity for
Australia to move towards a carbon-neutral future, employing anaerobic digestion technology
for treating its abundant organic waste, which amounted to 30 million tonnes in 2017, becomes ever more crucial. The overarching aim of this thesis was to develop and validate a more efficient way to anaerobically digest a major organic waste, chicken manure, with a focus on the intricacies of microbiology to better understand and optimise the underlying processes.</p
Understanding Cranial Injury – Virtual Forensics
No full textIn the essence of firearm-related crime scene investigation and its resolution in court, one of the critical aspects is to distinguish the cause of mortality between homicide and suicide. This context can be determined by investigating the nature and conditions leading to the injury and accessing viable biomechanical scenarios. In cases of cranial injuries, a comprehensive understanding of principal biomechanics, biomaterial dispersion, and wounding mechanisms can promote the development of valid methods to distinguish the cause of fatality.
The virtual crime scene investigation employing numerical models holds significant potential to assist in forensic investigations of firearm-related fatalities, particularly in cases where ethical concerns and resource constraints limit physical experiments. Furthermore, traditional experiments involving animal models and human cadavers have limitations in investigating various factors influencing different cranial ballistic injuries, which require repetitive experimental conditions, unlike biomaterials. However, research on ballistic impacts on accurate numerical-based human cranial models using three primary anatomical components (skin-skull-brain) is still scarce. This includes the usage of appropriate biomaterials that replicate the properties of the human scalp and their properties at dynamic strain rates, which are essential for ballistic applications, thus hindering the reliability of numerical models.
Therefore, there is a need for appropriate models to investigate cranial ballistic responses. The present research aims to i) develop suitable human scalp biomaterials across a wide range of dynamic loadings and ii) enhance understanding of cranial ballistic responses through numerical and experimental approaches involving various ballistic factors. The human cranial numerical model developed in this study integrates three primary layers: skin, skull, and brain, with dynamic properties of biomaterials, addressing previous limitations in incorporating high strain rate skin biomaterial properties. The present study also develops and characterises silicone-based composite skin biomaterials with short polyethylene fibre and bioactive glass reinforcements across a range of strain rates through the micromechanics design of composites. The experiment and numerical analysis reveal their suitability for both quasi-static conditions and high-speed dynamic events. Notably, the study identifies a 3% reinforcement composite as the optimal skin simulant compared to pure silicone, and it also demonstrates a large variation in material stiffness with strain rate. The skin biomaterial subjected to a strain rate of 4000 s-1 demonstrates a 9-time greater stiffness than the same biomaterial subjected to a quasi-static strain rate of 0.48 s-1. This variation underscores the necessity of incorporating high-strain rate properties to yield accurate cranial injury simulations in ballistic impact scenarios.
The importance of considering the biomaterial strain rate becomes evident when employing different skin material properties in the numerical human cranial head model impacted by a 9mm projectile, where the skin with quasi-static properties exhibits excessive cranial injuries compared to its physical counterpart. In contrast, using high-strain rate properties yields results similar to those of the existing experiments. Additionally, the necessity of incorporating blood vessels and pressure was determined to be insignificant through the micromechanics finite element analysis, and a simplified human cranial model without a circulatory system was deemed sufficient.
The simulation results of ballistic impact on the human cranium demonstrate a correlation with ballistic experiments conducted on physical counterparts and animal models, establishing the reliability of the numerical model. The study establishes relationships between post-ballistic impact cranial injuries, such as wound diameter, wound shape, and temporary cavity, and various ballistic factors, namely impact location, projectile velocity, and angle of impact. This analysis facilitates the elimination of specific crime scene situations from reassessment consideration.
The research offers comprehensive numerical and experimental approaches to study cranial biomechanics and injury mechanisms that generate primary evidence in crime scenes. It integrates detailed anatomical human cranial geometry, including skin, skull, and brain components, and employs biomaterials closely resembling real human skin properties. This study constitutes the first analysis of various ballistic factors using an anatomical human cranial simulation, providing valuable insights into the interplay between ballistic factors and resultant cranial injuries. Moreover, it stands as the inaugural study characterising a specific composite skin biomaterial that integrates both mechanical and bio-integrative properties. The comprehensive exploration of strain rates highlights the importance of adopting high-strain rate material properties in dynamic simulations.
Overall, this research advances the understanding of cranial biomechanics and injury mechanisms, with implications for composite biomaterials design, impact dynamics, and forensic investigation.</p