2,721 results for 2000, Doctoral

  • Inverse problems in astronomical imaging

    Johnston, Rachel Anne (2000)

    Doctoral thesis
    University of Canterbury Library

    The imaging of astronomical objects is limited by atmospheric turbulence, which consists of layers of varying refractive index surrounding the earth. These refractive index fluctuations are a direct consequence of the warming and cooling of air and water vapour in the atmosphere. Wavefronts entering the atmosphere acquire phase distortions, which when propagated result in amplitude fluctuations known as scintillation. Hence the practical manifestation of the atmosphere is a degradation of the signals passing through it, for example it severely limits the resolution of images captured by ground-based telescopes. A variety of solutions, or inverse problems, have been proposed and trialed in the attempt to obtain the best possible images from astronomical telescopes. An orbiting telescope (for example the Hubble space telescope) is one solution. In this case light is captured before it is distorted by the atmosphere. Less expensive ground-based solutions include the post processing of short exposure images and real-time compensation using adaptive optics, both of which are investigated in this thesis. However, the success of an inverse problem lies in the accurate modelling of the processes that give rise to the corresponding forward problem, in this case the random refractive index fluctuations that characterise the atmosphere. Numerical simulation of atmospheric turbulence is achieved using phase screens in which the assumption of Kolmogorov statistics is often made. A previously presented method for modelling Kolmogorov phase fluctuations over a finite aperture, the midpoint displacement method, is both formalised and improved. This enables the accurate generation of atmospheric speckle images for the development and testing of post processing methods. Another aspect of the forward problem is the accurate simulation of scintillation, resulting from the propagation of phase distorted wavefronts. Commonly used simulation methods achieve this by assuming periodic boundary conditions. A technique for the accurate modelling and simulation of scintillation from an aperiodic Kolmogorov phase screen is presented. The more physically justifiable assumption of smoothness is shown to result in a propagation kernel of finite extent. This allows the phase screen dimensions for an accurate simulation to be determined and truncation can then be used to eliminate the unwanted spectral leakage and diffraction effects usually inherent in the use of finite apertures. Deconvolution methods are popular for the post processing of atmospheric speckle images to compensate for the effects of the atmosphere. Conventional deconvolution algorithms are applied when the distortion is known or well-characterised, whereas, blind deconvolution algorithms are used when the distortion is unknown. Conventional deconvolution techniques are not often directly applied to astronomical imaging problems as the distortion introduced by the atmosphere is unknown. However, their extension to blind deconvolution is straightforward and hence their development is valuable. The ill-conditioning of the deconvolution problem requires the addition of prior information, such as positivity, to enable its solution. It is shown that the conventional deconvolution problem can be reformulated as an equivalent quadratic programming problem. Consequently, an accelerated quadratic programming approach is applied and shown to be an improvement to an existing method used for enforcing positivity in deconvolution applications. The main algorithmic differences of the new method are implementation via the fast Fourier transform (FFT) and guaranteed convergence to the constrained minimum. Blind deconvolution is also an interesting problem that may arise in many fields of research. It is of particular relevance to imaging through turbulence where the point spread function can only be modelled statistically, and direct measurement may be difficult. The extension of the quadratic programming method to blind deconvolution, combined with Tikhonov-Miller regularisation (energy constraints), smoothness constraints, penalty terms and statistical priors produced a series of new algorithms. The performance of these algorithms is illustrated on simulated astronomical speckle images. Ground-based adaptive optics (AO) technologies are an alternative to post processing methods and aim to compensate for the distortion introduced by the atmosphere in real-time. Knowledge of the vertical structure of the atmosphere combined with AO provides the potential for compensation over a wide field of view. However, the continually changing nature of atmospheric turbulence places strict requirements on techniques for determining the turbulence structure. The remote sensing of scintillation data to estimate this information is known as scintillation detection and ranging (SCIDAR). Application of SCIDAR methods to the capture and analysis of experimental data, as demonstrated in this thesis, highlighted a number of problems with the technique. Methods for overcoming these difficulties are discussed and demonstrated. Finally, alternative approaches to the estimation of atmospheric turbulence profiles and a proposed new technique are investigated.

    View record details
  • Informed preferences in forest-based land use planning in Indonesia : a methodological case study.

    Rahardja, Teguh (2005)

    Doctoral thesis
    University of Canterbury Library

    Indonesia is large and rich in natural resources. Its forest extends over 60 per cent of the country's land and contains many other natural resources. There are many stakeholders, often with conflicting interests. The demands placed on the forest have resulted in declining quantity and quality of the forest lands. People have recognised the need for reviewing and improving the forest-based land use plan, and, in so doing, promoting the participatory approach rather than the traditionally centralistic one. This has been attempted, but there were difficulties in the participatory evaluation of land use options' impacts. Therefore, this study aims to develop a method to help forestry-based land use planning take into account stakeholders' preferences after considering land use scenario consequences. Based on the situation in Indonesia and existing options, this study adopted the mixed rational-participatory approach. The rational side was attempted by FOLPI simulation of land use scenarios. An interview survey of opinions suggested eight scenarios of varying emphases on the economic, ecological and social aspects, which were simulated in FOLPI with area and resource data of each land use. The results were graphs of land use changes and their economic, ecological and social impacts. The participatory aspect was promoted by Q methodology applications. Q was used to analyse respondents' sorts of a set of statements about different aspects of land use planning, and revealed the typology and preferences of stakeholders with regard to land use planning. Using verbal statements in such exercises discovered the typology and normative preferences, while using the FOLPI application graphs as the statements disclosed the positive preferences. In tandem, they provide useful information as inputs to stakeholder deliberations towards a new, rational, and acceptable land use scenario. This study, therefore, recommends a method to help forest-based land use planning stakeholders. The method includes FOLPI simulation of the broad-scoped land use scenarios, and Q applications both the conventional verbal way and the innovative graphical way.

    View record details
  • Advanced surface texturing for silicon solar cells

    Ganesan, Kumaravelu (2005)

    Doctoral thesis
    University of Canterbury Library

    The multi-crystalline silicon (me-Si) solar cell is considered to be one of the most promising cells capable of achieving high efficiency at low cost and high reliability. Improving solar cells efficiency using low cost materials requires careful design considerations aiming to minimise the optical and electrical losses. In this work plasma texturing was employed to reduce optical reflections from silicon surfaces well below 1%. Plasma texturing is used to form light trapping structures suitable for silicon solar cells. Several plasma texturing methods are investigated and associated defects are analysed. Masked as well as mask-less texturing techniques are investigated. Conventional parallel plate Reactive Ion Elching (RIE), Inductively Coupled Plasma (ICP) and Electron Cyclotron Resonance (ECR) plasma system are used to compare the plasma induced defects in silicon. The influence of various plasma etch parameters on plasma induced defect is investigated. A correlation between the minority carriers lifetime and surface area increased by texturing is established. Effective lifetime measurements using Quasi Steady State Photo Conductive (QSSPC) technique is mainly used to estimate the plasma induced defect in textured silicon substrates. Sinton lifetime tester is used to measure the effective lifetime of the substrates. The implied open circuit voltage is calculated from the lifetime data for textured substrates. In this work low temperature photoluminescence spectroscopy is also used to analyse the defect caused by plasma on me-Si substrates. Photoluminescence (PL) data is obtained using the 514.5 nm line of an Ar⁺ laser as an excitation source. The luminescence is dispersed with SPEX 1700 spectrometer with a liquid nitrogen cooled Germanium detector. Reflectance measurements are performed on textured surfaces usmg a purpose built integrating sphere attachment of a high accuracy spectrophotometer. Modelling is also performed using PV-optics software to compare the experimental and theoretical results. Finally, silicon solar cells are fabricated with measured efficiency around 18% . The efficiency is estimated from the I-V characteristics data obtained using a calibrated halogen lamp and a HP semiconductor parameter analyser. Spin-on-dopant source as well as solid diffusion source is used to form the ewitter junction of the solar cells fabricated on p-type silicon wafers. Multicrystalline silicon, CZ- silicon and FZ silicon wafers are used to fabricate solar cells in this thesis. The effect of single and double layer antireflection coatings on diffused reflections is also investigated.

    View record details
  • Multi-level voltage and current reinjection ac-dc conversion.

    Liu, Yonghe (2003)

    Doctoral thesis
    University of Canterbury Library

    This thesis describes a new concept of multi-level reinjection ac-dc conversion, its main purpose being a further reduction of the harmonic content, a solution of dynamic voltage balancing for direct series connected switching devices and an improvement of high power converter efficiency and reliability. It is a combination of the multi-level, soft switching and reinjection concepts. A variety of configurations are proposed, based on the new concept, to achieve efficient voltage and current conversion. For each configuration the firing sequences, waveform analysis, steady and dynamic performances and close-loop control strategies are presented, and particular applications suggested. The ideal reinjection waveforms are first derived for perfect harmonic cancellation and then fully symmetrical approximations are made for more practical implementations. This is followed by a description and comparison of the generation circuits required for the implementation of the multi-level symmetrical reinjection waveforms. A three-level voltage reinjection scheme, implemented by adding a reinjection bridge and a reinj ection transformer to the standard twelve-pulse converter, is discussed in great detail, both for the series and parallel connections. This is followed by an investigation into the possible application of these converters to Back to Back VSC HV de interconnection; the analysis is validated by EMTDC simulations. A multi-level voltage reinjection VSC is also proposed, which uses a controllable de voltage divider to distribute the de source voltage to the two main bridges and produces high quality output waveforms. The voltage and current waveforms, the firing sequences and the capacitor voltage balancing are analyzed and verified by EMTDC simulations. In particular, the proposed VSC is shown to be an ideal solution for the STATCOM application. The multi-level reinjection CSC alternative is also described and shown to exhibit an excellent performance in the STATCOM application.

    View record details
  • Improved shortest path algorithms for nearly acyclic graphs

    Saunders, Shane (2004)

    Doctoral thesis
    University of Canterbury Library

    Dijkstra's algorithm solves the single-source shortest path problem on any directed graph in O(m+nlogn) worst-case time when a Fibonacci heap is used as the frontier set data structure. Here n is the number of vertices and m is the number of edges in the graph. If the graph is nearly acyclic, then other algorithms can achieve a time complexity lower than that of Dijkstra's algorithm. Abuaiadh and Kingston gave a single source shortest path algorithm for nearly acyclic graphs with O(m + nlogt) worst-case time complexity, where the new parameter t is the number of delete-min operations performed in priority queue manipulation. For nearly acyclic graphs, the value of t is expected to be small, allowing the algorithm to outperform Dijkstra's algorithm. Takaoka, using a different definition for acyclicity, gave an algorithm with 0 ( m + n log k) worstcase time complexity. In this algorithm, the new parameter k is the maximum cardinality of the strongly connected components in the graph. This thesis presents several new shortest path algorithms that define trigger vertices, from which efficient computation of shortest paths through underlying acyclic structures in the graph is possible. Various definitions for trigger vertices are considered. One definition decomposes a graph into a unique set of acyclic structures, where each single trigger vertex dominates a single corresponding acyclic structure. This acyclic decomposition can be computed in O(m) time, thus allowing the single source problem to be solved in 0 ( m + r log r) worst-case time, where r is the resulting number of trigger vertices in the graph. For nearly acyclic graphs, the value of r is small and single-source can be solved in close to O(m) worst-case time. It is possible to define both monodirectional and bidirectional variants of this acyclic decomposition. This thesis also presents decompositions in which multiple trigger vertices dominate a single acyclic structure. The trigger vertices of such decompositions constitute feedback vertex sets. If trigger vertices are defined as a set of precomputed feedback vertices, then the all-pairs shortest path problem can be solved in O(mn + nr2 ) worst-case time. This allows all-pairs to be solved in O(mn) worst-case time when a feedback vertex set smaller than the square root of the number of edges is known. For suitable graph types, these new algorithms offer an improvement on the time complexity of previous algorithms.

    View record details
  • Formulations of New Zealand identity : re reading Man alone, The bone people and Once were warriors.

    Stachurski, Christina (2001)

    Doctoral thesis
    University of Canterbury Library

    John Mulgan's Man Alone (1939), Keri Hulme's the bone people (1983) and Alan Duffs Once Were Warriors (1990) are considered in terms of developments in settler acculturation. These three novels' contents and implications can be thought of as marking distinct stages in the general formulation and experience of collective Pakeha identity through cultural discourse: the late colonial disregard and distancing of Maori; the anti-colonial embrace of the Maori as a means of claiming indigeneity; and the stage of internal de-colonisation in which Maori are once again cast as other and scapegoated. My study focuses upon personal identity in tandem with collective identity, as representations of race and/or ethnicity are commonly enmeshed with constructions of sex, gender and sexuality (as well as what can loosely be called geography) in these novels. As a part of this process, I test in particular the thesis that the various versions of Maoriness represented in Man Alone, the bone people and Once Were Warriors are a crucial factor in these novels' cultural significance for Pakeha. At the same time, I focus upon the ways in which these various versions of Maoriness signify both the concept "Pakeha" - or otherwise - and complexities within Pakeha.

    View record details
  • Multi-level voltage and current reinjection ac-dc conversion

    Liu, Yonghe (2003)

    Doctoral thesis
    University of Canterbury Library

    This thesis describes a new concept of multi-level reinjection ac-dc conversion, its main purpose being a further reduction of the harmonic content, a solution of dynamic voltage balancing for direct series connected switching devices and an improvement of high power converter efficiency and reliability. It is a combination of the multi-level, soft switching and reinjection concepts. A variety of configurations are proposed, based on the new concept, to achieve efficient voltage and current conversion. For each configuration the firing sequences, waveform analysis, steady and dynamic performances and close-loop control strategies are presented, and particular applications suggested. The ideal reinjection waveforms are first derived for perfect harmonic cancellation and then fully symmetrical approximations are made for more practical implementations. This is followed by a description and comparison of the generation circuits required for the implementation of the multi-level symmetrical reinjection waveforms. A three-level voltage reinjection scheme, implemented by adding a reinjection bridge and a reinj ection transformer to the standard twelve-pulse converter, is discussed in great detail, both for the series and parallel connections. This is followed by an investigation into the possible application of these converters to Back to Back VSC HV de interconnection; the analysis is validated by EMTDC simulations. A multi-level voltage reinjection VSC is also proposed, which uses a controllable de voltage divider to distribute the de source voltage to the two main bridges and produces high quality output waveforms. The voltage and current waveforms, the firing sequences and the capacitor voltage balancing are analyzed and verified by EMTDC simulations. In particular, the proposed VSC is shown to be an ideal solution for the STATCOM application. The multi-level reinjection CSC alternative is also described and shown to exhibit an excellent performance in the STATCOM application.

    View record details
  • Hollow-core floor slab performance following a severe earthquake.

    Matthews, Jeffrey (2004)

    Doctoral thesis
    University of Canterbury Library

    Hollow-core floor slabs are the dominant flooring systems used in New Zealand since the 1980's. This study experimentally investigates the seismic performance of precast hollow-core floors including three-dimensional effects of an entire floor system within a two-way moment resisting frame. In order to experimentally assess the seismic perfo1mance of a large super-assemblage a new type of self-equilibrating loading frame was designed and built. A full-scale super-assemblage based on a multi-storey prototype was constructed and tested under quasi-static cyclic loading. The capacity designed precast concrete frame perf01med very well but the performance of the floor itself was quite poor. Incipient failure of the precast floor occurred at an interstorey drift of 1.9 percent, while complete collapse of the floor occurred at a drift of 2.5 percent. A rainflow counting method is developed to enable the amount of beam elongation to be predicted during an earthquake. This is particularly impmiant in determining the required seating length for the precast hollow-core flooring units. Based on the results of this investigation new connection (seating) details are proposed for attaching the hollow-core units to the supporting beams to try to improve the performance of the hollow-core units.

    View record details
  • An organisation theory perspective on choice of franchising form

    Floyd, Callum (2001)

    Doctoral thesis
    University of Canterbury Library

    This research addresses the issue of diversity of organisational forms in franchising, which despite its prevalence, is poorly understood. The research focuses on the choice between three categories, consisting of five common types, of franchising: single-unit franchising, direct multi-unit (sequential franchising and area development) and indirect multi-unit (area representation and subfranchising) franchising forms. The thesis presents and tests a contingency model that explores the influence of environmental (munificence, complexity and dynamism) and task (task complexity) uncertainties on choice of franchising form. Six factors were operationalised to represent environmental and task uncertainties. These factors included demand size and growth (environmental munificence), demand dispersion and heterogeneity (environmental complexity), intensity of rivalry (environmental dynamism) and task complexity (task uncertainty). A multi-case study research strategy was conducted to test the contingency model. The strategy involved interviews with founders, other franchisor executives and franchisees, and also considered documentation and direct observations. The sample comprised a heterogeneous selection of seven New Zealand founded franchise systems. Companies were theoretically selected to ensure all five types of franchising were represented. The findings illustrated general support for the thesis that environmental and task uncertainties do influence choices made between alternative franchising forms. Most companies adopted types of franchising that were consistent with expectations derived from the model. Importantly, however, the overall fit was not neat and conclusive. The explanatory power of individual factors varied and in some situations form choices occurred contrary to expectations. This research produced further important findings. The qualitative methodology employed helped uncover five further drivers of franchising form choice. These additional factors related to individual choice and the firm, and included incentives, growth aspirations, need for control, resource constraints and franchisee aspirations. The findings also confirmed that no one factor or theory was sufficient to explain form choice, and the factors important in one company's decision might have little relevance to another's. Consequently, multiple perspectives were necessary to understand the decisions made by franchisors.

    View record details
  • Formulations of New Zealand identity : re reading Man alone, The bone people and Once were warriors

    Stachurski, Christina (2001)

    Doctoral thesis
    University of Canterbury Library

    John Mulgan's Man Alone (1939), Keri Hulme's the bone people (1983) and Alan Duffs Once Were Warriors (1990) are considered in terms of developments in settler acculturation. These three novels' contents and implications can be thought of as marking distinct stages in the general formulation and experience of collective Pakeha identity through cultural discourse: the late colonial disregard and distancing of Maori; the anti-colonial embrace of the Maori as a means of claiming indigeneity; and the stage of internal de-colonisation in which Maori are once again cast as other and scapegoated. My study focuses upon personal identity in tandem with collective identity, as representations of race and/or ethnicity are commonly enmeshed with constructions of sex, gender and sexuality (as well as what can loosely be called geography) in these novels. As a part of this process, I test in particular the thesis that the various versions of Maoriness represented in Man Alone, the bone people and Once Were Warriors are a crucial factor in these novels' cultural significance for Pakeha. At the same time, I focus upon the ways in which these various versions of Maoriness signify both the concept "Pakeha" - or otherwise - and complexities within Pakeha.

    View record details
  • Modelling and measurement of diode rectifiers and their interaction with shunt active filters

    Laird, Hamish Duncan (2001)

    Doctoral thesis
    University of Canterbury Library

    The use of power conditioning equipment, such as the shunt active filter, is becoming more popular to remedy power quality problems associated with non-linear rectifier loads. Typically in the shunt active filter situation the non-linear load is considered to be insensitive to the terminal voltage variation. This assumption is made tacitly by modelling the non-linear load as a constant current source. In fact all loads, including non-linear rectifier type loads, are sensitive to voltage variation at their terminals. Since a shunt active filter changes the terminal voltage when it operates by changing the current flowing in the AC system, the shunt active filter can change the current in the non-linear load. The interaction of the non-linear load with the shunt active filter has implications for the power quality delivered by and to that load. In order to determine the interaction of the non-linear load with the shunt active filter and the AC system a control system approach is taken. This shows that the non-linear load contributes to the control behaviour of the shunt active filter because it forms part of the forward control transfer. A suitable model for the non-linear load is required because classical linear system cannot accurately represent the non-linearity. A small-signal frequency domain model is used to represent the non-linear load. This model completely and accurately includes the modulation with a frequency transfer matrix by including the phase dependent nature of the modulation results. The frequency domain model ensures that modulation is represented in a linear way. This means the evaluation of the non-linear load response is achieved by the solution of the linear equation set given by its frequency transfer matrix. The very common single-phase rectifier is analysed using this frequency domain approach. The two rectifier small signal transfer mechanisms, by which it connects its AC and DC sides, are the base switching and the switching instant modulation. Partitioning the rectifier into appropriate partial transfers and then using the small signal approach allows both these mechanisms to be analysed. The effect of the switching instant modulation is found to be second order and so is ignored in this analytic model. The component transfers and the total transfer are validated by time domain simulation. The model shows excellent accuracy. Experimental measurements of the single-phase rectifier are made by injecting AC current perturbations with a DSP controlled converter and measuring the small signal load current response. These measurements show good correlation to those predicted by the analytic model. The analytic model also allows the calculation of the effect of shunt active filter operation on the single-phase rectifier current. Measurements of a threephase rectifier are made and the first order sequence coupling nature of this device for both continuous and discontinuous DC side current conduction is demonstrated. This shows that the three-phase rectifier behaviour can be modelled and analysed using the same frequency transfer matrix approach as used for the single-phase rectifier.

    View record details
  • Hollow-core floor slab performance following a severe earthquake

    Matthews, Jeffrey (2004)

    Doctoral thesis
    University of Canterbury Library

    Hollow-core floor slabs are the dominant flooring systems used in New Zealand since the 1980's. This study experimentally investigates the seismic performance of precast hollow-core floors including three-dimensional effects of an entire floor system within a two-way moment resisting frame. In order to experimentally assess the seismic perfo1mance of a large super-assemblage a new type of self-equilibrating loading frame was designed and built. A full-scale super-assemblage based on a multi-storey prototype was constructed and tested under quasi-static cyclic loading. The capacity designed precast concrete frame perf01med very well but the performance of the floor itself was quite poor. Incipient failure of the precast floor occurred at an interstorey drift of 1.9 percent, while complete collapse of the floor occurred at a drift of 2.5 percent. A rainflow counting method is developed to enable the amount of beam elongation to be predicted during an earthquake. This is particularly impmiant in determining the required seating length for the precast hollow-core flooring units. Based on the results of this investigation new connection (seating) details are proposed for attaching the hollow-core units to the supporting beams to try to improve the performance of the hollow-core units.

    View record details
  • Terahertz diffractive optics

    Walsby, Edward David (2003)

    Doctoral thesis
    University of Canterbury Library

    Terahertz radiation research is an emerging field with significant work only being possible in the last decade with the development of new high speed laser equipment. Sources of terahertz are a continuing problem however several methods now available are making this technology more reliable and less expensive. With these developments much research is being conducted into the potential applications of this window in the electromagnetic spectrum. An integral part of any such application is the ability to focus and manipulate beams. Traditional optics systems achieve this with reflective, refractive or diffractive elements the later of which has not been utilised in the terahertz frequencies and is the subject of this thesis. The wavelength range of terahertz waves of the order of hundreds of micron make fabrication difficult with milling or machining. Planar fabrication techniques using microfabrication technology have been used to create micro-optics for much shorter wavelengths. This project has involved developing a microfabrication process that can be applied to the large structure sizes and depths required at terahertz frequencies. High diffraction efficiency demands the structuring of multiple level structures. This was achieved using a repeated binary fabrication process in silicon. A set of Fresnel lenses were produced and some anti-reflection structures to demonstrate the application of this technology to the formation of high quality terahertz diffractive optics. Different applications may demand the use of either a continuous wave or pulsed emission and detection. An appreciation of the devices' performance in a single frequency and pulsed broadband system is given. These systems were set up for the measurements of anti-reflection gratings and both single and pairs of Fresnel lenses of varying complexity. These results demonstrate potential uses as frequency and or spatial filters and focusing elements for tomography applications. Using Fresnel and Fraunhofer diffraction theory lenses have been simulated to further explain the results measured and to give an appreciation of how design modifications can be used to improve the efficiency. Because the fabrication process described is tailored for the construction of a broad range of optics in the terahertz frequency range there are a wide range of potential feature sizes and etch depths. A general guide to how processing defects inherent in the microfabrication process can effect lens performance is investigated to give a designer tolerances to remain within during production.

    View record details
  • Improved shortest path algorithms for nearly acyclic graphs

    Saunders, Shane (2004)

    Doctoral thesis
    University of Canterbury Library

    Dijkstra's algorithm solves the single-source shortest path problem on any directed graph in O(m+nlogn) worst-case time when a Fibonacci heap is used as the frontier set data structure. Here n is the number of vertices and m is the number of edges in the graph. If the graph is nearly acyclic, then other algorithms can achieve a time complexity lower than that of Dijkstra's algorithm. Abuaiadh and Kingston gave a single source shortest path algorithm for nearly acyclic graphs with O(m + nlogt) worst-case time complexity, where the new parameter t is the number of delete-min operations performed in priority queue manipulation. For nearly acyclic graphs, the value of t is expected to be small, allowing the algorithm to outperform Dijkstra's algorithm. Takaoka, using a different definition for acyclicity, gave an algorithm with 0 ( m + n log k) worstcase time complexity. In this algorithm, the new parameter k is the maximum cardinality of the strongly connected components in the graph. This thesis presents several new shortest path algorithms that define trigger vertices, from which efficient computation of shortest paths through underlying acyclic structures in the graph is possible. Various definitions for trigger vertices are considered. One definition decomposes a graph into a unique set of acyclic structures, where each single trigger vertex dominates a single corresponding acyclic structure. This acyclic decomposition can be computed in O(m) time, thus allowing the single source problem to be solved in 0 ( m + r log r) worst-case time, where r is the resulting number of trigger vertices in the graph. For nearly acyclic graphs, the value of r is small and single-source can be solved in close to O(m) worst-case time. It is possible to define both monodirectional and bidirectional variants of this acyclic decomposition. This thesis also presents decompositions in which multiple trigger vertices dominate a single acyclic structure. The trigger vertices of such decompositions constitute feedback vertex sets. If trigger vertices are defined as a set of precomputed feedback vertices, then the all-pairs shortest path problem can be solved in O(mn + nr² ) worst-case time. This allows all-pairs to be solved in O(mn) worst-case time when a feedback vertex set smaller than the square root of the number of edges is known. For suitable graph types, these new algorithms offer an improvement on the time complexity of previous algorithms.

    View record details
  • Effect of varying hysteresis models and damage models on damage assessment of r/c structures under standard design level earthquakes obtained using a new scaling method

    Dong, Ping (2003)

    Doctoral thesis
    University of Canterbury Library

    A seismic damage assessment of a real ductile framed structure of reinforced concrete requires not only realistic damage indices for members, storeys and the whole structure, but also a hysteresis rule simulating strength, stiffness and energy dissipation characteristics of the member hysteretic behaviour. Many member damage indices and hysteresis models have been developed [Carr 1998]. Varying the member damage indices and the hysteresis models may result in significant variations in the predicted damage indices for storeys and structures subjected to design level earthquake excitations at ultimate limit state specified in New Zealand loading standard [NZS4203 1992]. This could greatly influence correct engineedng decisions of a structural design engineer, hence the effect of varying member damage indices and hysteresis models on the damage indices for the storeys and structures should be identified. To this purpose, three main aspects studied include the earthquake scaling, the relationship between the member curvature ductility and structural displacement ductility, and the effect of varying hysteresis models and member damage indices on the damage indices for the storeys and structures respectively. Three structures, four earthquakes, eight hysteresis rules and four member damage indices were employed in this study for structural models, earthquake inputs, modelling the inelastic behaviour in members and member damage respectively. The three reinforced concrete ductile frames are 6, 12 and 18 storeys respectively, designed according to the current New Zealand Standards [NZS4203 1992, NZS3101 1995] using the capacity design philosophy [Paulay 1992] with a structural displacement ductility of 5.0. The four different past earthquake excitations are Bucharest (1977-NS), El Centro (1940-NS), Northridge (Sylmar-949NW) and Kobe (1995-NS). The eight hysteresis models are the Elasto-Plastic, Bilinear, Modified Takeda (α=0.0, β=0.6), Degrading Bilinear (α=0.5), Clough, Modified Takeda (α=0.3, β=0.4), Q-Hyst (α=0.5) and Origin-Centred hysteresis models. The four member damage indices are the Park & Ang, Roufaiel & Meyer, Cosenza et al and Banon & Veneziano. The storey and structural damage indices, quantifying the storey and structural damage measures respectively, were calculated as the energy weighted average of all the inelastic member damage indices in the storeys and overall structures respectively (Park & Ang method). To have appropriate scaled earthquakes matching design-level requirements, six different scaling methods are used to scale the four earthquakes for the three structures. By comparing the maximum responses (the maximum base shears, interstorey drifts and spectral accelerations), resulting from carrying out elastic dynamic time-history analyses to the scaled earthquakes with those at design level, a new procedure for earthquake scaling is proposed. To check whether the member curvature ductility demand to the design level earthquakes is less than the member curvature ductility capacity, the relationship between the member curvature ductility and the structural displacement ductility was studied and identified by carrying out inelastic dynamic time-history analyses. For this purpose, the Carr & Tabuchi trend-line approach for defining the structural yield displacements was used. By comparative studies of the storey and structural damage indices for a specified member damage index and a specified hysteresis model, the effect of varying the hysteresis models and member damage indices on the storey and overall structural damage indices, i.e. damage evaluations, are identified. Finally a procedure for the seismic damage analyses of reinforced concrete ductile framed structures to design-level earthquake excitations is proposed, which is illustrated by computing storey and structural damage indices for evaluations of 6- and 12-storey structures responding to the El Centro (1940-NS) excitation.

    View record details
  • Writing/Righting Menstruation: a Feminist Analysis of New Zealand Women's Knowledge of the Menstrual Cycle

    Culling, Vicki Marie (2001)

    Doctoral thesis
    Victoria University of Wellington

    This thesis investigates New Zealand women's menstrual knowledge within a cultural, social and historical context. An analysis or dominant menstrual discourses and their impact on women's menstrual knowledge was undertaken from a feminist poststructural perspective. At the outset, my initial objectives were to examine and record the social construction of menstruation and to determine the extent to which it impacted on New Zealand women's menstrual knowledge. Following a reading of feminist poststructuralism, the initial objective of measuring and quantifying women's menstrual knowledge changed to an approach focusing on discourse. Similarly, I moved to a new methodological focus on feminist epistemologies. As a result, the thesis examines the effects of New Zealand cultural practices and social meanings on women's 'knowing' about menstruation. It seeks to establish the boundaries and markers that both construct and constrain women's menstrual knowledge. Thirty-seven New Zealand women ranging in age from fourteen to eighty-six years contributed their narratives during open-ended interviews. The women's stories located various discursive practices that impacted on their menstrual knowledge and on their adherence to a common or popular menstrual etiquette. Discourses that construct and confine what, and how, women know about their menstrual cycle are identified and discussed. These scientific, medical, and consumerist discourses intersect and overlap to constitute a dominant menstrual discourse. Menstrual product advertising is identified as a prevailing context that surrounds young women as they become menstruants. Discursive practices such as euphemisms, notions of cleanliness and hygiene, authority through technology, and the commodification of feminist imagery contribute to representations that devalue and stigmatise menstruation. This dominant menstrual discourse can be maintained or disrupted through the way mothers impart menstrual knowledge to their daughters. Mothers are faced with the contradiction of preparing their daughters for an experience that is presented as normal yet constructed within strategies of concealment that menstruating women are expected to follow. When young women do become menstraunts, they are faced with the menstrual 'script' that includes the emotional themes of embarrassment, anxiety and ambivalence. The formal acquisition of menstrual knowledge takes place in our schools and again is positioned within a contradictory framework. Menstruation is conveyed as 'ordinary' yet the teaching of the menstrual cycle is often 'extraordinary' 'Menstruation' is routinely taught in sex-segregated classes, in the evening, in the company of parents and often located within scientific and medical discourses. This thesis offers new insight into the different ways New Zealand women construct knowledge about our bleeding bodies. Its uniqueness rests with die theoretical framework used to analyse research data. A feminist poststructuralist discourse analysis enabled the positioning of the women's accounts within a social, historical and cultural context, and the identification of a new way of analysing the impact of discursive practices upon meaning and experience of menstruation.

    View record details
  • Studies in marine natural products

    MacLean, Warren Jon (2005)

    Doctoral thesis
    University of Canterbury Library

    The marine environment continues to be a prolific source of structurally diverse and biologically active natural products. Over the past few decades both macroand micro-organisms have been extensively studied for such compounds. Annual reviews are dominated by novel compounds isolated from marine sponges, even though investigations have more recently been focused on micro-organisms. After decades of research into the development of marine natural products as potential pharmaceuticals, many marine compounds now feature prominently in current pre-clinical and clinical trials. This thesis represents a continuation of work in the area of isolation and structural elucidation of novel and/or biologically active natural products from both marine macro- and micro-organisms. The four novel C₃₁ polyacetylenic compounds 2.23, 2.24, 2.25 and 2.26 were isolated as the major components of the cytotoxic extract of the New Zealand marine sponge Rhabderemia stelletta, using bioassay-guided separation techniques. However, these structures were not unambiguously assigned due to the low resolution of the NMR signals associated with the long-chain alkyl protons and carbons. Elucidation was also hindered by compound degradation during structural analysis. The partial structures were identified via mass spectrometry and NMR spectroscopy and comparison of structural data for similar compounds reported in the literature. The novel pteridines 3.13 and 3.15 were isolated as the major and minor components, respectively, from the cytotoxic organic extract of an as yet unidentified Antarctic marine sponge 02WM01-33. Identification was achieved using mass spectrometry and NMR spectroscopy, and comparison of these data to those published in the literature for similar structures. Although pteridines are distributed widely throughout nature, 3.13 and 3.15 represent a chemically interesting group of compounds as they possess a mono-oxygenated methyl side chain, which has not been previously reported among any of the naturally occurring pteridines. The novel hydantoin compound 4.8 was isolated as the major biologically active component of the organic extract of the Antarctic marine sponge Suberites sp. using microtitre plate P388 assay guided fractionation. The relative stereochemistry of 4.8 was determined using lD-NOESY experimental data. This work represents the first isolation of the hydantoin 4.8 as a natural product from a marine sponge. This compound has been reported previously in the literature, but only as a synthetic product. The two novel anthraquinone compounds 5.35 and 5.41 were isolated as minor components of the cytotoxic organic extract of the Antarctic marine spongederived fungus Aspergillus sp., along with the three known anthraquinones 5.32, 5.34 and 5.40. This series of work revealed several areas in the literature that require revising: these include, (i) the compounds 5.31 and 5.41 were both reported as a new metabolites in 1985 and 2003, respectively, however, both compounds had been previously isolated and reported in 1966; (ii) the compound 5.36 requires the correct structural assignment in the literature; and finally, (iii) the compound 5.35 has two isomeric structures reported in the AntiMarin database, however, neither of these structures match the experimental ¹H NMR data supplied by the author of the database. The three novel compounds 6.9, 6.13 and 6.15 were isolated from the organic extract of an Antarctic marine sponge-derived fungus Ulocladuim sp., along with the previously reported compounds 6.6, 6.7 and 6.16. These compounds were identified using NMR spectroscopy and mass spectrometry. This work highlights an inconsistency in the literature regarding the structural assignment of two carbons for the known compound 6.16. The structural assignment of these carbons was confirmed by the independent isolation of 6.16 by another member of the Marine Group.

    View record details
  • Selecting and/or processing wood according to its processing characteristics

    Chauhan, Shakti Singh (2004)

    Doctoral thesis
    University of Canterbury Library

    In this study, three issues associated with the segregation of wood according to their processing characteristics have been addressed. In the first part of the study, the influence of inhomogeneity on the acoustic velocity measured by a resonance based acoustic tool "WoodSpec" and a transit-time based tool "Fakopp-2D" was investigated. Four laminated panels with varying degree of inhomogeneity were prepared and acoustic velocities were measured by both tools. The acoustic velocity measured by the Fakopp-2D tool was always higher than that measured by the WoodSpec tool. The difference in the two velocities was found to differ with the magnitude of inhomogeneity in the panel. It was demonstrated that the velocity measured by the resonance frequency obeys the Law of Mixtures and is controlled by the volume-weighted average stiffness of wood. In contrast the velocity measured by the Fakopp-2D tool was influenced by the amount of energy propagating at the dilatational speed. The progression of the wave-front in panels was investigated using Fakopp-2D tool by picking up the arrival time of the signal at off-axis probes. The wave-front profile was found to be determined by the stress-wave speed (stiffness) at different angles from the axis symmetry. In logs, the transit-time velocity was found to be sensitive to the localised stiffness of the wood lying in between the measurement probes with in the propagation path. A strong correlation was observed between the velocities measured by WoodSpec and the Fakopp-2D tool in young radiate pine and eucalyptus logs; however the acoustic velocity by Fakopp-2D was higher than the WoodSpec velocity. The results imply that Fakopp-2D could be used to rank young trees according to their stiffness. In the second part of the study, two trials were conducted to test the hypothesis that acoustics can be used to segregate pulp logs into categories which will require different amounts of energy during mechanical pulping and will produce pulps with different strengths. Pinus radiata logs of varying age, length, SED, LED, taper, and volume were measured for acoustic velocity, segregated into four different velocity groups and chipped separately. It was shown that acoustics could segregate logs into groups that perform very differently in terms of pulp and paper properties when refined to a given freeness or at a certain energy input. At a given target freeness there was a 20% difference in energy requirement between the lowest and highest velocitylogs. Similarly there was a 17% difference in tensile strength between the lowest and highest velocity logs for a given specific energy. In the third part of the study, investigations were made to explore the potential of acoustic velocity in ranking young Eucalyptus nitens trees according to growth stress level. Longitudinal growth strain was measured in 155 selected trees at the breast height and acoustic velocity was measured in the same trees using the Fakopp tool. Measurements in the first 34 trees showed some relationship between growth strain and acoustic velocity which eventually vanished as the measurements were progressively extended to all 155 trees. The results indicated the dangers of drawing possible inferences on the basis of small sample sizes. A large variation observed in growth strains along the tree height and on the two opposite sides in logs suggested that a single strain measurement is not sufficient enough to assess the mean tree strain level even in young and small diameter eucalypts trees and measurements on two opposite sides at a specific height is an approach for the screening purpose. The relationships between longitudinal growth strain and certain key wood properties were also investigated. Green density, green moisture content, basic density, radial shrinkage, outerwood and corewood densities, volumetric shrinkage and dynamic MoE at 12% me and length-weighted fibre length were determined. Amongst all the studied wood properties, only shrinkage-related properties were found to have some association with the mean growth strain in trees. The mean growth strain was moderately but significantly related to the volumetric shrinkage of the outerwood while the corewood shrinkage was not related. However, the volumetric shrinkage differential (difference between outerwood and corewood shrinkage) was strongly related to the growth strain (r=0.70) suggesting that the growth stress gradient might be related to the shrinkage property variations within the stem. Fisher's LSD test indicated statistically significant lower volumetric shrinkage, lower outerwood MoE and less collapse in the wood from trees with the lowest growth strains as compared to those from the highest growth strains. The results suggest that Eucalyptus nitens trees with low strains could exhibit a lower degree of the drying defects like collapse and internal checking during processing The large variation in wood quality characteristics in plantation grown timbers makes some screening for wood quality necessary for the effective management of wood resources and for the allocation of logs according to "Fitness for Purpose" to capture greater value of forest product value chain.

    View record details
  • An 'American dream' in the 'England of the Pacific' : American influences on New Zealand architecture, 1840-1940

    McEwan, Ann Elizabeth (2001)

    Doctoral thesis
    University of Canterbury Library

    This thesis investigates the American influence upon New Zealand architecture between 1840 and 1940. Taking a thematic approach, it argues that architecture in New Zealand, as a case study of a broader historical discourse, reveals the significant impact of American styles, construction methods, publications and educational practices in the century prior to World War Two. Chapter One canvases the general history of N.Z.-U.S. relations before 1940. Chapter Two reviews the same chronology to identify specific architecturally designed buildings that reveal an American influence upon New Zealand practitioners. Divided into two parts this extended discussion also explains why American architecture offered useful paradigms to designers in the South Pacific. Chapter Three demonstrates how publications were the principal means by which American architectural ideas were communicated to architects and builders in New Zealand. In a similar vein Chapter Four examines the ways in which architectural education brought practitioners in the United States and New Zealand together in common cause. Chapters Three and Four address the 'how' of American architectural influence in New Zealand. Chapter Five considers how New Zealand architects reacted to the presence of American influences within their local cultural domain. In addressing the question, why has greater attention not been paid to American architectural influences in New Zealand before now, it highlights a degree of resistance within the profession that was informed by both class sensitivities and imperial loyalties. In the Conclusion it is argued that the perception of World War Two as a watershed in New Zealand's cultural and social history, during which New Zealanders became aware of the United States for the first time, is ignorant of the longstanding relationship between the two nations. The construct of New Zealand's cultural isolation in the century or more before World War Two can be dismantled in the course of examining New Zealand's architectural history.

    View record details
  • The structure and eruptive history of Rotorua Caldera, Taupo Volcanic Zone, New Zealand

    Milner, David M. (2001)

    Doctoral thesis
    University of Canterbury Library

    Mamaku Ignimbrite erupted 220 -230 ka, is a voluminous deposit that is exposed over c. 3100 km² predominantly NNE, NW, and SW of Rotorua Caldera, and has a maximum measured outflow thickness of 145 m. Rotorua Caldera is a 21x22 km complex collapse structure in south west Bay of Plenty, Taupo Volcanic Zone, New Zealand. The Mamaku Ignimbrite sequence can be divided into basal tephra and main ignimbrite sequences. The basal tephra sequence comprises pyroclastic fall and density current deposits and records progression of the eruption from wet to dry. The main ignimbrite sequence has been separated into lower, middle and upper parts based on crystal content, welding, and extent of vapour phase alteration and devitrification. Boundaries between lower, middle, and upper ignimbrite are always gradational, and lithic and pumice content and size also vary stratigraphically throughout the deposit. Lateral variations in upper Mamaku Ignimbrite include areas of high lithic concentration, and lithic lag breccias, at and within the Rotorua Caldera margin with size and concentration of lithic fragments decreasing away from the caldera. At medial distances the ignimbrite is internally massive with occasional zones of pumice clast concentration, and varying lithic content. Boundaries of these lithic and pumice concentration zones are gradational and no sharp flow unit boundaries exist. At distal localities a crystal-rich band is present at the base of the ignimbrite and crystal-rich lenses are present at higher levels. These crystal-rich lenses suggest that distally the pyroclastic flow was moving in a dominantly non-particulate manner. XRF analysis has revealed three silicic pumice types in Mamaku Ignimbrite that range from dacite to high silica rhyolite (66-76 wt% SiO₂). No large compositional gaps exist between the pumice types and they provide evidence for the evacuation of a gradationally zoned magma chamber. The more silicic pumice compositions were probably derived by 20 % crystal fractionation of the dacitic magma. Andesitic blebs in upper Mamaku Ignimbrite are a fourth juvenile component which had a different petrogenesis to the silicic pumice types. The andesitic magma probably resided as a sill towards the base of the dacitic magma. Maximum lithic size variation shows Rotorua Caldera to be the source of upper Mamaku Ignimbrite. Gradational contacts between lower, middle, and upper ignimbrite suggest that the ignimbrite was deposited during a single eruptive event from one source. Coexistence of the three pumice types at all stratigraphic levels is further evidence for a single eruptive source. Variations in lithic content, and coexistence of different pumice types through the ignimbrite stratigraphy, indicate that caldera collapse occurred throughout the eruption, but particularly during the eruption of middle Mamaku Ignimbrite and in later stages of the eruption of upper Mamaku Ignimbrite. Rhyolite domes of Rotorua Caldera can be separated into seven groups, two of which are related to adjacent volcanic centres. Fragments of rhyolite lava are also a major component of lithic Jag breccias at caldera margin sites. The domes that are related to Rotorua record the eruption of five different rhyolitic magma bodies. The largest rhyolite dome complexes are located around the area of deepest caldera subsidence. These domes have similar phenocryst assemblages and phenocryst chemistries to silicic Mamaku Ignimbrite pumice clasts suggesting that they are from the same magma. Rhyolite lava fragments in the lag breccias also have similar phenocryst assemblages and phenocryst chemistry to Mamaku pumice clasts and may represent initial extrusions from the Mamaku magma chamber. All other rhyolite lavas in Rotorua Caldera have different phenocryst assemblages to Mamaku Ignimbrite pumice clasts and are probably not consanguineous with them. Mokai Ignimbrite outcrops in an 8 km wide (east to west) band between lakes Taupo and Whakamaru. It has the same unusual paleomagnetic direction as Mamaku Ignimbrite, is the same age, and has similar pumice chemistry. Upper Mokai Ignimbrite is vapour phase altered and has a similar appearance to upper Mamaku Ignimbrite. Three distinct units define Mokai Ignimbrite's stratigraphy, each separated by ash deposits. Internal variations suggest that Mokai Ignimbrite formed a compound cooling unit. The three flow unit compound cooling unit stratigraphy, phenocryst assemblage of juvenile mafic fragments, presence of mafic blebs at all stratigraphic levels, and thickness of Mokai Ignimbrite suggest that it is distinct from Mamaku Ignimbrite. Rotorua Caldera is described as a rhyolitic, single event, asymmetric, multiple block, single locus caldera on the basis of published geophysical data, caldera geomorphology and geology, location and thickness of Mamaku Ignimbrite and the nature of intracaldera rhyolite domes. The timing of collapse is deduced, from Mamaku Ignimbrite stratigraphy, to have started at least by the time that middle Mamaku Ignimbrite erupted, but probably occurred throughout the eruption of the main ignimbrite sequence. The caldera formed during a single eruption, but cannot be well described by current caldera classifications, as it has characteristics that can be attributed to trapdoor, downsag, piecemeal and piston processes of collapse. Taupo Volcanic Zone is dominated by NE-SW striking faults with more widely spaced NW-SE striking structures. This fault pattern inhibits the formation of simple caldera structures. The regional fault pattern is interpreted to have cut the Rotorua Caldera floor into a number of blocks that subsided to varying depths towards a single collapse locus.

    View record details