Search

Recently, carbon fibre reinforced thermoplastic composites (CFRPTs) have received increasing interest in various industries, particularly in aerospace, due to their good recyclability/reparability and high production turnover. Carbon-fibre-reinforced-polyetherketoneketone (CF/PEKK) stands out as a highly promising CFRTP material due to its outstanding mechanical properties/thermal stability. Creating high-quality holes in CF/PEKK for aircraft assembly through riveting remains a challenge due to the material's heterogeneous structure and thermal sensitivity. To date, the hole making performance and damage mechanisms of CFRTPs, particularly CF/PEKK, remain a significant but understudied area. In this work, the hole making performance of CF/PEKK has been investigated for the first time. The impact of different hole making conditions (conventional drilling and helical milling) and effects of drilling parameters (feed rate, spindle speed) on the resulting thrust force, temperature and chip morphology, hole delamination/microstructural damage and material removal mechanisms, have been thoroughly investigated and wherever possible, compared with that of the conventional CF/epoxy. Multi-objective optimization has been conducted to identify the optimal drilling parameters for CF/PEKK. In addition, temperature controlled orthogonal cutting was carried out to reveal the fundamental temperature-dependent cutting physics of CF/PEKK. This PhD project has created a significant body of new knowledge in relation to CFRTPs machining process and the associated cutting physics, which is expected to provide important guidance for low-damage, high-efficiency machining of CFRTPs.

The reduction of atmospheric reactive N (Nr) and S (Sr) species is a key objective for air quality control policies because they contribute to the formation of PM2.5 (particulate matter with aerodynamic diameter ≤2.5 μm), which have significant effects on human health and climate, and their deposition affect ecosystem biodiversity. As a range of emission sources and atmospheric chemical and physical processes contribute to Nr and Sr concentrations, atmospheric chemistry transport models (ACTMs) are an essential tool to identify important processes controlling their impacts and effective mitigation. Here, the EMEP MSC-W ACTM coupled with WRF meteorology at 1º×1º resolution is used. The chemical climate for Nr and Sr pollution has changed dramatically in the past two decades, it is therefore necessary to update our understanding of global Nr and Sr chemistry and investigate their mitigation under current atmospheric state. Thus, the aims of the thesis are (i) to evaluate the global model simulation of reduced N (RDN), oxidised N (OXN), and oxidised S (OXS) against ambient measurements; (ii) to analyse current global and regional budgets, fluxes, and lifetimes of RDN, OXN, and OXS, quantifying important processes controlling their regional distributions in the atmosphere; and (iii) to quantify the sensitivities of emissions reductions for the mitigation of PM2.5, Nr and Sr species. Firstly, the evaluation of the EMEP MSC-W model is conducted both spatially and temporally, covering 10 monitoring networks worldwide measuring surface concentrations and wet deposition. Model simulations for 2010 compared use of both HTAP and ECLIPSEE (ECLIPSE annual total with EDGAR monthly profile) emissions inventories; those for 2015 used ECLIPSEE only. Simulations of primary pollutants are somewhat sensitive to the choice of inventory in places where regional differences in emissions between the two inventories are apparent (e.g., China), but much less so for secondary components. Comparisons of 2010 and 2015 annual concentrations between model and measurement demonstrate that the model captures well the overall spatial and seasonal variations of the major inorganic pollutants NH₃, NO₂, SO₂, HNO₃, NH₄⁺, NO₃⁻, SO₄²⁻ , and their wet deposition in East Asia, Southeast Asia, Europe, and North America. The model shows better correlations with annual average measurements for networks in Southeast Asia (Mean R for 7 species: /R₇ = 0.73), Europe ( /R₇ = 0.67) and North America ( /R₇ = 0.63) than in East Asia ( /R₅ = 0.35) (data for 2015), which suggests potential issues with the measurements in the latter network. The evaluation of wet deposition shows that the greatest consistency is in North America (R: 0.75-0.82), followed by Southeast Asia (R: 0.51-0.68), Europe (R: 0.61-0.64), and East Asia (R: 0.13-0.59). Model-measurement bias varies between species in different networks. The greater uniformity in spatial correlations than in biases as revealed in this work suggests that the major driver of model-measurement discrepancies (aside from their differing spatial representativeness and uncertainties in measurements) are shortcomings in absolute emissions rather than in modelling the atmospheric processes. Secondly, the EMEP-WRF model is used to undertake a present-day (2015) global and regional quantification of Nr and Sr concentrations, fluxes, and lifetimes, which are quantities that cannot be derived from measurements alone. In areas with high levels of RDN (NH3 + NH4+), OXN (NOx + HNO3 + HONO + N2O5 + NO3- + Other OXN species), and OXS (SO2 + SO42-), RDN is predominantly in the form of NH3 (NH4+ typically < 20%), OXN has majority gaseous species composition, and OXS predominantly comprises SO42- except in areas near major SO2 sources. Most continental regions are now 'ammonia rich', and more so than previously, which indicates that whilst reducing NH3 emissions will decrease RDN concentration it will have little effect on mitigating secondary inorganic aerosol (SIA). South Asia is the most ammonia-rich region. Coastal areas around East Asia, northern Europe, and north-eastern United States are 'nitrate rich' where NH4NO3 formation is limited by NH3. These locations experience transport of OXN from the adjacent continent and/or direct shipping emissions of NOx but NH3 concentrations are lower. The least populated continental areas and most marine areas are 'sulfate rich.' Deposition of OXN (57.9 TgN yr-1, 51%) and RDN (55.5 TgN yr-1, 49%) contribute almost equally to total nitrogen deposition. OXS deposition is 50.5 TgS yr-1. Dry deposition of NH3 is the largest contributor to RDN deposition in most continental regions except for remote areas where NH3 emissions are small and RDN deposition is mainly determined by transport of NH4+. The two largest contributors to OXN deposition in all regions are HNO3 and coarse NO3- (via both wet and dry deposition). The tropospheric lifetime of NH3 (1.6 days) is much shorter than that of NH4+ (8.9 days), consistent with a global NH4+ burden (68% of total RDN burden) almost double that of NH3 (32%). Fine NO3- only constitutes 10% of global nitrate burden, albeit fine NO3- dominates in eastern China, Europe, and eastern North America. It is therefore important to account for contributions of coarse nitrate to tropospheric nitrate budgets. Lifetimes of RDN, OXN, and OXS species vary by a factor of 4 across continental regions with East and Southeast Asia generally having the shortest lifetimes. South Asia is the largest net exporter of RDN (2.21 TgN yr-1, 29% of its annual emission) and OXS (1.62 TgS yr-1, 37%). Africa is the largest net exporter of OXN (1.92 TgN yr-1, 22%). Despite having the largest RDN emissions and deposition, East Asia has only small net export and is therefore largely responsible for its own RDN pollution. Finally, the sensitivity of Nr, Sr, and PM2.5 to 20% and 40% individual and collective reductions in anthropogenic emissions of NH3, NOx, and SOx relative to the 2015 baseline is investigated. Regional comparisons reveal that the individual emissions reduction has multiple co-benefits and small disbenefits on different species, and those effects are highly geographically variable. A 40% NH3 emission reduction decreases regional average NH3 concentrations by 47-49%, but only decreases NH4+ by 18% in Euro_Medi, 15% in East Asia, 12% in North America, and 4% in South Asia. This order follows the regional ammonia-richness. A disbenefit is the increased SO2 concentrations in these regions (10-16% for 40% reductions) because reduced NH3 levels decrease SO2 deposition through altering atmospheric acidity. A 40% NOx emission reduction reduces NOx concentrations in East Asia by 45%, Euro_Medi and North America by ~38%, and South Asia by 22%, whilst the regional order is reversed for fine NO3-, which is related to enhanced O3 levels in East Asia (and also, but by less, in Euro_Medi), and decreased O3 levels in South Asia (and also, but by less, in North America). Consequently, the oxidation of NOx to NO3- and of SO2 to SO42- is enhanced in East Asia but decreased in South Asia, which causes a less effective decrease in NO3- and even an increase in SO42- in East Asia, but quite the opposite in South Asia. For regional policy making, it is thus vital to reduce three precursors together to minimise such adverse effects. A 40% SOx emission reduction is slightly more effective in reducing SO2 (42-45%) than SO42- (34-38%), whilst the disbenefit is that it yields ~12% increase in NH3 total deposition in the four regions which further threatens ecosystem diversity. This work also highlights important messages for policy-makers concerning the mitigation of PM2.5. More emissions controls focusing on NH3 and NOx are necessary for regions with better air quality such as northern Europe and eastern North America. In East Asia, the three individual reductions are equally effective, whilst in South Asia only SOx reduction is currently effective. The geographically-varying non-one-to-one proportionality of chemical responses of Nr, Sr, and PM2.5 to emissions reductions revealed by this work show the importance of both prioritising emissions strategies in different regions and combining several precursor reductions together to maximise the policy effectiveness. In summary, the comprehensive model evaluation, as the opening study of this thesis, supports the application of this model framework for analysis of current Nr and Sr budgets (second study) which then provides theoretical explanations for responses of Nr and Sr to potential emission controls (third study). By demonstrating the model's capability of simulating global atmospheric chemistry and transport and presenting an update of current atmospheric state in terms of particle formation, this thesis provides useful suggestions for global and regional policymakers to mitigate PM2.5 pollution and reduce N and S deposition.

This thesis presents a longitudinal investigation of spontaneous speech by child Mandarin heritage speakers (n=6, 2;00−5;02) in an English-speaking country with a focus on the acquisition of several syntactic phenomena: wh-questions, object realization, sentence final particles (SFPs) and code-switching. The study on heritage speakers' languages could contribute to both L1 and L2 language acquisition since they acquire heritage language as L1 and may show some similarities with L2 learners in the acquisition of social majority language. Furthermore, as a type of bilingual that takes place in a specific sociolinguistic environment, their language development could also provide evidence for the debates among bilingual studies and shed more light on input factor. Mandarin and English are chosen to investigate since they are distinct from each other in several domains, which could provide a good testing ground for those hypotheses in language acquisition. Overall, in child Mandarin heritage speakers' languages, I only find a little of evidence for the transfer from Mandarin to English and then suggest that it might be owing to the societal environmental language support. Meanwhile, no transfer from English to Mandarin is found, which seems to prove the asymmetry in the direction. I also find that Mandarin heritage speakers use much less code-switching utterances than simultaneous bilinguals (Poeste et al., 2019) and thus suggest that it might be benefited from the relatively monolingual environment the heritage speakers received for each language. The result of this study also contradicts to the Interface Hypotheses (Sorace, 2011), since I find that heritage speakers are able to fully acquire Mandarin null objects and SFPs (e.g. SFP le, ma ne1, ba1 and a), which lie on syntax-semantics-discourse interface. However, these children are found to take longer time than monolinguals to acquire Mandarin wh-questions as well as the SFP ne2, which is in line with previous observations that reduced input in a language lead to a protracted acquisition. Moreover, I notice that they cannot properly use Mandarin SFP ba2 in questions, similar to that of adult Mandarin heritage speakers (Yan, 2020). Thus, I argue that this provides evidence of divergent attainment rather than attrition and agree that this problem is related to the limitation in processing resources (Polinsky and Scontras, 2020).

Peer-to-peer (P2P) lending creates an environment in which lenders and borrowers are directly matched without financial intermediary. Since the first P2P platform Zopa.com was founded in 2005, it has been one of the most emerging FinTech innovations. Transparency Market Research indicates that the potential market volume of global P2P market will be worth $897.85 billion by 2024. The unsecured loans funded by individual investors play profound roles in the alternative banking system. This doctoral thesis uses three empirical chapters to investigate the P2P lending market from both lenders and borrowers perspectives. The first chapter investigates expert bidding imitation in peer-to-peer lending platforms. Differs from herding behaviour which associates certain actions of an individual to those of the whole crowd, we question whether an individual's bidding behaviour is related to the decisions made by expert. The experts are defined as investors who either have more central roles or who spend more time or money on the network. We employ data from Renrendai.com, which contains information of about 170,000 investors who placed almost four million bids on 111,234 loan listings from 2010 to 2018. Our dataset suggests that an average investor mimics the bids of expert lenders. Inactive lenders learn top investors' lending behaviour through observational learning and then follow their actions, although they do not know the experts' identity. Finally, we show that experts rarely imitate other experts, yet they exhibit herding behaviour. The second chapter examines the reaction of individual investors to news arrival. In particular, we explore how Spanish COVID-19 (C19) information affects decisions of European investors holding Spain originated P2P loans. This study employs loan transaction data from Bondora secondary market, one of the leading P2P lending platforms in continental European. We find that asset holders react to the ongoing Spanish official C19 announcements by reducing asset prices. Also, the higher agreement on asset valuation between sellers and buyers is attributable to the ascending infections. Interestingly, investors process more country-wide C19 information compared to regional updates. This could be explained by the insufficient European media attention to Spanish region-level C19 topics. Moreover, the lockdown enhances the negative effects of C19 crisis. In addition, the dispersion of asset valuation between sellers and buyers is constricted in response to the ongoing cases. The third chapter uncovers the impact of public holiday on the investor attention on financing issues in P2P markets. Differs from typical financial markets such as stock and futures market, P2P market never closes. Therefore, P2P platforms create a unique environment which allows investors to trade and invest in public holidays. Using unique datasets obtained from Renrendai.com and Bondora.com, two prominent peer-to-peer lending platforms in China and continental Europe, our estimation suggests that the celebration of holiday increases the investor inattention as the investor decision time delays during public holidays. Also, the investment performance tends to decrease under limited investor attention. The mismatch on the asset valuation between sellers and buyers on asset valuation also expands during the holidays.

This thesis includes three essays in corporate reporting and auditing. Chapter 1 studies how institutional blockholdings affect firm voluntary disclosure. We document that concentrated institutional ownership reduces firms' voluntary disclosure measured by the propensity to issue management forecasts, comprehensiveness of guidance, propensity to engage in conference calls, and the number of 8-K filings. We identify two channels through which blockholders affect firms' voluntary disclosure. First, blockholders have easier access to managers and substitute private for public information acquisition. Second, a higher proportion of non-monitoring blockholders with low demand for voluntary disclosure, such as passive blockholders, reduces the firm's incentive to provide voluntary disclosure. The results are robust to endogeneity and reverse causality concerns. Our study identifies an important effect that concentrated ownership has on firm corporate disclosure. Chapter 2 examines how an increase in the audit market competition affects the incumbent accounting firms' audit quality. Our setting is a quasi-natural experiment related to the government-supported emergence of second-tier domestic accounting firms in China. We use this shock to perform a difference-in-differences test focused on the incumbent Big 4 firms' response to new competition, which we instrument by the industry-level variation in second-tier firms' market share change. We find that the audit fees and audit quality of Big 4 firms decreased in response to the increased competition from second-tier firms. The result of audit quality reduction is mainly driven by industries with lower initial Big 4 dominance. Our results highlight that increased competition, as captured by a larger number of audit firms, can reduce average audit quality, which contrasts the frequent regulatory opinion that more competition in the audit market will improve audit quality. Chapter 3 identifies a new channel through which audit market competition affects audit quality of incumbent Big4 audit firms-auditor to client assignment. We use the same setting as in Chapter 2 which is the government-supported emergence of second-tier domestic accounting firms in China as a shock to the incumbent Big 4 firms' audit market competition. To identify how the Big 4 re-assign auditors to clients in response to changes in audit market competition, we make use of the signing-auditor level data and measure the auditor's experience and workload. We find that Big 4 firms assigned less experienced auditors to the industries with high competition and burdened them with more workload. These auditor re-assignments help to explain the negative effect of competition on Big 4 firms' audit quality.

Scientists have developed a variety of ultrafast pump-probe experimental techniques to investigate gas-phase molecular structural dynamics over the past few decades. Amongst those techniques is Coulomb explosion imaging, which can provide direct information on nuclear positions. Coulomb explosion imaging takes advantage of ionisation-dissociation of gas-phase molecules and velocity map imaging of fragment ions. In our research group, an NIR laser was used to perform strong field ionisation on small molecules for Coulomb explosion imaging. A new laser beam system was built for this purpose, and a peak laser intensity of 1300 TW/cm² was achieved in the interaction area with molecules. Meanwhile, a theoretical methodology for simulating the Coulomb explosion dynamics of small molecules was developed. This simulation methodology is fast, accurate enough, and very informative. The Coulomb explosion dynamics of 1,2-dichloroethene, methyl iodide, and thiophenone simulated using this methodology are shown and discussed in detail in this thesis. The simulation results are also compared with the experimental results for 1,2-dichloroethene and methyl iodide to reveal more details about the Coulomb explosion dynamics of these molecules, and the comparison finds both similarities and discrepancies between the experimental and the simulation results. The experimental and theoretical work on 1,2-dichloroethene shows that the angular distributions of the ion images are the only thing to distinguish the cis- and trans-isomers. The experimental and theoretical results for methyl iodide suggest that there could be intramolecular charge transfers during the dissociation. The results for the simulations of Coulomb explosion dynamics of thiophenone suggest that several significantly different structures along the (photo-induced) ring-opening reaction pathway of thiophenone can be distinguished by examining the simulated asymptotic speeds of fragment ions, suggesting that Coulomb explosion imaging is potentially a technique for investigating ring-opening dynamics of organic molecules.

Over 100 genetic associations have been identified for Rheumatoid arthritis (RA) and although the enrichment and functional relevance of these associations in immune cells have been well established in recent literature, the same cannot be said for Fibroblast-like synoviocytes (FLS). This thesis maps the epigenomic landscape of FLS by integrating multi-omics data (ATAC-seq, Hi-C, CHi-C, RNA-seq, ChIP-seq), revealing transcriptionally active regulatory elements across the genome. Cross validation of these multi-omics datasets also revealed their interconnected nature in response to TNF stimulation. The epigenomic landscape of FLS confirms the functional relevance of FLS in RA pathogenesis, by demonstrating that FLS regulatory elements account for up to 24% of RA heritability. By overlapping fine-mapped RA associations with both FLS and publicly available epigenomes, the cell types which are functional active at these associations were identified. This also further refined the list of potential causal variants at each RA association. Each RA association was assigned putative target genes in FLS and immune cells, by providing multi-layered evidence using several multi-omics datasets available: each gene assignment to an association were in 3D proximity to the each other as demonstrated by Hi-C and/or CHi-C, and their activity was confirmed through overlapping analysis with histone modifications of activity (ChIP-seq), assigned genes were also significantly expressed, as demonstrated by RNA-seq. Overall, the results presented provide RA associated genes for further study/validation, and a useful resource for future epigenetic studies into FLS.

With the development of smart technologies built on big data analysis, the vision of smart city, that aims to improve quality of life, reduce energy consumption and pollution and drive economic growth, has been enhanced. One important component of this vision is the urban transport network which is continuously challenged with increasing number of vehicles on roads. This results in issues including long travel time, increasingly persistent traffic congestion, accidents and road safety concerns. The research work in this thesis aims to predict traffic in urban transport networks which are capable of solving those issues. Understanding and control of traffic patterns are fundamental towards this aim. For this, we develop a short-term traffic flow prediction model, named EM, on linear roadways based on machine learning technology. EM is able to analyse and extract spatial and temporal features from original traffic data for the final prediction. This short-term traffic prediction could give drivers traffic situation in advance and guide them to avoid congested roads to reduce travel time and also relieve traffic congestion. Moreover, armed with accurate short-term traffic prediction models on linear roadways, we further develop a novel deep learning model, named ALLSCP, for short-term traffic flow prediction on intersections where traffic situations are more complex compared to linear roadways. The difference to EM is that ALLSCP can analyse more detailed features (i.e. global- and local-spatial and short-, medium- and long-term temporal features) for the final prediction. Due to the fact that traffic flows are almost controlled by traffic lights in most cities around the world, this could benefit the optimisation of traffic light strategies so that more vehicles are allowed to pass in a short duration. After analysing traffic data on linear roadways and intersections, we consider and solve traffic prediction problem on large-scale road networks and develop a deep learning model (named SAGCN-SST) based on Graph Convolution Network (GCN) and Attention mechanism for multiple traffic speed prediction on large-scale road networks. This SAGCN- SST is able to capture dynamic-spatial and temporal features for the final prediction. The predicted traffic speed on large-scale road networks could be used to find vulnerable roads and then optimise routing strategies for reducing traffic congestion and long travel time. Furthermore, we design another deep learning model based on Virtual Dynamic Graph Convolution Network and Transformer with Gate and Attention mechanisms (VDGC- NTGA) for multi-step traffic speed prediction on large-scale road networks, which can explore dynamic and hidden spatial and temporal features in network-wide for the final prediction. VDGCNTGA has ability to generate and update a virtual dynamic road graph each batch to represent dynamic and hidden spatial dependencies that are not described in the real road graph. Compared to our last SAGCN-SST model, VDGCNTGA can exploit more useful hidden spatial dependencies of road segments in network-wide. Finally, we develop a deep learning model based on Sequence-to-Sequence architecture with an embedded module using Graph Convolution Neural Network and Transformer, named GCNT-Seq2Seq, for long-term traffic speed prediction in large-scale road net- works. This model is able to analyse and extract local- and global-spatial and long-term temporal features for the final prediction, which benefits the optimisation of traffic strategies for improving road network efficiency. For this research, we use real transportation data collected from urban transport networks of different characteristics to evaluate our proposed models and also compare them against well-known existing traffic prediction models.

Graphs are important data structures that can be found in a wide variety of real-world scenarios. It is well recognised that the primitive graph representation is sparse, high- dimensional and noisy. Therefore, it is challenging to analyse such primitive data for downstream graph-related tasks (e.g., community detection and node classification). Graph representation learning (GRL) aims to map graph data into a low-dimensional dense vector space in which the graph information is maximally preserved. It allows primitive graphs to be easily analysed in the new mapped vector space. GRL methods typically focus on simple connectivity patterns that only explicitly model relations between two nodes. Motif structures that capture relations among three or more nodes have been recognised as functional units of graphs, and can gain new insights into the organisation of graphs. Therefore, in this thesis we propose new GRL methods modelling motif structures for different graph-related tasks and applications along three directions: (1) a method to learn a spectral embedding space with both edge-based and triangle-based structures for clustering nodes; (2) a graph transformer by unifying homophily and heterophily representation for role classification and motif structure completion; (3) a method to tackle noises in knowledge graph representations with motif structures for recommendations and knowledge graph completion. Experimental studies show that the proposed methods have outperformed related state-of-the-art methods for targeted tasks and applications.

Maritime shipping, while acting as the most important role in the world transportation system, witnesses the boom of Operational Research (OR) that has been applied to support the decision making towards not only higher profitability from the financial perspective but also better sustainability in the sense of green shipping, due to the sharply fluctuations of fuel price and newly approved emission measures. However, due to the complexity of operational scenarios, existing speed optimisation modelling frameworks are developed on a case by case basis, and the preference between them is yet well-explained in literature. This has now caught the attention to some researchers in line with the needs of evaluating other related problems, e.g., environmental measures, from the perspective of profit-seeking. This thesis, in particular, adopts the Net Present Value (NPV) cash-flow approach that explicitly accounts for each payment made for the decision maker and it can also be used to study the classic modelling frameworks via the technique of NPV Equivalent Analysis (NPVEA) developed in literature. Using our NPV framework, this thesis established how to explicitly consider contracts details, e.g., for time charter contract and voyage-alike charter contract, into speed optimisation models, and how these affect optimal decisions and profitability for a ship operator (or the ship owner). The focus lies on decisions in the deterministic setting that maximises the decision maker's profitability. In particular, we look at the comparison between our proposed NPV modelling framework with classic speed optimisation models from both the mathematical analysis and the numerical experiments. Factors other than contractual details are considered in this thesis as well. In particular, the decision maker's expectations about future is modelled as the Future Profit Potential (FPP), which is the NPV of profits within the relavent future. We thus aim to demonstrate how the setting of FPP will impact the optimisation, and clarify what is the corresponding underlying assumption. Through this work, we also examine some of the selected topics that are marked in literature or yet studied in the context of maritime shipping to provide a modelling framework as well as managerial insights that what is important for the decision maker to consider.

Penetration testing (PT) is a well-established proactive approach to evaluating the security of digital assets by actively identifying and exploiting existing vulnerabilities. It is a widely used approach to improving the information security level of the target system. However, the use of PT has been restricted to advanced security experts who have many years of experience. Furthermore, the complex manual process is costly and time-consuming. Automation can significantly reduce the time, cost and human labour required in the stage of information gathering, analysis and exploitation. In terms of privacy protection, automated PT can prevent the leakage of sensitive information by human testers. To date, there has been little academic research on automated PT, and the field is still in its infancy in security. Many studies or implementations of automation merely map the results of vulnerability scanners to the corresponding exploitation tools. Most research treats PT as a planning problem expressed in terms of an attack tree, an attack graph, a planning domain definition language (PDDL) or a partially observable Markov decision process (POMDP). However, these solutions either cannot handle incomplete knowledge, uncertainty and a dynamic environment, or they exhibit poor scaling. The contribution of the thesis is to achieve real-time automation of PT based on the belief-desire-intention (BDI) model, and to validate the model, which can work well in the Internet of Things (IoT) environment. An ontology for PT was built based on semantic web rule language (SWRL) rules for knowledge reuse and better reasoning ability. The experiment results illustrate that the model's performance is better than the manual PT and other existing approaches.

The overarching aim of this dissertation is to unfold the practices of family businesses looking at the role of family involvement in leveraging the intangible aspects that can lead to success over time. Investigating the role of knowledge, the use of history, and the authenticity of strategy, which I explores throughout my PhD journey, I offer a novel and relevant contribution to family business literature. This dissertation consists of five chapters, including a compilation of three individual papers that advance the debate on intangible resource in the family business field. In chapter 1, I introduce inspiration for my chosen research, and provide an overview of the three papers, which are included in chapters 2 to 4. In chapter 2, I present my first paper, a literature review of knowledge management in the succession process, co-authored with Dr. Giovanna Campopiano. This paper has been submitted to the Journal of Knowledge Management for the Special Issue for Family Business. It is at the stage of resubmission after the first review. Chapter 3 contains my second paper - Mining the Past: Scripted History and Competitive Advantage in a Family Venture, co-authored with Professor Alfredo De Massis and Dr. Josip Kotlar. This paper has been submitted to entrepreneurship, Theory and Practice. It is also at the stage of resubmission after the first review. Chapter 4 contains a third paper on strategy authenticity in a medium-sized family business. This paper aims at a strategy-related journal. In chapter 5, I summarize and conclude the achievements of my research. In addition, I explain my future research plan. This includes one of my draft paper about family practices, which was presented in a PhD workshop in Toulouse School of Management, as well as a grant I am applying for a project in the next three years to come.

My dissertation research explored responder behaviors (e.g., demonstrating response styles, carelessness, and possessing misconceptions) that compromise psychometric quality and impact the interpretation and use of assessment results. Identifying these behaviors can help researchers understand and minimize their potentially construct-irrelevant impact on the psychometric quality of measurement procedures. In three studies related to response quality, we analyzed item response data in terms of subgroups and in terms of missingness levels respectively to evaluate the sensitivity of item response theory (IRT) indicators to response characteristics that can compromise psychometric quality; we also validated a diagnostic classification model (DCM) aiming to diagnose the presence of misconceptions. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

This thesis carries out a series of empirical investigations into the nature and evolutionary process of asset bubbles in global stock markets. It also provides insight into the issue of market predictability with the consideration of price bubbles in the US, and how those results might inspire policymakers to prevent future bubbles. We start by reviewing the rational bubble theories which are used for modelling bubble process and discuss the rationale of the relevant testing methods for discovering bubbles. Overall, three main testing procedures are selected in Chapter 3 with the purpose of concluding whether bubbles exist in the global stock markets. Eventually, we confirm the presence of bubbles globally, and provide clear dates for each bubble's origination and collapse. The bubble dates obtained provide a timeline for stock market exuberance, and their overlapping periods suggest that bubbles can migrate between countries. However, there has been very little research on this latter issue. Therefore, in Chapter 4 we undertake a large-scale empirical analysis to investigate the bubble transmission mechanism. Our vector autoregressive (VAR) and volatility results confirm that for some countries a contagion effect exists, leading to bubble migration between countries. Finally, in Chapter 5, we are particularly interested in whether empirical results on the predictability of stock market data by the dividend-price ratio is affected by the presence of a bubble, and by borrowing Campbell-Shiller's model but adding selected monetary variables, we further assess the forecasting performance of common monetary policy indictors in predicting the movement of price-dividend ratios in both bubble and non-bubble periods.

The problems caused by urban congestion and population growth result in a growing need to construct railway networks in urban areas due to the advantages of safety, efficiency and reliability over alternative mass transportation options. However, railway systems can cause environmental issues due to the transit of vehicles, including railway noise and vibration. The problem of ground-borne vibration has gained increasing attention because people are becoming less likely to tolerate environmental disturbance and sensitive buildings and equipment could be significantly affected by nearby groundborne vibrations. Therefore, it is necessary to investigate those effects and provide a better understanding of ground-borne vibrations from railways. This research work was primarily concerned with the transmission of vibration from railways via a single friction pile through the soil-pile interface and the ground. In this study, dynamic centrifuge tests were conducted to study the dynamic pile behaviour, the soil-pile interaction and ground-borne vibration propagation characteristics generated from the vertical oscillation of a single pile in sand. The use of the centrifuge is attributed to its great advantage of reproducing the full scale stress field within a small scale model, thus correctly modelling soil behaviour and soil-pile interface characteristics. However, the boundary effects in centrifuge models on vibration propagation have not been fully understood. In addition to the boundary effects, centrifuge and shaker induced vibrations and the embedded piezoelectric accelerometors can also affect the measured vibration results in a centrifuge test; these were also considered in this research project. The main aims of this study were to provide insights into the soil-pile interaction and ground-borne vibrations in sand from the vertical oscillation of a single pile and to examine the use of centrifuge modelling as a tool for investigating this problem. To achieve these aims, centrifuge tests were performed to simulate the ground-borne vibration by applying small vertical harmonic loads to the top of a single rough pile (with sand stuck to the pile surface) embedded in sand. The effects of vibration absorbing material used in the centrifuge tests, Duxseal, on the soil stress field and soil stiffness were investigated experimentally by measuring the shear wave velocity of the soil using centrifuge based air hammer tests and triaxial based bender elememt tests. A series of numerical models were developed in FLAC3D to further the understanding of soil-pile interaction under vertical harmonic loads and to examine the effects of Duxseal and piezoelectric accelerometers on the centrifuge measurements. A series of combinations of dead loads and dynamic loads were used to investigate their effects on pile and soil behaviours. Three dead loads with a safety factor (ratio=ultimate bearing capacity of the pile/dead load) of approximately 8, 4 and 2 were employed in this study to investigate the effect of the static load amplitude on the dynamic pile and soil behaviour. The dynamic loads with two different amplitudes were applied by an electromagnetic shaker for each dead load. Centrifuge test results demonstrated that both dynamic pile and soil behaviour were in the linear range when the safety factor was greater than 2. Under all the combined forces applied in this study, the soil and the pile experienced the same displacement at the interface and the dynamic load applied to the pile top did not result in accumulated pile displacement. Therefore, the soil-pile interface can be treated as a perfectly bonded interface for the setup used in this study. Under the three dead loads (safety factor > 2), the pile top response amplitude indicated the pile top response decreased with the increasing dead load. It was also found that the pile was compressible and the displacement of the pile element under vertical harmonic loading was purely caused by the axial elastic deformation. The axial load within the pile decreased with depth, thus resulting in a decreasing axial elastic deformation of the pile and soil particle displacement along depth at the soil-pile interface. It was found that the vibration level at the same horizontal distance from the pile center decreased almost linearly with depth. The damping ratio of soil along depth was calculated by an empirical equation and the logarithmic decrement method using data measured by the accelerometers. It was found that within very small shear strain (less than 1.0 × 10−6 ), the material damping had a slight effect on the vibration attenuation at three depths. The vibration attenuation along radial distance from the pile center was evaluated by comparing the ratio of vibration responses measured by two accelerometers with the same embedded depth. It was found that test results agreed well with the theoretical estimation at most frequencies. The vibration attenuation curves at three depths (3.36 m, 10 m and 16.8 m at prototype scale) were calculated. It was found that within the radial distance of 15 m (prototype scale), the vibration attenuates very quickly. The curves tend to be flat when the radial distance from the pile is greater than 27 m (prototype scale). The original purpose of the use of Duxseal was to absorb vibrations and reduce reflections from the rigid boundaries of the centrifuge container so that more accurate model response can be measured in the centrifuge. However, it was found that Duxseal can change the lateral stress in the soil and the soil stiffness due to soil behaviour being stress dependent. It is suggested that side Duxseal should not be used to investigate the ground-borne vibration because the soil-pile interaction and measured ground-borne vibrations were affected and the Duxseal was not able to eliminate reflections. The effects of piezoelectric accelerometers embedded in the soil were examined by comparing the results from two FLAC3D models: one with piezoelectric accelerometers embedded in the soil and one without piezoelectric accelerometers. It was observed that calculated results with and without the piezoelectric accelerometers were close to each other at most frequencies. This indicated that the presence of piezoelectric accelerometers did not significantly change the model response.

Conflicts are inevitable in work teams. In diversity research, early studies have analysed the distribution of personal attributes among interdependent members of a working unit and investigated the different types of conflicts that are triggered by individual differences within the organisation. This body of research investigates the impact of the different dimensions of personal attributes on team behaviour (i.e., conflicts and cooperation) and explores how conflicts may have an impact on the group’s performance. Since the impact of such conflicts on team performance has not yet been proven, some studies have tried to figure out the conflict-performance relationship by analysing the mediator impacts; for example, group cohesion and group behavioural integration. Thus, the research focus has changed from analysing individual dynamics (i.e., personal characteristics) to analysing group dynamics (i.e., the relationships and interactions among the diversified members). The analysis of different levels, including the individual, group, organisation, industry and environment dimensions, makes conflict management research more integrated and dynamic in comparison with studies that examine the impacts of the behaviour of isolated individuals within a working team. The recent focus on group behaviour and the composition of group members has led to the concepts of subgroups and faultlines in conflict management research. Subgroups or faultlines are regarded as a central component of work teams; however, subgroups or faultline issues have remained largely unexamined by scholars. Previous studies have presented a typology of subgroups and examined the antecedents and consequences of subgroups. Faultlines are regarded as the antecedents of subgroups in the literature. They are hypothetical dividing lines that split a group into two or more subgroups based on the alignment of one or more individual’s attributes and they have been found to affect teams’ processes, teams’ performance and teams’ affective outcomes. Most faultline studies are interested in understanding the composition of faultlines and they focus on the demographic attributes of team members. Recent studies have shown an increasing interest in analysing the complex mix of attributes that generate faultlines. Other studies have tried to analyse the context of teams and organisations by exploring the group’s characteristics (i.e., group size and the number of subgroups), group-level moderators (i.e., openness to experience and the salience of subgroup differences) and organisational and national culture in order to examine the faultline-performance relationship. As is the case in many new research areas, the findings of faultlines studies are not consistent, and many empirical studies have neglected several aspects of faultlines that are critical to understanding the link between faultlines and group performance; for example, faultline activation and evolution. There is an insufficient understanding of the micro aspects of subgroup formation by which to explain ‘how’ individuals align themselves to form rivals in a team, and the reason ‘why’ individuals try to formulate faultlines is still underdeveloped. This area of interest is called faultline triggers. Only a few studies have provided limited categories of triggers and more research is needed. The majority of the work on faultlines has investigated how demographic faultlines affect group processes and outcomes. However, little research has investigated the faultlines’ interactions via a process perspective. In other words, the question about ‘how’ teams interact regarding these differences within the team when faultlines emerge or are present also requires further study. This study, therefore, draws on diversity research and conflict management studies, which have introduced a theoretical framework that integrates teams’ early and late conflict states, faultline activation, the conflict transition process and the conflict management process. By targeting the behaviours and interactions of Top Management Teams (TMT), this study employs both upper echelon theory and faultline theory to understand the faultline transition process. Firstly, this study reviews the recent research that investigates the interplay of team conflict types. This study advances the Team Conflict Dynamics Model to examine conflict types within a dynamic and changing viewpoint. This model considers dynamics by examining conflict transformations in Top Management Teams, the reciprocal effects of conflict management processes and the negative impacts on the emerging faultlines. Using current studies in the conflict management field, this study explores whether the two types of conflict states (i.e., task conflicts and relationship conflicts) can be transformed to faultlines. This study then explores TMT/organisational characteristics and events that will activate faultlines. Previous research incorporates contextual features involving team design and contextual factors. Together with transformed task conflicts and relationship conflicts in the transition process, the dimensions motivating the emergence of faultlines form a typology of faultline triggers. Using a process-state perspective, this study then proposes the conflict management procedure as a dynamic transition and action process. The measures dealing with task conflicts and relationship conflicts within Top Management Teams are examined as the pre-emptive procedures that prevent faultlines from emerging. This study explores how senior managers deal with subgroups’ conflicts, which are referred to as reactive procedures, after faultlines are activated within TMTs. CEOs’ leadership, in terms of pre-emptive procedures and reactive procedures, will be explored separately. Thus, this process-based study explores the interaction between team members in order to prevent and react to faultlines. The findings categorise three different types of faultlines based on interests, relationships and seniority. They confirm that task conflicts and relationship conflicts can be transformed into faultlines in a specific context. In addition, these two conflicts will result in different types of faultlines, as explored in this study. Other triggers besides existing conflicts may also activate faultlines in Top Management Teams. The results suggest that faultline triggers, including specific legitimising events (i.e., newcomers and successive CEOs), have a significant impact on TMTs’ team morale and cohesion. The emerging findings emphasise the issue of nepotism; namely, when ties to relatives and friends are present in TMTs. By dividing nepotism into successor nepotism, which is related to new CEOs; schism nepotism, which is related to member exit; and proximity nepotism, which is related to relationship distance, the findings argue that tensions between subgroups significantly affect team cohesion and activate faultlines. This study also provides evidence that conflict management approaches are affected by the type of conflicts that existed in the early and late life cycle stages. Thus, this study provides an overview of how top management teams manage early levels of conflict types and how these approaches affect the later levels of conflict type, which are referred to as faultline conflicts. After examining the CEO-TMT interface, it is found that TMT members are mainly engaged in pre-emptive procedures that use cooperative conflict management approaches, whereas in reactive procedures, CEOs’ leadership approaches are critical in determining whether activated faultlines are exacerbated or lessened. The findings highlight the importance of early intervention and acknowledging the different effects of CEOs’ personal leadership approaches and TMT approaches in pre-emptive procedures and reactive procedures. The findings suggest that conflict types and conflict management approaches should be modelled together in order to understand team conflicts better. This study advances the Team Conflict Dynamics Model. It does this by examining conflict transformations in TMTs, the reciprocal effects of conflict management processes and the negative impact of events on emerging faultlines. This study’s new typology of faultline triggers helps scholars to understand whether there are differences among teams with faultlines that are dormant, faultlines that are active or dormant faultlines that have been triggered to become active. This study recognises and pinpoints the detrimental effects of the involvement of relatives and friends in TMTs and introduces the idea that nepotism can apply in non-family owned organisations. This process-based study acknowledges the different impacts of TMTs’ managerial practices and CEOs’ leadership practices, according to the pre-emptive and reactive stages of conflict management that are distinguished throughout the study.

This thesis investigates the stability problem of controlling nonlinear systems. Nonlinear systems are hard to be controlled due to its nonlinear terms and sometimes it is hard to find feedback gains for the controller even with the application of advanced nonlinear control methods due to the conservativeness of these methods. Consequently, some of the nonlinear systems cannot be controlled due to the limitation of control methods. To well control these nonlinear systems, the research of interval type-2 (IT2) polynomial fuzzy-model-based (PFMB) control systems is conducted. For IT2 PFMB control systems, the nonlinear plant is represented by the polynomial fuzzy model as the PFMB control scheme demonstrates higher potential to represent the nonlinearity of the nonlinear system than the traditional Takagi-Sugeno (T-S) fuzzymodel-based (FMB) control method when polynomials are used in local systems. Meanwhile, the IT2 fuzzy logic is used to tackle the uncertainties because of its property that can incorporate uncertainties into the membership functions. Hence, both of system nonlinearity and uncertainties can be handled by using the IT2 PFMB control method. Then, aims of this thesis are to investigate the stability analysis and to relax the stability conditions of the IT2 PFMB control system. Firstly, this thesis investigates the stability conditions, in terms of sum-of-squares (SOS), for PFMB control systems with mismatched IT2 membership functions based on Lyapunov stability theory. To relax the stability conditions, the information of the membership functions is introduced into the analysis. By introducing different information of the membership functions, three approaches, namely membership functions approximated by piecewise linear membership functions (PLMFs), linear functions in sub-domains and polynomial functions in sub-domains, are proposed to conduct the stability analysis. All of these methods have feasible control results and can achieve larger feasible regions comparing with that of membership function independent (MFI) method, which is the main stream of analysis techniques in the literature. Secondly, this thesis employs a switching control scheme to further relax the stability conditions. An IT2 polynomial switching controller is introduced into the IT2 PFMB control system. With the application of the switching controller, the whole operation domain is divided into several sub-domains, and the stability analysis is conducted in each sub-domain. Although this method introduces a more complex controller, it relatively relaxes the stability conditions when the same information of membership functions is used. Thirdly, this thesis introduces a mean controller of the IT2 fuzzy controller to investigate the system stability. The mean controller is the average sum of some type-1 polynomial fuzzy controllers, whose type-1 membership functions are embedded into the original IT2 membership functions. When more embedded type-1 membership function information is considered the larger feasible region can be obtained and the stability conditions are more relaxed.

This study aims to develop novel diagnostic assays for zika virus (ZIKV) and equine influenza viruses (EIV) by • Generating pseudotyped ZIKV as a surrogate for wild-type virus in the virus neutralisation (VN) test • Producing plant-expressed ZIKV non-structural (NS) 1 protein as a surrogate for wild-type virus in the ELISA • Identifying lineage-specific antibodies against EIV using Next Generation Phage Display (NGPD) to replace antibodies produced by immunizing animals for use in ELISA. Co-circulation and cross-reaction among flaviviruses have become the major obstacle for an accurate diagnosis. Antigenic drift and shift of influenza viruses underline the importance of close monitoring and surveillance for precise diagnosis and vaccine production. This study aimed to develop novel diagnostic assays for emergent/re-emergent viruses, namely, Zika virus (ZIKV) and equine influenza virus (EIV) using various techniques including pseudotyping, plant expression and next-generation phage display (NGPD) to improve their current diagnostic assays. Vesicular stomatitis virus (VSV) backboned pseudotype virus (PV) was generated by co-transfecting producer cells with commercially available VSV plasmids. ZIKV envelope protein (ZIKV E) was used to replace the VSV glycoproteins to generate VSV-backboned ZIKV PV, which showed no infectivity. However, VSV backboned Chikungunya virus (CHIKV) PV used as a control exhibited similar luciferase activity as the VSV PV (~1×107 RLU/ml), indicating successful production of VSV-backboned CHIKV PV. The virus budding site of ZIKV is different from CHIKV and VSV, which was identified as one of the possible reasons for the failure of infectivity of ZIKV PV. Future work would be to mutate the ER-retention signals in the transmembranes of ZIKV prME for its translocation to the plasma membrane and to incorporate furin proteases during pseudotyping for the ZIKV E to achieve a mature and infectious form. Next, a ZIKV non-structural protein 1 (NS1) ELISA was developed using ZIKV NS1 proteins expressed in plants. Unlike envelope proteins, NS1 does not elicits flavivirus cross-reactive neutralisation antibodies but induces a non-neutralising yet virus-specific antibody response. NS1 gene was codon-optimised, synthesised and inserted into a plant expression vector pEAQ-HT to obtain expressed protein. Preliminary results using the resultant ELISA showed around 95% sensitivity and 95% specificity using 20 ZIKV positive human serum samples and 19 Dengue-only positive samples. For future work, a bigger sample size should be tested for the proper validation of this plant produced NS1 ELISA. Finally, antibody peptide sequences that specifically recognise a strain of EIV from clade 2 and not the strain from clade 1 were identified using NGPD and data analysis. It is important to be able to distinguish different clades to keep current equine vaccine effective for the circulating strain. Future work would include the recovery of the peptide sequences using inverse PCRs and testing their specificities against other EIV strains in the two clades. Identified antibody peptides could be used to replace ferret antisera in the haemagglutination test for characterising circulating EIVs.

With the arrival of the data explosion era, data modelling/prediction has become one of the most important research areas. Many datasets in our real-life are essentially data streams and functional data. This thesis investigates fuzzy system-based regression approaches for these two representative datasets. In line with the data types, this thesis is divided into two scenarios. The first part focuses on the data stream regression problem with input/output being real vectors. It is widely known that evolving fuzzy systems (EFSs) are effective approaches for solving data stream regression problems, in that EFSs are structurally self-organized, capable of updating structure/parameters in an online manner, and acting as the one-pass approaches without the requirement of storing historical data. However, current state-of-the-art studies indicate that the existing evolving structure/parameters approaches would impose a negative impact on the optimality of EFSs. To our best knowledge, research on proposing optimal EFSs was still rare. Furthermore, selecting predefined thresholds to control the structure/parameters evolution of EFSs is of importance, which has not been systematically investigated thus far. In this part, this thesis focuses on addressing the aforementioned two EFSs-related problems, which might provide implications for the research on the data stream. From an optimality viewpoint, EFS learning approaches for both Takagi-Sugeno and Mamdani fuzzy systems, that is, local error optimization approach (LEOA) and identifying evolving Mamdani fuzzy systems from the parameter optimization aspect (EMFSPO), are proposed. To automatically tuning the thresholds, two approaches, i.e. the self-evolving fuzzy system (SEFS), and an extended work based on SEFS, that is, EFS with self-learning/adaptive thresholds (EFS-SLAT), are proposed. Finally, through a wide range of benchmark examples, LEOA, EMFSPO, SEFS, and EFS-SLAT are shown to be capable of improving the accuracy compared with many state-of-the-art approaches. The second part of this thesis focuses on the functional data regression problem, where inputs/outputs are functions. Current techniques for functional data regression are mainly statistical models with certain assumptions and restrictions for a specific model. Since the kernel functions should be used in integral in these approaches, the resulting models were lack of interpretability and difficult to understand especially for the nonlinear case. More importantly, they were mainly offline models, which were not suitable to handle big functional data/functional streaming data considering the computational cost for storing a vast of historical information during model analysis. To improve these weaknesses, a pioneering functional regression approach, i.e. the functional fuzzy system (FFS), is developed. It acts as an operator between two functional spaces. FFS is an offline nonlinear regression model enabling its inputs/outputs to be functions. It has a flexible structure that can explain any nonlinear relationship between functions using an interpretable ``If-Then" fuzzy inference. In order to make the model applicable, the corresponding learning approach is proposed for identifying FFS. Since FFS is an offline model, it is not suitable for big functional datasets and functional streaming data. To overcome such bottlenecks, the first functional real-time regression model known as evolving functional fuzzy system (EFFS) is developed. EFFS is endowed with an exceptional advantage of learning from data online in addition to inheriting all the superior characteristics of FFS. EFFS provides a potential approach for handling big functional data by processing functional streaming data with each data point being a function. Finally, it demonstrates that FFS and EFFS outperform many of the state-of-the-art approaches through comparison experiments on various benchmark examples. These numerical results verify the effectiveness of FFS and EFFS.

This thesis presents a set of shaking table benchmark tests on six large scale RC bridge piers with/without corrosion damage. Three well-confined specimens are tested with three spectrally matched ground motion records (near-field without pulse, near-field pulse-like and far-field) to investigate the non-stationary properties of the ground motion. To examine the effects of confinement and corrosion, three low-confined specimens are tested by the same ground motion (i.e. far-field). Two of the low-confined columns are corroded by the accelerated corroding procedure. The mass loss ratios of the steel reinforcement in the two specimens are 51% and 65%. Four rounds of the main tests with different scale factors (25%, 300%, 500% and another 300% as aftershock) of the ground motion PGA are conducted. Before the first round of the main test and after each round of the main test, low-amplitude white-noise tests are conducted to explore the fundamental frequency variation before/after the test. A spatial investigation is carried out on the corroded specimens to investigate the effects of corrosion distribution. With the experimental data, a rapid damage assessment method based on time-frequency analysis (i.e. Wigner-Ville distribution and Hilbert transform) is proposed. This non-parametric method only relies on input and output acceleration time-histories due to the easy acquisition of acceleration in situ. The effectiveness of this method is benchmarked by the moving linear regression of hysteretic loops and white-noise tests. A system identification strategy on the extended Bouc-Wen model with unknown parameters is proposed. Genetic algorithm is used in this process. A numerical model of the RC column is developed. The performance of the model in terms of hysteresis and energy dissipation modelling is benchmarked by the experimental data. To demonstrate the application of the proposed numerical model in design process, incremental dynamic analysis (IDA) on uncorroded columns is carried out. The results compare the performance of engineering demand parameters (i.e. drift and energy dissipation) in sequential earthquake events.

Microbubble-enhanced ultrasound imaging, and the more recent advances in super-resolution imaging and molecular imaging, have shown great promise in a wide range of clinical applications. However, a number of challenges still exist during their clinical translations. For instance, in ultrasound superresolution imaging, the in vivo lifetime of microbubble is normally less than 5 minutes and the concentration of microbubbles in vivo is difficult to adjust after the injection. Nanodroplet, as a condensed version of microbubble, has a number of advantages in super-resolution imaging. Firstly, nanodroplet have a longer lifetime compared to microbubbles. This enables less contrast agents injected and a longer scanning time. Second, these nanodroplets can be controlled to be activated to provide the contrast signals both spatially and temporally on-demand. This potentially provides more flexibility during the ultrasound scanning. In the context of molecular imaging, the nano-size of droplets potentially allows extravasation into cancerous tissue due to its leaky vasculature and enhanced permeability and retention effects where microbubbles cannot. In the field of photoacoustic imaging, dye-coated nanodroplet was developed to provide a significant contrast enhancement in photoacoustic imaging upon the optical activation where microbubbles also cannot be used. Among a variety of research efforts on facilitating the pre-clinical translation, this work focuses on the applications of nanodroplets in ultrasound and photoacoustic super-resolution and molecular imaging. There are four main scientific contribution as follows. First, the feasibility of using low-boiling-point nanodroplets to perform ultrasound super-resolution imaging was investigated. 'Acoustic wave sparsely activated localization microscopy (AWSALM)', an acoustic counterpart of photo-activated localization microscopy (PALM), is developed to super-resolve structures which cannot be resolved by conventional B-mode imaging. AWSALM utilizes acoustic waves to sparsely and stochastically activate decafluorobutane nanodroplets by acoustic vaporization and to simultaneously deactivate the existing vaporized nanodroplets via acoustic destruction. This technique is less dependent on flow and does not require a low concentration of contrast agents, as is required by current ultrasound super resolution techniques. Acoustic activation and deactivation can be controlled by adjusting the acoustic pressure, which remains well within the FDA approved safety range. This study shows the promise of a flow and contrast agent concentration less dependent super-resolution ultrasound technique which has potential to be faster and go beyond vascular imaging. Second, to achieve super-resolved imaging frames with sub-second temporal resolution, fast acoustic wave sparsely activated localization microscopy (fast-AWSALM) was developed by using low-boiling-point octafluoropropane nanodroplets and high frame rate plane waves for activation, destruction, as well as imaging. The effects of the temperature and mechanical index on fast-AWSALM was investigated. The contrast signals were quantified as a function of acquisition time. The final results showed two orders of magnitude faster than the reported localization-based ultrasound super-resolution techniques, under a non-flow/very slow flow situations. Just as in AWSALM, fast-AWSALM is less dependent on flow, as is required by current microbubble based ultrasound super resolution techniques. This study shows the promise of fast-AWSALM, a superresolution ultrasound technique using nanodroplets, which can generate super-resolution images in milli-seconds. Third, to have a better design of targeted nanodroplets for imaging and therapeutic applications, the size and acoustic response of targeted nanodroplets under high-frame-rate ultrasound and optical imaging was investigated. A flow velocity mapping technique, Stokes' theory and optical microscopy were used to estimate the size of both floating and attached vaporized nanodroplets immediately after activation. The floating vaporized nanodroplets were on average more than seven times larger than vaporized nanodroplets attached to the cells. The results also indicated that the acoustic signal of vaporized FR-targeted-nanodroplets persisted after activation, with 70% of the acoustic signals still present 1 s after activation, compared with the vaporized NT-nanodroplets, for which only 40% of the acoustic signal remained. The optical microscopic images revealed on average six times more vaporized FR-targeted-nanodroplets generated with a wider range of diameters (from 4 to 68mm) that were still attached to the cells, compared with vaporized NT-nanodroplets (from 1 to 7mm) with non-specific binding after activation. The mean size of attached vaporized FR-targetednanodroplets was on average about threefold larger than that of attached vaporized NT-nanodroplets. Taking advantage of high-frame-rate contrastenhanced ultrasound and optical microscopy, this study offers an improved understanding of the vaporization of the targeted nanodroplets in terms of their size and acoustic response in comparison with NT-nanodroplets. Such understanding would help in the design of optimized methodology for imaging and therapeutic applications. Fourth, the feasibility of using dye-coated nanodroplets to perform photoacoustic super-resolution imaging was investigated. A photoacoustic super-resolution imaging technique was developed through imaging the activation of Cyanine 7.5-coated phase-change nanodroplets using a preclinical photoacoustic imaging system and localizing the activated droplets. As a proof-of-concept experiment, photoacoustic images of microfluidic channels were obtained with a cylindrically focused curvedarray, while dye-coated nanodroplets flowed through the channels. Experimental results showed that super-resolution images can resolve structures which cannot be resolved by conventional beamformed images in vitro. The results also showed that the dye-coated phase-change nanodroplets can be optically activated in vivo and the activation signals can be separated from the image background by applying singular value decomposition filtering, and be used for further super-localization processing. Nanodroplets offer better biocompatibility, as well as more flexible and controllable droplet activation rates, with potential for superresolution imaging without flow and of extravascular targets, compared to contrast agents used in existing localization-based photoacoustic superresolution imaging techniques.

Computer-aided design (CAD) models play a crucial role in the design, manufacturing and maintenance of products. Therefore, the mesh-based finite element descriptions common in structural optimisation must be first translated into CAD models. Currently, this translation either can be performed semi-manually or fails to reserve the structural optimality found by the structural optimisation due to the intrinsic difference in geometric representation between finite element mesh and CAD model. This thesis propose a fully automated and topologically accurate approach to synthesise structurally sound parametric CAD models from topology-optimised finite element models to fill the long-existing gap between structural optimisation and CAD systems. This approach successfully preserves the optimal structural performance during the mesh-CAD conversion. The solution provided in this thesis is to first convert the topology-optimised structure into a spatial frame structure and then to regenerate it in a CAD system using standard constructive solid geometry (CSG) operations. The obtained parametric CAD models are compact, that is, have as few as possible geometric parameters, which makes them ideal for editing and further processing within a CAD system. The critical task of converting the topology-optimised structure into an optimal spatial frame structure is accomplished in several steps. The first step is to generate a one-voxel-wide voxel chain model from the topology-optimised voxel model using a topology-preserving skeletonisation algorithm from digital topology. The undirected graph defined by the voxel chain model yields a spatial frame structure after processing it with the proposed graph algorithms. Subsequently, the cross-sections and layout of the frame members are optimised to recover its optimality, which may have been compromised during the conversion process. At last, the obtained frame structure is generated in a CAD system by repeatedly combining primitive solids, like cylinders and spheres, using boolean operations. The resulting solid model is a boundary representation (B-Rep) consisting of trimmed non-uniform rational B-spline (NURBS) curves and surfaces. The numerical studies in this thesis clarify that the converted spatial frame structures are with equivalent structural performance. Moreover, CAD models generated from the spatial frame structures have significantly fewer geometric degree of freedom compared to the topology-optimised structures. Though the numerical studies use topology-optimised structures as input and compact CSG models as output, this thesis also provides the way to extend the proposed generation process to taking other optimised meshes and producing outputs of various geometric representations. This offers a wide range of possible applications and brings new thoughts to industrial design and manufacturing.

This research is a comparative study on the works of Charles Dickens (1812-1870) and Zhang Tianyi 張天翼 (1906-1985). The former was one of the greatest novelists of the Victorian era; the latter, a Left-wing writer in Republican China. The study analyses five short stories from Zhang's corpus and compares his works with ten novles of Dickens. The study argues that Dickens is one among other writers that have parallels with Zhang, through the exploration of several aspects of their works. At the beginning of the twentieth century, Dickens's novels were introduced to China by Lin Shu. Since then, they have influenced a great number of Chinese scholars and writers. First, considering the contents of Zhang's short stories, I roughly categorise them into five modes, respectively focusing on bureaucrats, villains, social climbers, revolutionary propa-ganda and wartime intellectuals. One short story of each type is analysed in detail. Second, this study explores both the similarities and the discrepancies between Zhang and Dickens's inner worlds and the projections of those worlds within their texts. I find that the two authors have much in common in their character portrayals, humour, language, children's perspectives and focus on social issues, whereas they differ considerably in what they hoped to achieve with their works, partly due to their different social settings, and this in turn gives their works very different qualities. Where Dickens believed in the goodness of human nature and hoped for gradual reform, Zhang Tianyi was a proponent of radical change and revolution. This study expands the scope of the existing research on Zhang Tianyi, clarifies and establishes important factors in Zhang's literary development, and explores how Zhang imitated Dickens's artistic techniques and then creatively transformed them. Moreover, this study provides new perspectives and deepens existing studies. This investigation will thereby enable others to better understand Zhang's literary works, and will simultaneously help map Dickens's influence outside Britain.

There is an increasing awareness that a translation product is composed of both the textual part and the promotional materials so that it is commercialised and socialised based on the market demand and the profile of the publisher. As a mediation between the readers and the translated text, the promotional materials, known as paratexts, can be very influential in familiarising consumers with the product, indicating the genre of the text and determining the target readership. While they play an essential role in managing how readers perceive the translation before they begin the book, they also reflect the publisher's and the other producers' voices in depicting the product based on its position in the social context as well as their assumptions about the preferences of the market. Thus, a study of translational paratexts allows us to observe the participation of different social agents and institutions in the process of production as well as their joint efforts to make the product more readily accepted by the target culture. In addition, the heterogeneous nature of paratexts generates additional reflections on research methodologies, such as the integration of the visual material analysis in the field of translation studies. In terms of research objects, Chinese translations of Lady Chatterley's Lover and Lolita are selected as appropriate materials for case studies due to the fact that these two controversial works have received a great deal of attention from both the general public and the translation field in China since their publication. The long history of translation and retranslation of these two works makes them ideal for a diachronic study observing how the translation field and publishing industry have changed in the past several decades in China. At the same time, their controversial nature highlights the struggles and compromises of the publishers due to the socio-political context.

This thesis analyzes the performance, performance persistence and fund flows for samples of both domestic and international equity unit trusts (mutual funds) using a new specially constructed dataset consisting of 650 UK domestic equity unit trusts and 627 UK-based international equity unit trusts over a sample period from January 1998 to March 2015. The comparative analysis between domestic and overseas investments provides new evidence on the mutual savings market in the UK and incorporates the effects of the financial crisis post-2008. The thesis contributes to the existing literature by providing empirical evidence on the two-step flow-performance relationship outlined in Berk and Green (2004) model for UK domestic and international mutual funds. We find little evidence that UK domestic and international equity mutual funds earn positive abnormal returns on a risk-adjusted basis across a range of investment objectives and across many alternative benchmarks. Neither UK domestic mutual fund managers nor international mutual fund managers have the ability to time the market, nor generate positive alphas in recessions and post the 2008 financial crisis. To the extent that manager skill exists in UK mutual funds, they are limited to a smaller number of funds. The flow-performance relationship has stronger convexity for the best performing UK domestic funds compared with international mutual funds. In comparison to domestic investor flows, international fund flows are more sensitive to risk-adjusted returns than raw returns, especially the most recent risk-adjusted performance. There is evidence that search costs affect the flow-performance sensitivity for UK domestic and international mutual funds when they are proxied by fund fees and fund family size. The impact concentrate on funds in certain performance ranges, that is, the bottom quintile and middle three quintiles, rather than all performance ranges. We find evidence of performance persistence for UK domestic equity unit trusts at the short horizon of one month and some evidence of performance persistence for international equity mutual funds at intermediate term horizons. Long-term persistence in underperformance is also evidenced for UK domestic funds. Our results from testing for the existence of the equilibrating mechanism of fund flows outlined in Berk and Green (2004) suggests that at the one-year horizon, fund flows reduce performance persistence in UK domestic unit trusts However, we find no evidence that fund flows affect performance persistence in UK-based international equity mutual funds.

Ribosomally synthesized and post-translationally modified peptides (RiPPs) are a burgeoning family of natural products with desirable bioactivities. The cyanobactin class of RiPPs have demonstrated anticancer, anti-microbial activities and more, and have therefore attracted research attentions. Cyanobactins are synthesised as a precursor peptide (PatE homologues) containing recognition sequences and core peptides (sequences that become the final products), the latter of which undergo various modifications including heterocyclization, prenylation, oxidation and macrocyclization. The posttranslational modification enzymes (PTMEs) utilised to modify the cyanobactin precursor have demonstrated magnificent versatilities that are potentially exploitable for the creation of libraries of natural and unnatural compounds. The structure and function of each of these enzymes are reviewed in Chapter 1, along with the current application of the cyanobactin biosynthetic pathway. Chapter 2 describes my work on creating and characterising modified heterocyclases. The leader peptide was appended to heterocyclases that produce both thiazolines and (methyl)oxazolines. The modified catalysts are not only capable of utilising leaderless peptides, but also have higher yields than the native enzyme. The next chapter investigates the reaction order of heterocyclase enzymes. The leader peptide was found to be partially responsible for the reaction order. In Chapter 4 the same investigation was applied to the oxidase, which aromatises and stabilises the heterocycles. The oxidase reaction was determined to be independent of the leader. Chapter 5 delves into the phosphate chemistry of the heterocyclase, which uses ATP/Mg2+ via a kinase mechanism, but subsequently catalyses additional reactions that yield AMP and PPi. Finally, Chapter 6 describes the structural characterisation of Psychrobacter arcticus ATP phosphoribosyltransferase (ATPPRT), which is the first dedicated enzyme in histidine biosynthesis, and provides a key regulatory point for this pathway.

ATP-driven motors exist globally in almost all prokaryotic and eukaryotic organisms. The motors in viruses are essential for viral replication and assembly. In this thesis, two ATP-driven nucleic acid motors from viruses were studied: a DNA packaging ATPase present in Thermus thermophilus phage phiKo and an NS3 helicase from Zika virus. The production and purification of the DNA packaging ATPase (gp14) from the phage phiKo was investigated. The purified gp14 protein was crystallised for determination of the native structure as well as the complexes with several non-hydrolysable ATP analogues, representing a precise picture of the ATP hydrolysis cycle. Structural analysis reveals that gp14 possesses an FtsK-like fold and likely translocates DNA similarly to other FtsK/HerA superfamily translocases. Moreover, an additional catalytic sensor residue from the adjacent subunit is required for ATP hydrolysis and protein assembly. The appropriate assembly of gp14 may be achieved by co-assembling with an auxiliary protein component into a higher-order hetero-oligomer complex. Further characterisations, including thermal unfolding, ATPase activity and DNA binding assays elucidate that gp14 is thermodynamically stable, but its ATPase activity decreases over time. The purified Zika virus NS3 helicase (NS3h) was crystallised in its apo form and as complexes with several nucleotides, including pre-hydrolysis substrate, transition state, and post-hydrolysis product analogues, as well as in a complex with a single-stranded RNA (ssRNA) segment containing a 5' phosphate. The structural observations illustrate how a catalytically important motif mediates the coupling of ATP hydrolysis with RNA translocation. Due to the transition state analogue ADP-MgF3(H2O)- not being observed before, this complex structure was subjected to careful examination by crystallography and 19F NMR. Further functional characterisations, including protein stability, RNA binding and ATPase activity assays, demonstrate this NS3 helicase possesses weak RNA-binding mode and its ATPase activity can be stimulated by ssRNA.

Whole genome comparison of metazoan genomes reveals extremely high level of noncoding conservation over tens to hundreds of base pairs across distant species. These sequences are termed as conserved noncoding elements (CNEs). Arrays of conserved noncoding elements that span the loci of developmental regulatory genes and their span defines regulatory genomic blocks (GRBs). CNEs are currently known to be involved in transcriptional regulation and development as long-range enhancers. However, no molecular mechanism can yet explain their exceptional degree of conservation. As a first step towards the genome-wide study of these elements, I developed two R/Bioconductor packages CNEr and TFBSTools, to detect and analyse regulatory elements. Next, I designed a novel CNE detection pipeline for duplicated regions in the ameiotic Adineta vaga genome. Identification of CNEs in this genome suggests that the principal function of CNEs is regulation of developmental gene expression rather than copy number sensing. In addition, I performed a de novo genome annotation of European common carp Cyprinus carpio. This genome stands as an ideal candidate for comparative study of zebrafish genome. Its analysis revealed a wealth of previously undetected fish regulatory elements and their unexpectedly high level of conservation between the two genomes. Finally, I presented a computational method for the identification of GRB boundaries and prediction of the corresponding target genes under long-range regulation. The predicted target genes are implicated in developmental, transcriptional regulation and axon guidance. The disruption of regulation of these target genes is likely to cause complex diseases, including cancer. The GRB boundaries and predicted target genes are valuable resource for investigating developmental regulation and interpreting genome-wide association studies.

Informal settlements have become an essential part of urban habitation patterns both globally and in China. This research reviews the interaction of industrialisation, urbanisation, migration and the informal economy alongside institutional factors that have encouraged the development of informal settlements in Beijing. Then it asks and attempts to answer the research question: How should ‘informal settlement’ be understood in Beijing, China? This research develops the empirical study in three phases. Firstly, it explores government approaches to informal settlements and their spatial distribution in Beijing from political and academic elites’ perceptions. Secondly, it profiles the functions that different types of informal settlements perform currently in Beijing through a survey. Thirdly, it investigates the needs of informal settlement dwellers, and how Beijing fits with the global conceptual understandings of informal settlements. This research makes four key findings to our understanding of informal settlements in Beijing. First, it maps out the spatial configuration of informal settlements, which are dispersed throughout Beijing, and documents diversity in physical manifestation of informal settlements, including evolved new types and changing complications within each type. In this sense, it provides an overall picture of Beijing informal settlements geographically and typologically. Second, it expands the understanding of informal settlements that to define the informal is not only determined by norms, regulations or laws, but also by the extent to which they are enforced; and the intensity of their enforcement depends on how government officials choose to interpret policies. Third, it suggests informal settlements satisfy dweller’s housing need because they are merely a sleeping place rather than home, and dwellers care more about socio-economic needs such as earning/saving money and social networks. Finally, it predicts the unstoppable trend of informal settlements in Beijing considering the enormous income gap between urban and rural China, together with the existence of informal economy.

Since the 1990s, grammatical complexity is a topic that has received considerable attention in various fields of applied linguistics, such as English for academic purposes, second language acquisition, language testing, and second language writing (Bulte & Housen, 2012). Many scholars in applied linguistics have recently argued that grammatical complexity has primarily been represented by clausal features (e.g., subordinate clauses), and it is important to study grammatical complexity as a multidimensional construct based on both clausal features and phrasal features (Biber, Gray & Poonpon, 2011; Norris & Ortega, 2009). Thus, this dissertation is a corpus-based investigation on how the use of noun phrases is influenced by two situational characteristics of a university context: academic level and first- and second-English language background. I built my corpus by extracting 200 essays from British Academic Written English Corpus, which represents academic writing of (1) undergraduate and graduate students and (2) L1 and L2 students. Noun phrase complexity was then operationalized to the 11 noun modifiers proposed in the hypothesized developmental index of writing complexity features in Biber, Gray and Poonpon (2011). The 11 noun modifiers were extracted from the corpus and counted for statistical analysis via a set of Python programs. With a Chi-square test followed by a residual analysis, I found that both academic level and first- and second-English language background influenced noun phrase complexity but in distinct ways. The influence of academic level is primarily associated with three phrasal modifiers (i.e., attributive adjectives, premodifying nouns, and appositive NPs) and two clausal modifiers (i.e., relative clauses and noun complement clauses). The undergraduate corpus includes more of the two clausal modifiers, whereas the graduate corpus has more of the three phrasal modifiers. This suggests that, in these 200 essays, graduate students tend to build more compressed NPs than undergraduate students. However, the influence of first- and second-English language background derives from a much broader range of noun modifiers, including eight noun modifiers (e.g., attributive adjectives, relative clauses, infinitive clauses). More diverse NP patterns with different noun modifiers are in the L1 corpus than in the L2 corpus. Surprisingly, the L2 corpus has more phrasal noun modifiers (i.e., attributive adjectives, premodifying nouns), which has been argued to indicate advanced levels of academic writing. A qualitative analysis on selected essays reveals that some cases of attributive adjectives and premodifying nouns are repeatedly used by L2 students to help content development in their writing. Overall, this dissertation adds an additional piece of evidence on the importance of noun phrase complexity in writing research. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]

Milling is a common unit operation in industry for the purpose of intentional size reduction. Considerable amount of energy is consumed during a grinding process and much of the energy is dissipated as heat and sound, which often makes grinding into an energy-intensive and highly inefficient operation. Despite many attempts to interpret particle breakage during a milling process, the grindability of a material in a milling operation remains aloof and the mechanisms of particle breakage are still poorly understood. Hence the optimisation and refinement in the design and operation of milling are in great need of an improved scientific understanding of the complex failure mechanisms. This thesis aims to provide an in-depth understanding of particle breakage associated with stressing events that occur during milling. A hybrid of experimental, theoretical and numerical methods has been adopted to elucidate the particle breakage mechanics. This study covers from single particle damage at micro-scale to bulk comminution during the whole milling process. The mechanical properties of two selected materials, i.e. alumina and zeolite were measured by indentation techniques. The breakage test of zeolite granules subjected to impact loading was carried out and it was found that tangential component velocity plays an increasingly important role in particle breakage with increasing impact velocity. Besides, single particle breakage via in-situ loading was conducted under X-ray microcomputed tomography (μCT) to study the microstructure of selected particles, visualize the progressive failure process and evaluate the progressive failure using the technique of digital image correlation (DIC). A new particle breakage model was proposed deploying a mechanical approach assuming that the subsurface lateral crack accounts for chipping mechanism. Considering the limitation of existing models in predicting breakage under oblique impact and the significance of tangential component velocity identified from experiment, the effect of impact angle is considered in the developed breakage model, which enables the contribution of the normal and tangential velocity component to be rationalized. The assessment of breakage models including chipping and fragmentation under oblique impact suggests that the equivalent normal velocity proposed in the new model is able to give close prediction with experimental results sourced from the public literature. Milling experiments were performed using the UPZ100 impact pin mill (courtesy by Hosokawa Micron Ltd. UK) to measure the comminution characteristics of the test solids. Several parameters were used to evaluate the milling performance including product size distribution, relative size span, grinding energy and size reduction ratio etc. The collective data from impact pin mill provides the basis for the validation of numerical simulation results. The Discrete Element Method (DEM) is first used to model single particle breakage subject to normal impact loading using a bonded contact model. A validation of the bonded contact model was conducted where the disparity with the experimental results is discussed. A parametric study of the most significant parameters e.g. bond Young’s modulus, the mean tensile bond strength, the coefficient of variation of the strength and particle & particle restitution coefficient in the DEM contact model was carried out to gain a further understanding of the effect of input parameters on the single particle breakage behavior. The upscaling from laboratory scale (single particle impact test) to industrial process scale (impact pin mill) is achieved using Population Balance Modelling (PBM). Two important functions in PBM, the selection function and breakage function are discussed based on the single particle impact from both experimental and numerical methods. An example of predicting product size reduction via PBM was given and compared to the milling results from impact pin mill. Finally, the DEM simulation of particle dynamics with emphasis on the impact energy distribution was presented and discussed, which sheds further insights into the coupling of PBM and DEM.

Anurans have lived on earth for more than 300 million years. During this long period of time, they learned how to adapt to the environment and defend against microorganisms. Amphibian skin secretions generally contain complex peptidomes. These skin peptides constitute the first line of defence of amphibians against bacteria and fungi. Until now, more than 600 antimicrobial peptides (AMPs) from amphibians have been authenticated. AMPs selectively retard bacterial growth and kill bacteria. Compared with conventional antibiotics, AMPs have a number of clinical application advantages. Chapter 3 describes the study of two AMPs from Phyllomedusa sauvageiand Phyllomedusa hypochondrialis, named QUB2897 and QUB2869.These two peptides were synthesised by a solid phase peptide synthesis method and were purified by HPLC. A Q-TOF mass spectrometer was used to determine the peptides purity. Staphylococcus aureus (S.aureus), Escherichia coli (E.coli) and Candida albicans (C.albicans) were used to test the antimicrobial activities of QUB2897 and QUB2869 and horse blood was used to test their haemolytic activities. Both peptides were found to possess antimicrobial and haemolytic activities and they also had anti-cancer cell proliferation activity. Chapter 4 describes three novel peptides identified as QUB2897 and QUB2869 metabolism products from rat urinary bladder and ileum smooth muscle. The novel peptides, named QUB 2208 and QUB2179, were found as products of QUB2897 and QUB2869 metabolism of rat bladder smooth muscle, respectively. The novel peptide, named QUB501,was identified as a metabolite of both QUB2897 and QUB2869 by rat ileum smooth muscle. Using solid phase peptide synthesis (SPPS), these three novel peptides were successfully synthesised. QUB2208 and QUB2179 displayed more potent antimicrobial activities than their parent peptides, QUB2897 and QUB2869, respectively. The bioactivity of QUB501, however, was lower than that of the parent peptides. Chapter 5 describes the isolation and identification of two novel peptides, named QUB2582 and QUB1889, from the defensive skin secretion of the toad, Bombina maxima. The two peptides were synthesised by solid phase peptide synthesis and were purified by HPLC. The Q-TOF mass spectrometer was used to determine the peptides purity. Both peptides were found to possess anti-cancer cell proliferation activities.

This PhD project is focused on three lead-free ferroelectric solid solutions, which are specifically Na0.5Bi0.5TiO3-KNbO3(NBT-KN), Na0.5Bi0.5TiO3-NaNbO3(NBT-NN) and Na0.5Bi0.5TiO3-BaTiO3(NBT-BT), to evaluate the effects of composition, electric field and temperature on structural and electrical properties. Novel observations of both reversible and irreversible electric field-induced phase switching were made in both NBT-KN and NBT-NN ceramics. The NBT-KN solid solution is the primary focus of this thesis. All compositions were observed to be cubic in the as-sintered, unpoled state. However, a well-defined ferroelectric hysteresis P-E loop was obtained for compositions with low KN contents, indicating that an irreversible phase transition from a weak-polar relaxor ferroelectric (RF) to a long-range ordered metastable ferroelectric (FE) state had occurred during the measurement procedure. Both the unpoled and poled ceramic powders were examined using high resolution synchrotron XRD. For the poled state, a rhombohedral R3c structure was identified for compositions with low KN content, confirming the occurrence of the irreversible electric field-induced structural transformation from cubic to rhombohedral. In contrast, a cubic structure was retained for high KN contents, giving rise to reversible phase switching evidenced by constricted P-E hysteresis loops. Similar behaviour was observed for NBT-NN system. An 'in-situ' electric field poling experiment was conducted using high energy synchrotron XRD. In certain NBT-KN compositions the structural transformation, from cubic to mixed phase cubic+rhombohedral and finally single phase rhombohedral, occurred progressively with increasing cycles of a bipolar electric field. Similar behaviour was observed for NBT-NN compositions having low NN contents. Furthermore, the distributions of domain orientation and lattice strain over a range of orientations relative to the poling direction were determined for NBT-KN, NBT-NN and NBT-BT ceramics exhibiting the rhombohedral phase. By combining the structural information with the results of dielectric and ferroelectric measurements, a phase diagram was constructed to illustrate the influence of temperature and composition on the stability of the metastable ferroelectric and relaxor ferroelectric states for the NBT-KN system. Furthermore, the phase transition temperatures obtained from dielectric measurements were correlated with the ferroelectric and thermal depolarisation characteristics for each of the NBT-KN, NBT-NN and NBT-BT systems.

In this thesis, we introduce two different methods for determining noise covariance matrices in order to improve the stability and accuracy in state estimation and output prediction of discrete-time linear time varying (LTV) and nonlinear state space systems. The first method is based on the auto-covariance least squares (ALS) method, where the noise covariance matrices can be estimated by establishing a linear relationship between noise covariances and correlations of innovation sequence, hence solving a linear least squares problem. For LTV systems, we propose a new ALS algorithm that does not involve any approximations in the formulation. Our new ALS algorithm has fewer parameters to determine and can provide more accurate noise covariance estimation even when the historical output measurement window is not sufficiently long, comparing to an existing method. In addition to the noise covariance estimates, our ALS algorithm can also provide the estimate of the initial state error covariance, which is required by most state estimation methods. For higher-order systems, we also provide a much faster and less memory demanding formulation by splitting large Kronecker products with sums of smaller Kronecker or Schur products. For nonlinear systems, we have to approximate nonlinear parts as time-varying matrices by linearizing the nonlinear function around current state estimates. In addition to the extended Kalman Filter (EKF), our ALS algorithm also uses moving horizon estimation (MHE) to estimate the system state. MHE guarantees stability, is able to add state constraints and provides more accurate state estimates and local linearizations around the current state than the EKF. The second method is based on expectation maximization (EM), where the noise covariance matrices are determined by recursively maximizing the likelihood of covariance matrices, given output measurements. In our method, the noise covariance matrices are estimated using a semi-definite programming (SDP) solver, so that the results are more accurate and guaranteed to be positive definite. We propose a new EM algorithm that, combined with MHE and full information estimation (FIE) rather than a Kalman-based filter/smoother, allows the addition of state constraints, provides stable and more accurate estimates, so that the performance of noise covariance estimation can be significantly improved. Finally, we apply our noise covariance estimation methods to ocean wave prediction for the control of a wave energy converter (WEC), in order to approach optimal efficiency of wave energy extraction. We use a state space model representation for an autoregressive (AR) process, combined with noise covariance estimation, to simulate wave height forecasting based on data recorded at Galway Bay, Ireland. The simulation returns good wave predictions. Compared to existing wave prediction methods, our model has fewer parameters to tune and is able to provide more stable and accurate wave predictions by using a Kalman-based filter combined with the ALS or EM method.

This thesis reports on the development of quantitative measurement using micromachined scanning thermal microscopy (SThM) probes. These thermal probes employ a resistive element at their end, which can be used in passive or active modes. With the help of a review of SThM, the current issues and potentials associated with this technique are revealed. As a consequence of this understanding, several experimental and theoretical methods are discussed, which expand our understanding of these probes. The whole thesis can be summarized into three parts, one focusing on the thermal probe, one on probe-sample thermal interactions, and the third on heat transfer within the sample. In the first part, a series of experiments are demonstrated, aimed at characterizing the probe in its electrical and thermal properties, benefiting advanced probe design, and laying a fundamental base for quantifying the temperature of the probe. The second part focuses on two artifacts observed during the thermal scans – one induced by topography and the other by air conduction. Correspondingly, two devices, probing these artifacts, are developed. A topography-free sample, utilizing a pattern transfer technique, minimises topography-related artifacts that limited the reliability of SThM data; a controlled temperature ‘Johnson noise device’, with multiple-heater design, offers a uniform, accurate, temperature distribution. Analyzing results of scan from these samples provides data for studying the thermal interactions within the probe and the tip-sample interface. In the final part, the observation is presented that quantification of measurements depends not only on an accurate measurement tool, but also on a deep understanding of the heat transfer within the sample resulting from the nanoscopic contact. It is believed that work in this thesis contributes to SThM gaining wider application in the scientific community.

This thesis investigates the electromagnetic performance of interior permanent magnet (IPM) machines for electric power steering application and variable reluctance (VR) resolvers for electric/hybrid electric vehicles, with particular reference to the manufacturing issues. It aims to reduce the cogging torque in IPM machines, to improve the position detection accuracy of VR resolvers, and to identify the influence of manufacturing tolerances. For IPM machines, optimal step-skewing methods are firstly proposed to reduce the residual cogging torque component caused by the axial flux interaction between rotor steps, followed by corrections accounting for the machine materials and the tolerance of axial assembling gaps. Further, the influence of manufacturing tolerances is investigated, i.e. PM diversity, rotor eccentricities, stator out-of-roundness and skewing angle errors, resulting in lower orders of additional cogging torque components which cannot be effectively reduced by skewing. Amongst random distributions of different manufacturing tolerances, the most sensitive cases are identified respectively for predicting the ultimate values of potential cogging torques. Moreover, the influence is also investigated and compared amongst different IPM machine designs (i.e. different rotor contours and different slot/pole combinations, etc.), as a reference for selecting proper designs for applications with strict requirements on cogging torque and torque ripple. In addition, the field spatial harmonics are analyzed with and without manufacturing tolerances considered, in order to illustrate the origins of additional cogging torque components. For VR resolvers, the basic operating principle and the winding configurations are firstly analyzed, based on which optimal stator slot (Z) and rotor saliency (p) number combinations are obtained. Further, a novel VR resolver topology with non-overlapping tooth-coil windings (NTWVRR) is proposed to simplify the manufacturing process of conventional VR resolvers, while the identical stator (and windings) of the proposed design can be employed in three different poles of resolvers. With actual application conditions considered, the proposed VR resolver proves to be a strong candidate of position sensor for hybrid electric/electric vehicle (HEV/EV) systems. Moreover, a novel design of rotor contour by injecting auxiliary air-gap permeance harmonics (AAPH rotor) is further proposed to eliminate the output voltage harmonics, which effectively improves the detection accuracy of the proposed and the conventional (Z=4p) VR resolver topologies. In addition, the influence of typical manufacturing tolerances (i.e. asymmetric coil locations and assembling eccentricities) on the detection accuracy is investigated and compared for different resolver designs. As verification of the foregoing analyses, several IPM machine and VR resolver prototypes are fabricated and tested on the cogging torque and the position detection accuracy respectively.

Cell culture has been a fundamental tool for the study of cell biology, tissue engineering, stem cell technology and biotechnology in general. It becomes more and more important to have a well-defined physiochemical microenvironment during cell culture. Conventional cell cultures employ expensive, manually controlled incubation equipment, making it difficult to maximize a cultures yield. Furthermore, previous studies use qualitative methods to assess cell culture proliferation that are inherently inaccurate and labour intensive, thereby increasing the cost of production. In addition, three dimensional cell culture, in scaffold, has been shown to provide more physiological relevant information as it mimic more accurate conditions that are similar to the physiological conditions of the human body compared with two dimension, which has special interest to regenerative medicine. Therefore, a portable and automated total-analysis-system (μTAS) was proposed with microenvironment control and quantitative analysis techniques to monitor cell proliferation and metabolic activity. The automated portable heating system was validated to be capable to maintain a stable physiochemical microenvironment, with little margin of error, for cellular substrate outside of conventional incubation. A standalone platform system was designed and fabricated with accurate temperature control by employing an optically transparent ITO-film with a large heating area. The transparency of the film is critical for continuous in-situ microscopic observation over long-term cell culture process. Previous studies have attempted to use ITO-film as a heating element, but were unable to distribute the heat evenly onto the microbioreactor platform. This nagging problem in the literature was improved through a novel film design. As a result, the ITO-film based heating system was evaluated and constructed successfully to serve as a heating element for long-term static cell culture with facilitated proliferation rate in gas-permeable PDMS microbioreactor outside of conventional incubation. In addition to maintaining a stable microenvironment, a non-invasive in-situ technology for monitoring cell viability and proliferation rate was constructed and developed based on bioimpedance spectroscopy (BIS). It was primarily focused on making decisions for structure and specification of proposed system-on a chip BIS measurement. The miniaturization of BIS system on microbioreactor platform was achieved by utilizing and integrating switching matrix array, impedance analyzer chip with reliable analogue-front-end circuitry. The realized system was verified with the DLD-1 cells and its monitored data were validated with conventional bioassays. Three dimensional cell cultures with scaffold is a key to the success of tissue engineering. Engineered cornea collagen scaffold may be feasible using re-seeding proper human cells onto a decellularized corneal scaffold. The quality of the scaffold and the interaction of the cells are critical to the key function (i.e transparency, haze and total transmittance) of final products. An integrated corneal collagen scaffold quality assessment system, via optical property inspection unit, was innovatively designed and realized with non-invasive and non-destructive characteristics. The H1299 cells were seeded onto inspected corneal scaffold and BIS system, which were realized in the previous chapter, were used to validate its applicability for 3D cell culture. The cell adhesion as an outcome at different scaffolds with different optical properties has revealed the importance of the microstructure of scaffold on the cell functions. The results showed the developed technologies can be used for the quality control of corneal scaffold and the fabricated μTAS not only enabled environmental control but, with BIS-based in-situ assay, it also facilitate the function (i.e adhesion) and viability monitoring with quantitative and qualitative analysis in 3D-alike cell culture. Additionally, by considering its low decontamination and cost-effective nature with compatibility for high-throughput screening applications, the fabricated and integrated systems has significant applications in tissue engineering.

This thesis contains three essays in evolutionary game theory. In the first chapter, we study the impact of switching costs on the long run outcome in 2X2 coordination games played in the circular city model of local interactions. We find that for low levels of switching costs, the risk dominant convention is the unique long run equilibrium. For intermediate levels of switching costs the set of long run equilibria contains the risk dominant convention but may also contain conventions that are not risk dominant. For high levels of switching costs also nonmonomorphic states will be included in the set of LRE. We study the impact of location heterogeneity on neighborhood segregation in the one-dimensional Schelling residential model in the second chapter. We model location heterogeneity by introducing an advantageous node, in which a player’s utility is impartial to the composition of her neighborhood. We find that when every player interacts with two neighbors, one advantageous node in the circular city will lead to a result that segregation is no longer the unique LRE. When players interact with more neighbors, more advantageous nodes are necessary to obtain the same result. In the third chapter, we consider a model of social coordination and network formation, where players of two groups play a 2X2 coordination game when connected. Players in one group actively decide on whom they play with and on the action in the game, while players in the other group decide on the action in the game only. We find that if either group’s population size is small in comparison to the linking restriction, all players will choose the risk dominant equilibrium, while when both groups are sufficiently large in population, the players of two groups will coordinate on the payoff dominant action.

This thesis presents new methodologies to improve performance of wireless cooperative networks using the Golden Code. As a form of space-time coding, the Golden Code can achieve diversity-multiplexing tradeoff and the data rate can be twice that of the Alamouti code. In practice, however, asynchronism between relay nodes may reduce performance and channel quality can be degraded from certain antennas. Firstly, a simple offset transmission scheme, which employs full interference cancellation (FIC) and orthogonal frequency division multiplexing (OFDM), is enhanced through the use of four relay nodes and receiver processing to mitigate asynchronism. Then, the potential reduction in diversity gain due to the dependent channel matrix elements in the distributed Golden Code transmission, and the rate penalty of multihop transmission, are mitigated by relay selection based on two-way transmission. The Golden Code is also implemented in an asynchronous one-way relay network over frequency flat and selective channels, and a simple approach to overcome asynchronism is proposed. In one-way communication with computationally efficient sphere decoding, the maximum of the channel parameter means is shown to achieve the best performance for the relay selection through bit error rate simulations. Secondly, to reduce the cost of hardware when multiple antennas are available in a cooperative network, multi-antenna selection is exploited. In this context, maximum-sum transmit antenna selection is proposed. End-to-end signal-to-noise ratio (SNR) is calculated and outage probability analysis is performed when the links are modelled as Rayleigh fading frequency flat channels. The numerical results support the analysis and for a MIMO system maximum-sum selection is shown to outperform maximum-minimum selection. Additionally, pairwise error probability (PEP) analysis is performed for maximum-sum transmit antenna selection with the Golden Code and the diversity order is obtained. Finally, with the assumption of fibre-connected multiple antennas with finite buffers, multiple-antenna selection is implemented on the basis of maximum-sum antenna selection. Frequency flat Rayleigh fading channels are assumed together with a decode and forward transmission scheme. Outage probability analysis is performed by exploiting the steady-state stationarity of a Markov Chain model.

For thousands of years, Nature has been considered to be essential to human beings for the treatment of their diseases. Amphibians are an important component of the Animal Kingdom and have played a significant role in sourcing many active and some therapeutic compounds. The secretions of amphibians act as defensive substances spreading out over their skins to protect them from predator attack and infections by microorganisms. These skin secretions represent a spectacular library of bioactive molecules, including peptides. In recent years, many different kinds of antimicrobial and pharmacological peptides have been isolated from amphibian skin secretions and some of these are providing potential human therapeutic leads. In the course of this research, genomic and proteomic techniques have been used to investigate the bioactive peptides from the skin secretions of four species of frog: the Central American red-eyed leaf frog, Agalychnis callidryas; the Peruvian purple-sided leaf frog, Phyllomedusa baltea; the European edible frog, Pelophylax kl. esculentus; the Fujian large-headed frog, Limnonectes fujianensis. The peptides isolated in these studies could be divided into two antimicrobial peptide families (balteatide and QUB3276) and two pharmacological peptide families (the bombesin-related peptides and AcT-2). The latter peptide was found to have both antimicrobial and pharmacological functions. Synthetic replicates of each peptide were prepared and used in an extensive range of biological assays.These discoveries of novel peptides from amphibian skin secretions have enriched our knowledge of bioactive peptides from this source and may provide the basis for several drug development programmes.

A content delivery network (CDN) typically consists of geographically-distributed data centers (DCs), which are deployed within proximity to end-users who request Internet content. Content copies are stored at the DCs and are delivered to end-users in a localized manner to improve service availability and end-to-end latency. On one hand, CDNs have improved QoS experienced by end-users. On the other hand, the rapid increase in Internet traffic volume has caused the global DC industry's energy usage to skyrocket. Therefore, our focus in this thesis is to realize energy awareness in CDN management while assuring end-to-end QoS. First, we surveyed the literature on energy-aware DC and CDN management schemes. We highlighted the significance of dynamically provisioning server and network resources in DCs in order to reduce DC energy usage. We also recognized that in order to achieve optimal CDN energy saving, energy optimization should be performed both within each DC and among multiple DCs in a CDN. Second, we proposed a theoretical framework that minimizes server power consumption in cross-domain CDNs. The term "server" refers to any co-locating entity that can handle user requests, e.g., server clusters or DCs. Our strategy was to put a subset of servers to sleep mode during off-peak hours to save energy. In order to avoid deteriorated QoS caused by less live server resources, we enforced constraints on utilization of servers and network links respectively to avoid them from being overloaded. Third, we designed an energy-aware CDN management system. The strategy was not only to put a subset of servers within each DC to sleep, but also to put entire DCs to sleep during off-peak hours through load unbalancing among DCs. We showed how the proposed system can be integrated within a typical modern CDN architecture. We also developed a heuristic algorithm that allows CDN operators to quickly make decisions on server and DC sleeping, as well as energy-aware request resolution. QoS was assured through constraints on server response time and end-to-end delay. Fourth, we built an optimization model that minimizes the overall energy consumption of CDN DCs, including their servers and cooling systems. We derived a lower bound to its optimal objective. Through comparing with the lower bound, we showed that our earlier developed heuristic algorithm's energy-saving gain was guaranteed to be near-optimal. We also quantitatively studied the trade-off between CDN energy saving and QoS performance in terms of end-to-end delay and server response time.

In this thesis, we develop some novel models and methods for the analysis of both structural and functional brain images, and their joint analysis with genetic data. In the first project, we present a suite of methods to increase the power of whole-brain genome-wide association studies. We introduce a kernel-based multilocus model to capture the interactions between single nucleotide polymorphisms (SNPs) and model their joint effect on the imaging traits. We provide a fast implementation of voxel- and cluster-wise inferences based on random field theory to take full use of the 3D spatial information in images and account for multiple comparison problems. We also propose a fast permutation procedure to increase the efficiency of standard permutation methods and provide accurate small p-value estimates based on parametric tail approximation. We explore the relationship between 448,294 SNPs and 18,043 genes in 31,662 voxels of the entire brain across 740 elderly subjects from the Alzheimer's Disease Neuroimaging Initiative (ADNI), and show boosted power of our approach by making head-to-head comparisons with previous voxel-wise genome-wide association studies. The various advantages of our methods over existing approaches indicate a great potential offered by this novel framework to detect genetic in uences on human brains. In the second project, we present a Bayesian spatial model of multiple sclerosis binary lesion maps based on a spatial generalized linear mixed model with spatially varying coefficients. Our model fully respects the binary nature of the data and the spatial structure of the lesion maps, as opposed to existing massive univariate methods, and produces regularized (smoothed) estimates of lesion incidence without an arbitrary smoothing parameter. Our model also allows for explicit modeling of the spatially varying effects of subject specific covariates such as age, gender, disease duration and disabilities scores, producing spatial maps of these effects and their significance, as well as the (scalar) effect of spatially varying covariates such as the fraction of white matter in each voxel. We apply our model to binary lesion maps derived from T2-weighted MRI images from 250 multiple sclerosis patients classified into five clinical subtypes, and determine the spatial dependence between lesion location and subject specific covariates. We also demonstrate dramatically improved predictive capabilities of our model over existing methods.

The technique of damping torque analysis (DTA) is proposed on the basis of linearized Philips-Heffron model of a single-machine infinite-bus power system. Analysis is to examine the electric torque contribution from a particular source to the electromechanical oscillation loop of generator in the power system. An effective engineering method - phase compensation method - for the design of PSS was developed on the basis of DTA. In study of real large power systems, modeling of the whole power system is very complicated. The generators in the large power system and other equipments result in a very high order of system model. Therefore a proper simplification method has to be applied to reduce the high order of real large power system. The aim of system simplification is to reduce the order of the linearized model of the large power systems, maintaining the characteristics of the under studied low frequency. Power system stabilizers (PSSs) have been adopted to mitigate low-frequency oscillations in the grid. However, in concerning of efficiency of damping controller in the power systems, decomposed analysis of PSS is studied to be further evidence for selecting installing location of PSS in the power systems. With greater integration of wind turbine generators in the introduced realistic large-scale power systems, impact of doubly fed induction generators (DFIGs) has been investigate for further study of renewable energy connected power systems. In terms of coordinated design of PMU and PSS in real large-scale power systems, developments in communication systems have made it feasible to also use remote signals as PSS inputs provided by PMU units. So as to the selection of signals from multiple locations can be further improved..

The need for a generic and adaptable object detection and recognition method in static images is becoming a necessity today, given the rapid development of the internet and multimedia databases in general. Computer vision is out-performed by human vision in terms of efficiency, accuracy and depth of understanding, as the computerised recognition is achieved at contextual level, in contrast to human vision that also considers semantics and background knowledge. In order to achieve recognition at a semantic level, computer vision systems must not only be capable of recognising objects, regardless changes in appearance, location, and action, but also be able to interpret the relationship of the object with its surrounding environment, as it appears in the image. This work reviews the state-of-the-art techniques in object recognition and proposes a supervised learning method based on adaptable models for detecting thematic categories of objects. Investigation is performed extracting visual features, which can be used to distinguish the object; and analysis is performed examining the semantic relations, which are embedded within the object’s inner-parts and its surrounding. The systematic identification and measurement of associations between components forming an object and the relationship of the object with its surrounding environment, allows for the automatic generation of object models. This method goes beyond classical image indexing and retrieval techniques, based on visual features only, and introduce semantics as a way of recognising and retrieving objects. The ultimate goal of this work is to develop a knowledge guided object recognition method. The proposed method is able to achieve objects recognition in real life images with reasonable high accuracy (83-87% for foreground objects recognition and 87-94% for background objects recognition). More importantly, semantic reasoning is introduced to guide and assist the recognition process. Instead of relying on purely visual similarity measures, objects are recognised based on their key co-observed components and their relations with the surrounding environment. The ability to filter out the background objects allowed the foreground objects to be recognised with much higher accuracy (10%-15% improvement). This method also improves the recognition speed dramatically, as the method will only be applied on specific areas of the image.

The growing requirements of a diversity of applications necessitate the deployment of large and powerful computing systems and failures in these systems may cause severe damage in every aspect from loss of human lives to world economy. However, current fault tolerance techniques cannot meet the increasing requirements for reliability. Thus new solutions are urgently needed and research on proactive schemes is one of the directions that may offer better efficiency. This thesis proposes a novel proactive failure management framework. Its goal is to reduce the failure penalties and improve fault tolerance efficiency in supercomputers when running complex applications. The proposed proactive scheme builds on two core components: failure prediction and proactive failure recovery. More specifically, the failure prediction component is based on the assessment of system events and employs semi-Markov models to capture the dependencies between failures and other events for the forecasting of forthcoming failures. Furthermore, a two-level failure prediction strategy is described that not only estimates the future failure occurrence but also identifies the specific failure categories. Based on the accurate failure forecasting, a prediction-based coordinated checkpoint mechanism is designed to construct extra checkpoints just before each predicted failure occurrence so that the wasted computational time can be significantly reduced. Moreover, a theoretical model has been developed to assess the proactive scheme that enables calculation of the overall wasted computational time.The prediction component has been applied to industrial data from the IBM BlueGene/L system. Results of the failure prediction component show a great improvement of the prediction accuracy in comparison with three other well-known prediction approaches, and also demonstrate that the semi-Markov based predictor, which has achieved the precision of 87.41% and the recall of 77.95%, performs better than other predictors.