13553 research outputs found
Sort by
Thrust and Efficiency Characterization of a Low-Power Applied-Field Magnetoplasmadynamic Thruster With a Superconducting Magnet
No full textSuperconducting magnets can become an enabling technology, offering new potential for applied-field magnetoplasmadynamic thrusters (AF-MPDTs). This study investigates the role of high, Tesla-level magnetic fields (up to 1 T) in AF-MPDTs operating in the low-power regime. More specifically, the AF-MPDT is operated with a lanthanum-hexaboride hollow cathode using argon flow rates below 2 mg/s, and anode discharge currents below 10 A. The reported results aim to characterize thruster performance in relation to both anode voltage and thrust. The thruster delivers up to 27.8 mN of thrust, with a peak thrust efficiency of 22% and a thrust-to-power ratio of 32 mN/kW. The highest reported discharge power was 1.73 kW. In the low-field regime (< 250 mT) the magnetic field increases thrust and thruster performance. Thrust is seen to increase linearly reaching a maximum at 500 mT–750 mT, depending on the combination of anode discharge current and argon flow rate. The ability of the thruster to convert swirl energy into axial acceleration is shown to decrease with increasing field, limiting performance in the high-field regime. In addition, it has been shown that the applied magnetic field can be used as an efficient mechanism for thrust modulation.</p
Mapping Noise Pollution Using Modelled and Crowdsourced Urban Noise Data
No full textTransport-generated noise pollution significantly burdens population health. Quantifying noise distribution and identifying areas with elevated noise levels is crucial for designing effective policy measures. We apply the CNOSSOS-EU numerical trans-port noise model to Pōneke Wellington, mapping traffic-induced noise exposure with good spatial coverage from public data. Crowdsourced noise data (CND) maps all noise sources and we spatially compare results. While unsuitable as a standalone method, CND highlights the impact of missing noise sources and offers a limited but promising validation tool for modelled data when alternatives are lacking
The double-edged sword of acculturation: Navigating work-family conflict among immigrants
No full textNo description supplie
The Impact of BCL6 Expression on Cancer Cell Behaviour in a Glioma Model
No full textGlioblastoma (GBM) is the most aggressive and malignant brain tumour, exhibiting some of the lowest survival rates among various cancers, even following interventions such as surgical resection, chemotherapy, and radiation therapy. GBM is characterised by its heterogeneity, invasive phenotype, and resistance to conventional treatment modalities. Key drivers of tumorigenesis play crucial roles in facilitating the ‘hallmarks’ of cancer, including uncontrolled proliferation, evasion of growth-regulatory mechanisms, and, critically, resistance to therapeutic interventions. A comprehensive understanding of these genetic drivers is essential for the development of innovative therapeutic strategies aimed at improving patient outcomes.BCL6, an established oncogenic driver in diffuse large B-cell lymphomas, also plays a significant role in solid tumours by repressing tumour suppressor genes and promoting survival under stress conditions. Preliminary studies indicate that the inhibition of BCL6 in certain tumour models, including GBM, correlates with reduced tumour growth and increased sensitivity to standard therapies.In this study, I utilized a newly developed genetically engineered mouse cell line model to investigate the role of BCL6 in GBM. The findings demonstrated that BCL6 overexpression significantly enhanced the proliferation rate of transfected GBM cells without affecting their migratory capacity. Furthermore, the data suggested that BCL6 expression contributes to improved survival rates following chemotherapy exposure. These results substantiate the hypothesis that BCL6 functions as a pivotal driver in the pathophysiology of GBM, positioning it as a promising target for therapeutic intervention. Future investigations will focus on determining whether these BCL6-overexpressing cell lines can initiate tumorigenesis and exploring the potential of BCL6 inhibitors as targeted therapeutic agents.</p
Topics in Computability
No full textThis thesis studies two topics in computability. The first is about computable metric and Polish spaces. We compare different notions of effective presentability and construct some spaces that are 'almost computable', in the sense that they do not have a computable presentation but they do have both left-c.e. and right-c.e. presentations. The second part studies c.e. Quasi-degrees (Q-degrees) and c.e. strong Quasi-degrees (sQ-degrees), which have interesting connections to algebra. We show that the c.e. sQ-degrees are not distributive, embed the lattice into them and show that no initial segment forms a lattice. We construct a non-computable c.e. set that has no c.e. simple set Q-below it. We also briefly study the relationship of sQ-degrees to wtt-degrees. Finally we show there is a minimal pair of sQ-degrees within the same Q-degree, and that if a degree is half of a minimal pair in the Q-degrees, it is also half of a minimal pair in the Turing degrees.</p
Probing the Evolution of Galaxy Clusters using SZ Effect and Non-thermal Emission
No full textGalaxy clusters are the largest gravitationally bound objects that are stable. They can contain hundreds or even thousands of galaxies, and can weigh as much as 10^15 times the mass of the Sun. About 15% of a cluster’s total mass is made up of the intracluster medium (ICM), while the remaining 5% consists of stars, gas, and dust found within the galaxies themselves. The majority of the cluster’s mass, around 80%, is thought to be made up of dark matter. Cluster mass, along with redshift, can link observations and theory allowing us to derive cosmological constraints from cluster number counts. However, measuring the mass of a cluster is still a challenge. Calibrating mass from ICM observables such as the Sunyaev-Zel’dovich (SZ) effect is subject to uncertainty and biases. The cause of biases and uncertainty is the assumption of hydrostatic equilibrium, while additional non-thermal pressure is not accounted for. On the other hand, merging cluster systems have been shown to exhibit radio emission which implies the presence of non-thermal electrons and a link with disturbances from hydrostatic equilibrium. In this thesis, I present work using a sample of clusters with SZ effect data from the Arcminute Microkelvin Imager and Planck, along with lower-frequency radio data from the Murchison Widefield Array and the LOw Frequency ARray (LOFAR).Using the SZ effect data, I study deviations of the galaxy cluster gas pressure profile from the average (universal) pressure profile. Meanwhile, with the low-frequency radio data, I investigate the presence and properties of non-thermal radio emission. By comparing the multiwavelength cluster properties, I investigate the connection between thermal and non-thermal electron populations, working toward the ultimate goal of obtaining unbiased and robust estimates of cluster mass.</p
The Graphic Novel as a Means to Explore Philosophies of the Subconscious and Memory. Helping Chinese youth contribute to the current discourse on the “Lying Flat” Crisis.
No full textCurrently, Chinese young people are under a lot of pressure, this has caused a series of social phenomena to emerged, such as the 996 working system (overtime hours), Lying flat (Zhang & Li 2022) (refusal to work), a rise in civil servant examination applicants (He 2022) (for secure jobs), and temple visits (Xiecheng Website 2023) (of a manifestation of hopeful thinking). In this thesis, I want to explore how using visuals and storytelling in a comic can contribute to the current discourse on these social issues confronting Chinese youth. To achieve this, I have read academic literature and graphic novels, integrating a broad spectrum of insights, and forming a solid grounding for the resulting comic. The philosophies of The Big Other (Jacque Lacan), trauma and healing (Slavoj Žižek), and The Angel of History (Walter Benjamin) form the theoretical background, help to explain the social issues Chinese young people are currently experiencing from a different angles. This thesis is a practice-based research project. Based on the literature review, my design output is a comic reflecting these social phenomenon from the perspective of a 25-year-old person. The research portfolio, including the comic and the thesis, serve as an example of how to use graphic novels to contribute to current social phenomenon.</p
Exploring the Micro Dynamics of Absorptive Capacity: A Systemic Mixed Method Approach
No full textManagement of absorptive capacity has the potential to increase the capability of organisations through the effective use of new knowledge generation and its use within organisations. There is a growing call from academics and practitioners to better understand the underlying structures at play within an absorptive capacity process.The broad aim of this research project is to model and explore the micro dynamics of an absorptive capacity process using a systemic mixed method approach. Located within absorptive capacity literature, this research seeks to investigate some of the key historical, contextual, and micro-dynamic aspects of absorptive capacity. On this account, absorptive capacity is taken to be a dynamic and emergent property of organisations that arises through complex interactions in a specific historical context and is not reducible to a specific set of causes.To accomplish this a sequential mixed method systemic framework was used consisting of system dynamics as the dominant method and soft systems methodology as the supplemental method applied to a business unit within The New Zealand Customs Service in the quest to increase their ability to respond to both internal and external forces. The study began with soft systems methodology at the problem structuring and qualitative causal loop modelling stage, which in turn, was used to inform the construction of the quantitative stock flow system dynamics model. Soft systems methodology was then used again with system dynamics at the final implementation and organisational learning stage of the systems thinking and modelling process proposed by Maani & Cavana (2011). A business unit within The New Zealand Customs Service, and other relevant interested parties, took part in the group model building session to generate the variables used to construct the qualitative causal loop diagrams of an absorptive capacity process using Vensim PLE software. The causal loop diagrams were then used to inform the construction of the system dynamics computer simulation model, constructed using Powersim Studio 10 software. Model validation consisted of feedback from personnel that included existing, past, present and those with a close working relationship to The New Zealand Customs Service.iii The qualitative group model building session resulted in a richly debated expression of ideas, culminating in a shared understanding of the barriers each face in building capability within The New Zealand Customs Service. Validating the model and testing different scenarios with interested New Zealand Custom Service parties helped them learn how their actions affect the accumulation and use of absorptive capacity, and identify behaviour changes they intend to make.In conclusion this research contributes to the management of absorptive capacity, and the research fields of absorptive capacity and systems thinking, through the application of a sequential mixed method approach to understand and inform how the micro level dynamic behaviours can affect the identification, assimilation and exploitation of new knowledge within organisations, not otherwise possible without a combined qualitative and quantitative systemic approach.</p
Prompting A Beautiful Young Woman: Gender Stereotypes and the Discursive Power of Image Generative AI
No full textImage generative artificial intelligence (AI) represents a troubling new tool in the social construction of gender. Text prompts, such as an attractive woman with […] perfect anatomy, perfect posture (Midjourney, 2023), create AI images at the rate of 34 million per day (Attie, 2023), indicative of this technology’s ascendant role in the global transmission of gender stereotypes. To investigate and illuminate the extent to which these gender discourses are perpetuated through generative AI platform Midjourney’s text prompts and generated images, I harness the affordances of a linguistic approach. Guided by a critical feminist stance, I utilise corpus linguistic (CL) tools to identify broad discursive patterns across user text prompts. I then perform a social semiotically-informed multimodal analysis (MMA) of AI-generated images, exemplifying and highlighting the CL results. Key findings reveal emerging discourses of normative femininity constructed on a vast scale, alongside those of confinement and hegemonic masculinity – the data analysis of which is deepened through attention to the “mini-narratives” within (Kress & Hodge 1979: 109). Discourses revolve around women as young, white, passive, and epitomising normative beauty standards while men are attributed action and strength-based qualities. Those beyond the binary or who do not align to ‘normative’ representations of appearance are all but erased. In an exciting analytical turn, the emergence of a discourse phenomenon I have labelled ‘algotext’ is discussed, a linguistic strategy platform users employ to evade content filters and elicit explicit images. My research aims to alert current and future image generative AI users to the technology’s sordid training origins, embedded gender biases, and the problematic content it is capable of eliciting and (re)producing. As a result, this research stands to benefit scholars from all linguistic, gender, and media studies fields.</p
YOLO Models for Instance Segmentation of Individual Tree Crowns from Aerial Imagery in Wellington
No full textThe instance segmentation task of individual tree crowns is an important real-world application that facilitates forest management, carbon storage estimation, and biodiversity modelling. Recently, Convolutional Neural Networks (CNNs) have achieved great success in computer vision. Several efforts have applied CNNs to perform instance segmentation of tree canopies. Unlike typical segmentation scenarios, aerial imagery of tree crowns often features densely distributed small and medium crowns, overlapping crowns, varied species, and challenging backgrounds. This variability poses significant challenges to traditional instance segmentation methods. You Only Look Once (YOLO) has recently gained popularity as a rapid and powerful approach for object detection and instance segmentation. Its one-stage design is computationally efficient and particularly appealing for large-scale imagery analysis. Nevertheless, standard YOLO models may struggle with very small or overlapping tree crowns, scale variation, and subtle inter-class differences, underscoring the need for specialized enhancements when applied to canopy segmentation tasks.The overall goal of this thesis is to address these challenges by leveraging YOLO-based instance segmentation methods and tailoring it to the unique requirements of aerial tree crown identification and species classification. Specifically, this research focuses on designing robust detection frameworks optimized for small and medium objects, devising novel multi-scale feature extraction and fusion techniques, and crafting effective yet efficient network architectures for aerial canopy data.First, this thesis proposes a new YOLO method for identifying individual tree crowns based on YOLOv7. It introduces a specialized detection mechanism for small and medium tree crowns, employs dense connectivity in the backbone to reuse feature maps across layers, and incorporates an efficient attention module to capture long-range dependencies. Experiments on the Wellington, New Zealand (NZ), aerial canopy dataset demonstrate that this method achieves higher detection and segmentation accuracies than other commonly used baselines.Second, this thesis proposes a new efficient YOLOv8 method, optimized for precise instance segmentation and species classification of tree crowns. This method includes new schemes for selecting candidate positive samples for each instance and a refined network design tailored for small and medium tree crowns. Adjustments in hyperparameters, particularly within the Task-Aligned Assigner, are also discussed to better suit canopy segmentation tasks. Comprehensive experiments conducted on the canopy dataset demonstrate that the new approach not only outperforms a number of advanced methods in terms of the Box AP and Mask AP metrics but also achieves a substantial decrease in parameters and model complexity.Finally, this thesis proposes a feature fusion technique based on the YOLOv8 architecture to address diverse canopy sizes. The new method incorporates a feature fusion mechanism that includes both cross-scale and same-scale fusion methods, enhancing the model's ability to integrate information across different layers and scales. Large convolution operations are employed to effectively extract key features, helping the model capture richer and deeper global information in the image. Experimental results on the canopy dataset demonstrate that the new method further advances performance, marking a promising solution for accurate and efficient instance segmentation of individual tree crowns in aerial imagery.</p
Access Repository Dashboard
Do you manage Victoria University of Wellington? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!