Infrastructure Engineering - Theses
Permanent URI for this collection
Search Results
1 - 10 of 65
-
ItemResponse measurement approaches towards damage identification using evolutionary algorithms( 2013)Despite the extensive research conducted in the past on damage identification using different damage parameters, objective functions and optimization techniques, the search for damage sensitive parameter and effective algorithms is still in progress. There have been many different studies employing traditional modal parameters and frequency response functions for detecting, localizing and estimating the severity of damage. Very few research studies have been devoted to using some other damage sensitive parameters for damage identification. However, previous studies that implemented modal parameters and frequency response functions for structural condition assessment, have not demonstrated consistent levels of sensitivity to structural damage. Moreover, damage identification problems usually require dealing with non-smooth, noisy stochastic problems and non-differentiable, multi-dimensional functions in which “local identifiable” optimization algorithms, are not capable of locating the global maxima or minima. In order to deal with complex high-dimensional, non-linear, non-differentiable and ill-condition problems, global optimization algorithms would be required. This research study presents new vibration-based non-destructive global structural damage identification and condition monitoring techniques that can be used for detection, localization and quantification of damage. Both two-stage as well as one-stage structural damage identification approaches are developed and implemented using power spectral density analysis based damage sensitive parameters, principal component analysis and a Bayesian probabilistic objective function. The power spectral density damage sensitive parameters can be obtained in the time-domain, spectral domain, modal domain and wavelet domain and are applied to detect, localize and estimate the severity of structural damage. In the first stage of the two-stage damage identification technique, damage is localized using damage sensitive parameters based on power spectral density analysis using a non model-based technique. The second stage of the two-stage method deals with the estimation of the damage severity using a model-based method and takes advantage of a Bayesian probabilistic objective function that take into account the main sources of uncertainties corresponding to measurement noise and modelling errors. In this stage, an extensive study is conducted to investigate the performance of the proposed damage identification technique using various conditions; single and multiple locations of damage, noise-free and even noisy incomplete response data. Numerical experimental data obtained from a one-dimensional structure, are used for extensive illustration of the proposed approach. In the proposed one-stage damage identification technique, damage is detected, localized and estimated in a one-step process using a model-based method. A parameterized finite element model along with a damage sensitive parameter and a Bayesian probabilistic objective function are implemented to evaluate an inverse problem in order to obtain the optimal model parameters. In this method, the performance of two stochastic evolutionary optimization algorithms; the genetic algorithm and the covariance matrix adaptation evolution strategy, is compared and the one with the highest rate of convergence and the most accurate solution for structural damage identification is proposed as the preferred algorithm. Moreover, a comparison study is conducted to assess the performance of five known objective functions in a one-stage damage identification process. The effectiveness of the objective functions is investigated for various damage condition scenarios and complex and noisy response situations. Furthermore, in this research, a one-stage wavelet analysis based structural damage identification technique is proposed and implemented in order to overcome the practical limitations of the Fourier transform in the condition of time-varying response data. Finally, the results obtained demonstrate the impressive performance of the proposed two-stage and one-stage damage identification techniques using Bayesian objective function in the presence of significant levels of noise and complex damage condition states. As well as the significant capabilities of the proposed methods, the covariance matrix adaptation evolution strategy is found to be computationally efficient compared to the genetic algorithm. Moreover, difficulties associated with inverse problems such as multi-dimensionality, ill-conditioning and non-linearity are overcome via employment of the two-stage and one-stage damage identification approaches.
-
ItemDevelopment of an inter-seasonal thermal storage system( 2013)The use of a ground-coupled heat pump (GCHP) has the potential to significantly reduce the amount of energy required for heating and cooling buildings. In order to reduce the impact of unbalanced heating and cooling load on the system performance and the ground temperature, an inter-seasonal thermal storage system integrated with unglazed solar collectors is proposed. This study examines the viability of using an inter-seasonal thermal storage integrated with a GCHP and unglazed solar collectors (ISTS-GCHP-SC) for the heating and cooling of residential buildings in Australia. Technicality and practicality of the proposed system has been assessed by experiment and simulations. The financial viability has been assessed by using life cycle cost (LCC) analysis. An experimental rig for ISTS-GCHP-SC system (heating capacity of 4.8 kWh and cooling capacity of 6.1 kWr) was set up at the University of Melbourne, Burnley campus. The system capacity, which depends on the peak building heating and cooling loads, was determined by simulating the building. The building model was verified by comparing simulated temperatures of the building zones with the measured temperatures. The borehole length of the ISTS depends on the effective ground thermal conductivity which is not directly measurable. Therefore, it was derived from the secondary measurements of heat transfer rate and temperature by performing an in-situ thermal response test. The set-up was tested with charging and discharging experiments conducted over a period of one and half years. The proposed system was found to be workable in Melbourne climatic conditions and there are no major issues with the installation and operation of the proposed system. The unglazed solar collector was able to charge with both heat and coolth. The average cooling rate of the unglazed solar collector was found to be 120 W m-2. However, a single borehole configuration of the ISTS, is not able to retain the heat and coolth until the next season. As the system was to be used for predicting the performance of the system in different climatic zones, the experimental results were used to validate the TRNSYS components. A TRNSYS model was developed to simulate the performance of the proposed system for different types of loads based on various climatic conditions. The load types investigated were cooling-only (Darwin), cooling-dominated (Alice Springs), heating-dominated (Melbourne) and heating-only (Hobart). The TRNSYS components (Types 257, 559, 919, 39 and 31) and the system simulation model were validated using the experimental data. Results show that the TRNSYS components are able to predict the performance of the corresponding system components. The system model was also found to be able to predict the performance of the system set-up. The financial viability of the proposed system was assessed through LCC analysis of the optimised system. The TRNSYS simulation model coupled with GenOpt was used to minimise the LCC for single and multi-building applications. The optimisation process was repeated for four cities representing four climatic zones using the same building as a load. Results showed that the optimisation process was able to find the optimal set of borehole length and solar collector area for each city except Darwin due to its high cloud cover of more than 46%. The LCC and the performance of the ISTS-GCHP-SC system were compared with the optimised conventional GCHP (without solar collectors) system. For a single residential building application, the results show that by using unglazed solar collectors to recharge heat and coolth, the borehole length can be reduced between 2.4% and 26.5% depending on the location. However, it was found that cost saved from the reduction of borehole length was offset by the operating cost of pumps and the solar collector cost. Therefore, except in the case of Hobart which has heating-only load, the ISTS-GCHP-SC system was found to be more expensive than the conventional GCHP system. This was also proven by the system coefficient of performance of the ISTS-GCHP-SC being lower than that of the conventional GCHP system except in the case of Hobart. The unit cost of heating a single residential building using the ISTS-GCHP-SC system was found to be higher than the air-source heat pump (ASHP) system and liquefied petroleum gas (LPG) heating system in Alice Springs and an ASHP and ducted natural gas heating system in Melbourne. In Hobart, it was found that heating with the ISTS-GCHP-SC system is cheaper than the ASHP and ducted natural gas heating system. However, the unit cost of cooling using the ISTS-GCHP-SC system is higher than an ASHP in both Alice Springs and Melbourne. The analysis was repeated for a multi-building application. The ISTS-GCHP-SC system was optimised for a multi-building application comprising of 30 residential buildings. Three water-to-water heat pumps were used with a separate heating and cooling coil units for each building. The system was found to be able to supply the heating and cooling load to 30 buildings with a constant system coefficient of performance (COP) over a period of 20 years. The LCC per building is 40-50% lower than for a system used for a single building application. The unit cost of heating multi-building using the ISTS-GCHP-SC system is lower than the ASHP and LPG heating system in Alice Springs and ASHP and ducted natural gas heating system in Melbourne and Hobart. However, the unit cost of cooling is higher than that of an ASHP. As the cost of the ISTS-GCHP-SC system was based on a one-off cost in Melbourne, a sensitivity analysis was performed. A sensitivity analysis was conducted to investigate the impact of varying the cost of the components and the grout thermal conductivity on the LCC and the unit cost of heating and cooling for a single building application. The results showed that the LCC is most sensitive to the heat pump and borehole drilling cost and is least sensitive to the solar collector area except in the case of Alice Springs. The LCC was also found to be highly sensitive to the grout thermal conductivity. For a multi-building application, the LCC was found to be more sensitive to the borehole length and the discount rate. Therefore, in order to reduce the cost of ISTS-GCHP-SC system, the focus should be on the minimisation of cost of drilling vertical boreholes and heat pumps. The greenhouse gas (GHG) emission savings depends on the type of fuel used for heating and climatic zones where it is used. The results were found to be similar for both single and multi-building heating applications. In Alice Springs, it was found that using ISTS-GCHP-SC system could result in lower GHG emissions compared to the ASHP and LPG heating systems. However, in Melbourne a ducted natural gas heating system results in the lowest GHG emissions and in Hobart, it is the ISTS-GCHP-SC system that results in the lowest GHG emissions. For a single building cooling application, the ISTS-GCHP-SC system results in lower GHG emissions in both Alice Springs and Melbourne but for a multi-building, it is the ASHP that results in lower GHG emissions.
-
ItemLandfill phytocap development and performance evaluation using Australian native plants( 2010)The aim of the research presented in this thesis is to improve Australian landfill phytocap performance outcomes by enhancing our understanding of factors influencing the function of the phytocap plant community. The landfill phytocap presents a soil-plant alternative to the traditional barrier cap paradigm and relies on the capacity of a porous layer of soil to store water, and the combination of evaporation and the natural transpiration abilities of vegetation to control the percolation of water into a landfill. Despite the vegetation component having at least an equal role as the soil in determining phytocap performance outcomes, to date there has been a far greater emphasis in phytocap research on the configuration of the soil profile relative to the plant community. Furthermore, the majority of full-scale field studies on phytocap performance have been conducted in semi-arid or arid sites in the northern hemisphere with different climatic patterns and plant species. This research evaluates the hydrological behaviour and performance of a phytocap system within an Australian context. Greater emphasis is placed on the phytocap plant community and the potential effects that landfill processes have on performance outcomes including the impact of a lysimeter. Initial chapters focus on the selection and trial of native plant species in the field and glasshouse as part of phytocap development while later chapters evaluate the performance of the developed phytocap using full-scale test sections. The first stage of phytocap development was substrate characterisation and site assessment to inform native plant selection. A transferable methodology was demonstrated that can be applied to other landfill sites to improve plant selection outcomes, and was strongly supported by the successful establishment of a native plant community. Responding to the need for greater consideration of plant performance during phytocap design, a glasshouse trial was conducted to assess the species specific effects of substrate compaction on plant water use and root growth. This trial enabled the identification of an optimum phytocap substrate compaction range (75-83% of standard maximum dry bulk density) which optimises plant water use and effective rooting depth. This specification can be broadly prescribed to future phytocap designs. The hydrological performance of a monolithic and capillary barrier design were evaluated with lysimeters and found to perform exceedingly well. As both designs recorded very low percolation and were capable of effectively limiting percolation within the second year of monitoring, no distinction was made between designs in terms of performance. Evapotranspiration was found to be the most significant element of the water balance and the system was strongly limited by the availability of water such that evapotranspiration was high throughout the year, including the winter months. Limitations of lysimeters were assessed through comparative monitoring of unlined test sections installed directly on top of the landfill. Lysimeters were found to exclude adverse landfill gas and temperature effects, prevent assessment of plant sensitivity to the landfill environment and obstruct the interaction between the substrate design and the landfill including the landfill gas extraction system. Another significant limitation of lysimeters identified, was their inability to differentiate between phytocap designs; while there was no distinction between the performance of the monolithic and capillary barrier designs when monitored with lysimeters, when the designs were monitored on unlined test sections the performance of the capillary barrier design was clearly hydrologically and ecologically inferior. These limitations have significant implications for the interpretation of data obtained from studies based solely on lysimeter monitoring. They also suggest that comparative modelling of designs based on lysimeter data may not be meaningful for the employment of the designs on a full-scale phytocap. Overall, this thesis highlights the complexity of the interactions present within a full-scale phytocap and suggests that the inability of lysimeters to incorporate interactions which. strongly impact upon phytocap performance makes them an inadequate tool for performance evaluation of these systems. For that, alternative performance evaluation tools to lysimeters need to be considered.
-
ItemThe investigation of land cover change and trends through the reconstruction of historical landscapes( 2012)The thesis presented on a new and novel method of reconstruction of historical landscapes for investigation of land cover change and trends. It also discusses and accentuate the importance of land cover change thus the reasons it should be known. Remote sensing and geographical information systems (GIS) method is the only available method for the land cover change investigations at present. This modern techniques method of remote sensing and GIS is reviewed and its inadequacies established. Its major problem is that it has only been available since 1972 therefore for periods that spans more than 40 years there is no satellite data. Then, since land cover change is a slow process, it is necessary to access evidence about the land cover conditions of past eras, data from 40, 50, 60 years ago and beyond. Therefore, it is imperative that a method is found that can be used to assess land cover change in the historical times. The major question is whether reconstruction of historical landscape can be used successfully for the land cover change investigation to fill up the gap where remote sensing is unavailable or inadquent. The proposal of this new method to counteract the inadequacies of remote sensing and GIS method is set forward. Its blue print (Table 3-2 and Figure 3-1) is submitted and tested against the well used, tested and trusted land cover change investigation method of remote sensing and GIS. The new method uses the elements of traditional ecological knowledge of oral history, songs, poems, praise singing and relics; archival data of maps, photographs and aerial photographs; written records and archeological research results. Since this approach does not rely on technology in its original data capture phase, it can be applied anywhere and long time after land cover changes has taken place. For the development and evaluation of the alternative method, test areas in Botswana were used because in these areas, it was possible to draw on the strong African tradition of oral history. The study was made between three time periods of – 1970-1979 – Sir Seretse Khama‟ Era, 1980-1990 – Sir Ketumile Masire‟ Era, and 1991-2003 – Festus Mogae‟s Era. Besides that, on the local setting of Botswana, three minor questions were asked: one was on the political landscape - how did Botswana come into being? And what were the political historical landscapes governing the southern African region? And whether these had any effect on shaping land use patterns and development as seen today? The other question was on population - What are the variability of population changes of Botswana within this period of study and their impact on its land use and therefore land cover changes? And the third question was on climate - What is the extent and variability of climatic changes in Botswana within this period of study and what assessment can be made by correlation of temperature/precipitation data with the years of drought and those of floods? The creation of land cover change maps by the new method followed some procedures. First, the informal data from the traditional ecological knowledge voice data recorded from interviews; the archival maps and photographs graphical data; the written records and the archaeological research findings text and tabular data were deciphered. Second, the epicentres of land cover change point of reference data of the areas were deduced from these data, which were then extracted, and tabulated. Third, the field visits to these points of reference were made and coordinates of their exact locations were measured using a global positioning system (GPS). Fourth, these single items, large scale field data was categorised into land cover classes similar to those seen from smaller scale satellite images of the same area. After these steps it was only then that these informal data were used to create databases which were then transformed and translated into land cover maps for different periods. The land cover maps of consecutive periods were then compared and from that change were detected land cover change map was formed. Other important findings were that for the two methods, that of remote sensing and that of the reconstruction of historical landscape sees the world differently. First it is about remote sensing and GIS intrinsic characteristics. Remote sensing sees the land cover surface from the above, a bird‟s eye view. Although it captures everything and is unbiased some details can be obscured by those above them and are missed out. It is also sees the Earth‟s phenomena from great distances above. Geostationary Operational Environmental Satellites (GOES) 8 orbit the Earth at 35 790 kilometres altitude, the latest of Landsat Earth resources group of satellites system, Landsat 7‟s orbiting altitude is 705 kilometres and the very high resolution IKONOS orbiting altitude is 681 kilometres. Consequently, the satellite data is always viewed as land cover class categories at small scale like built-up-area town/village, or field/forest/vegetation or river/lake/swamp not as single land cover items like a building or a tree or water. The view can change in incidence that depends on factors acting on the land surface; land change by the effects of seasons. The areas which can be viewed at one time are enormous. Good in showing the land cover change and what the land class has really changed to become and this is recorded and archived. Remote sensing views the world far beyond human eye; they utilize the visible electromagnetic spectrum as well as the infra red – they can even see heat; the microwave – RADAR satellites. But need trained people to interpret and analyze. Second, the natural way human beings see the world, the reconstruction of historical landscape method. The world is viewed piecemeal – large scale single item and in oblique, sideways fashion. Also what is viewed is the understory only. The items remain the same to the human mind, it does not change by season or by the way it looks. The world is categorized artificially into names, land uses, ownerships and governed by rules which sometimes has nothing to do with what is on the Earth‟s surface, the land cover its self. People also have the ability to know why something happed and also the ability to draw from their experience, rightly or wrongly. Therefore, as a consequence to that, if these two methods have to be used interchangeably a land cover class specification commensurate to the two methods has to be created. Third, besides all that, the remote sensing platforms and sensors have improved tremendously over the years. Consequently, land cover change from consecutive time period studied from data of satellite from different generation could be more pronounced not only on account of the changed landscape but also on account that better equipment are being employed. So, if a trans-generation satellite data has to be employed some form of standardization of the different generation data has to be made. Various elements of reconstruction of historical landscape subscribe to the land cover change investigation endevours at different levels which complement each other. The traditional ecological knowledge is the most effective giving information which could nearly complete the land cover change. Archaeology although it has brought life to the historical landscape is sporadic and falls apart by age and by space and it also depends on preservations. Others like the songs and the written records depend on whether they are available and also, the information they include is at the author‟s discretion. This thesis has sufficiently proved that the method of reconstruction of historical landscape for the investigation of land cover change can be applied to fill in the gap where modern technologyis inadequate or unavailable. And that indeed climatic change, population density and historical, political and economical landscape are factors that affect land cover change. This is, without any doubt, a welcome accomplishment for the field of engineering and will greatly facilitate further national and regional as well as global scientific undertakings of this nature.
-
ItemRunoff generation mechanisms, hydrologic connectivity and nutrient export from a steep dairy farming catchment, Victoria, Australia( 2013)Effective catchment management aims to improve water quality, primarily through reducing excess nutrient (nitrogen and phosphorus) concentrations in streamwater. This thesis aims to investigate the role of hydrology in controlling streamwater chemistry including major ions (Na+, K+, Ca2+, Mg2+ and C1-), isotopes (18O and D) and N and P species in streamwater. In particular, the thesis develops a conceptual model that reconciles the hydrologic and chemical behaviour that relates nutrient export to a quantitative understanding of flow pathways in a small headwater dairy farming catchment in Victoria, Australia. In order to achieve the aims of this research, a field study was undertaken in a small hillslope of 1.37 ha. Hydrometric data collected included rainfall, flow, shallow groundwater levels and climate variables. Water samples were collected from the stream (event flow and low flow), shallow groundwater and soil water samples and were analysed for hydrochemical tracers, isotopes and nutrients (NH4+, NO3-, TN, FRP and TP). Stream flows and water chemistry were also monitored for a 125 ha neighbouring catchment. Results are presented in three parts: a hydrometric analysis, a hydrochemical and flow tracing analysis, and an analysis of the nutrient dynamics. The hydrometric results show a mixture of runoff mechanisms, including Hortonian overland flow, saturation overland flow, and subsurface storm flow, and relates these to variations in hydrologic connection and rainfall intensity. These results showed that initiation of flow typically depends on a specific threshold of the combination of antecedent soil moisture and total rainfall. The flow is highly seasonal due to variations in soil moisture, which controls hydrologic connection between the hillslope and riparian zone via subsurface flow pathways. Hillslope flow was intermittent and usually ceased between events; however, during the wettest conditions, the hillslope was continuously connected to the riparian zone and hillslope flow was persistent. Event runoff was generally a combination of subsurface stormflow and saturation excess overland flow; however, for events which had high maximum hourly rainfall intensities (> 15 mm/hr) runoff was produced from a relatively dry catchment, probably by a combination of Hortonian overland flow and/or preferential flow pathways initiated by the high intensity rainfall. Then tracer analysis was used to test the conceptual model developed from the hydrometric data. Primary analyses of flow-tracer concentration relationships for all tracers showed chemostatic behaviour. Tracer analysis was able to separate flow contributions between the saturated zone and unsaturated zone. This revealed that saturated zone water was typically the main runoff source during events, although unsaturated zone water often made a significant contribution during the rising limb and peak. A new heuristic model for hydrograph separation was developed because end-member mixing analysis (EMMA) failed due to very large geographic variability in end members. The results of the hydrochemical analysis suggest that the catchment mixes a wide spectrum of stores with concentrations that change spatially and whose relative contributions vary temporally. Isotopic analysis showed that much of the event flow is old water but that for some events, new water is important during the rising limb and peak of the hydrograph. The results of nutrient analysis revealed that most N and P species, except NO3-, showed no seasonality. The concentrations of NO3- changed seasonally. During events, concentrations of NH4+, FRP, and TP were proportional to the fraction of unsaturated zone water contribution, suggesting they are transported with this flow pathway. For NO3-, a strong hysteresis with the proportion of unsaturated zone water was found. More detailed analysis revealed that variations in flow pathway through the riparian zone coupled with spatial variations in NO3- concentrations in the riparian zone was the main controller of NO3- concentrations in the subsurface flow and high NO3- concentrations bypassed most of the riparian zone when the hillslope was highly connected to the riparian zone. In the discussion and conclusion chapter, a conceptual model of the catchment was discussed from the perspective of new concepts in hydrology including the double paradox and the catchment chemostatic behaviour. A conceptual model of the hydrometric, hydrochemical, isotopic and nutrient dynamics is discussed, followed by an explanation of the possible reasons for the observed differences between the hillslope and the headwater catchment response. The key implications and contributions of the study are discussed from various perspectives including: the double paradox, chemostatic behaviour of the catchment, and implications for management and modelling.
-
ItemUsing land administration for land risk management( 2013)The impact of risk and disaster events on land and property within the developed world in the last decades has highlighted a significant problem in the ability of citizens and governments to address and respond to these threats. A breakdown in the process of identifying, analysing, evaluating and treating these risks has occurred, leaving communities exposed and vulnerable to a range of very real risk possibilities. The integration of land administration information and risk management processes is considered essential for achieving effective land risk management practices and community resilience for risk events. However, in most countries, land administration and risk management are usually disparate disciplines. This research addresses this problem with the overall aim of facilitating improved risk management of land and property for all stakeholders. This research investigates how land administration could support the process of managing risk to land and property for a range of stakeholders. Its primary objective is to develop a land risk management model which illustrates how these two elements, land administration and risk management, could be integrated to enable the implementation of effective land risk management practices by all stakeholders and to facilitate the development of a resilient community. A mixed methods research design was utilised which included the use of a case study approach focusing on developed countries with established land administration systems. The research developed: an understanding of the issues which impact upon the ability of land administration agencies to contribute to land risk management as well as the factors which motivate them to participate; an understanding of the stakeholder roles and responsibilities in the process of land risk management; and finally, a land risk management model which illustrates a ‘to be’ situation for how land administration could support land risk management if the issues and factors identified were addressed. The model is realised as a prototype system which demonstrates how land administration information can facilitate the effective implementation of land risk management processes and strategies. This research goes beyond the disaster risk reduction and disaster risk management strategies which have emerged from the integration of traditional disaster management models with the process of risk management. In these new models, only specific elements of the risk management process are incorporated and the focus remains largely on the response and recovery elements. This research focuses more heavily on the entire risk management process and all of the elements within the model and is applied specifically to the problem of risk affecting land and property and how this risk can be managed. Applying more attention to the risk management process enables the development of a more resilient community through thorough identification, acknowledgement, assessment and treatment of risks affecting land and property. The integration of land administration facilitates the process enabling stakeholders to better understand the risks which affect their land and property through a user centred view. The study concludes that the current land risk management processes are not sufficient and that improvements are required to achieve community resilience to risk events. The findings reveal that land administration systems have the potential to support land risk management practices and have significant motivational factors however changes to policy, legal, institutional and technical arrangements are first required. It is expected that land risk management initiatives will continue to be high profile issues as climate change brings more frequent and severe weather events. The success of future community resilience will therefore rely heavily on improved management processes for managing risk to land and property through the utilisation of land administration information and engagement of all stakeholders.
-
ItemImpact resistance of polyurea coated aluminium plates subjected to low and high velocity impacts( 2013)With the increasing possibility of structural damage due to natural disasters, collision of vehicles, and blast and ballistic impacts, the demand for protective measures for structures is on the rise. Localised impacts are among one of the most common loading mechanisms that many modern structures undergo during their life span. Over the years, many investigations have been conducted in order to mitigate the structural damages caused by impact actions from different objects such as flying projectiles, debris, hail and collision of vehicles. Due to their unpredictability and complex nature, impact loads are considered one of the most intricate loading mechanisms to mitigate in the field of structural engineering. Engineers have been depending on high strength and high hardness materials to achieve the required resistance against such severe loadings. Better performance was achieved by either increasing the strength of the materials, or by increasing the dimensions of the structural elements. However, in recent years, attention has been tilted towards using lightweight materials in structural components in areas such as the armour industry. Due to their low density, there is comparatively high demand for lightweight materials in many industrial applications. Metal–elastomer composites can be considered as one of the main alternatives in such applications. Over the last decade, the use of elastomers as composite materials to mitigate blast loads has been extensively investigated. However, less attention has been paid to study the use of composites subjected to projectiles impacts. Therefore, in this study, aluminium–polyurea composite (metal–elastomer) has been selected as an alternative structural material to improve the performance of structures under impact loading. This thesis presents experimental, analytical and numerical investigations of aluminium–polyurea composite plates subjected to low and high velocity projectile impacts. The research has been divided into three main parts: (a) an experimental study of material behaviour under different strain rate loadings, (b) experimental, analytical and numerical investigation of aluminium–polyurea composite plates subjected to low velocity impacts, and (c) a comprehensive study of the behaviour of aluminium–polyurea composite plate systems subjected to high velocity projectile impacts. Each section consists of two peer reviewed journal papers that have either been published or are under review in international engineering journals. Over the years, aluminium alloys have been considered as a possible partial substitute for structural steel. Aluminium alloys AA5083-H116 and AA6061-T656 have been investigated for their behaviour under different strain rates, ranging from 10-3/s to 104/s. Aluminium alloy AA5083-H116 was selected as the base material in the formation of the composite in this study. With the advancement of alloying techniques, the manufacturers have been able to produce aluminium alloys with ultimate tensile strengths in the range of 450–750 MPa. Especially in the personal armour industry, aluminium AA5083-H116 is a commonly used alloy due to its considerably high ductility and moderate strength. Strain rate sensitivity and energy absorption capacities were obtained from the experimental data and are discussed in this thesis. In the present study, polyurea has been selected to form a composite with the aluminium alloy. Due to its ability to bond well with materials such as concrete and steel, and its advancement in manufacturing techniques over the last few years, polyurea is considered as one of the prospective materials to be used in composites, especially in the application of impact mitigation. High strain rate behaviour of polyurea under tensile loading has been studied, and a constitutive model to predict high strain rate behaviour of polyurea under different loading conditions has been proposed. This simplified model is capable of predicting strain rate sensitivity of polyurea reasonably well. Low velocity impact test programs on both uncoated and coated aluminium plates have been conducted to investigate the ability of polyurea to resist permanent damage caused by such impacts. An analytical model to predict plastic deformation of aluminium plates subjected to low velocity impacts was also proposed. The same experimental program was extended to investigate the effects of polyurea coating in terms of energy absorption and reduction in permanent structural damage. A comprehensive numerical study has been performed to understand the behaviour of composite plates under the impact of projectiles travelling at low velocities using the advanced finite element code LS-DYNA. One of the main objectives of this study was to investigate the applicability of layered aluminium–polyurea composite plate systems to mitigate ballistic impacts. An experimental study was conducted in order to evaluate the performance of aluminium–polyurea composite plate systems subjected to penetration by NATO standard ammunitions (high velocity impact). A broad study was performed using finite element code LS-DYNA on numerical simulation of the experimental program. The models were validated with the experimental results. An analytical model to predict the residual velocities of the composite plate systems has been proposed and validated with both experimental and numerical results. The results of this research, which have been summarised in six journal papers, have been used to show the applicability of polyurea–aluminium composites in mitigation of low and high velocity projectile impacts. Strain rate sensitivity of the mechanical properties of aluminium and polyurea has been highlighted. A strain rate dependent constitutive material model was proposed based on the well-known Mooney-Rivlin model. Applicability of polyurea or a similar type of elastomer in reducing the residual velocity and acting as a protective shield in composite plate arrangements was highlighted. This finding can be used effectively in armour and many other industries in manufacturing light weight structures. An analytical model to predict the residual velocity of projectiles penetrating through composite plate systems has been proposed. This model can be effectively used to predict the residual velocities of multi-layer composite systems. In addition applicability of polyurea-aluminium composite in mitigation of low velocity impact has been highlighted. Polyurea has shown good ability in energy absorption, and subsequently has reduced the permanent deformation of the composite structure. Therefore, the outcomes of this research study can be used in the future in the protective structures design industry.
-
ItemGround heat exchanger design for direct geothermal energy systems( 2013)Direct geothermal energy systems use the ground to heat and cool buildings. Ground-source heat pump (GSHP) systems are the most widespread form of direct geothermal energy system (although currently rare in Australia). Geothermal energy is a sustainable energy source, and because the temperature of the ground is more stable than that of the air (warmer in winter and cooler in summer), GSHP systems are typically more efficient than conventional heating and cooling systems. However, GSHP systems are typically more expensive to build than conventional systems, largely because of the need to construct ground heat exchangers (GHEs) to extract heat from or reject heat to the ground. In GSHP systems, a primary (ground) circuit is linked to a secondary (building) circuit via a heat pump. The GHE is an integral part of the primary circuit and typically comprises a closed-loop of absorber pipes embedded in the ground. Closed-loop vertical GHEs (boreholes and energy piles) are the focus of this research. Heat is transferred from the ground to the heat pump (or vice versa) using a fluid which circulates in the absorber pipes. The manner in which the GHE is constructed has a significant influence on the efficiency of heat transfer between the fluid and the surrounding ground, and hence GSHP system efficiency and cost. Despite the importance of the GHE to GSHP system performance, there are no Australian (and few international) field experiments that study GHE heat transfer in the ground in detail. Consequently, existing GHE design is conservative. The ultimate goal of this research is to increase the performance and reduce the cost of GSHP systems, by improving the understanding of GHE heat transfer. To pursue this goal, field experiments have been performed on several 30m deep energy pile and borehole GHEs, in which factors such as absorber pipe diameter and spacing, grout thermal conductivity, and heat transfer fluid flow rate have been varied. Experiments have been performed on both water GHEs (water circulating in HDPE absorber pipe) and refrigerant GHEs (refrigerant circulating in copper pipe). Ground and GHE temperatures have been measured at various depths within and around the GHE to observe the effect of thermal loading on the surrounding ground. Power has been applied to the GHEs using either an electrical heating element or a heat pump capable of extracting and rejecting heat to the ground. Simple but robust new numerical models have been developed on the basis of the experimental results. These models have been validated against the experimental results, and used as a guide for the likely influence of GHE depth on GSHP system performance in the long term. The results of this research suggest that GSHP systems are suitable for local conditions, and that there is the potential to significantly improve GHE efficiency, especially with regards to the influence of GHE depth on the ground’s ability to replenish heat extracted from or rejected to it in the long-term, and on the combined influence of fluid flow rate and absorber pipe length. This thesis also presents an improved method for accurately assessing the thermal properties of the ground (an improved thermal response test analysis method), and by describing the experimental works and their analysis in detail, provides a framework for future experimental studies of direct geothermal energy systems.
-
ItemStochastic modelling for real-time GNSS positioning( 2013)Satellite positioning refers to the process of obtaining positions on or near the Earth’s surface by measuring ranges to a number of Earth-orbiting satellites. As the positions of the satellites are known at any given time, the observations can be combined in a set of simultaneous equations to determine the coordinates of the receiver, along with some measure of coordinate quality. The most prevalent technique for computing parameters from a set of observations is least squares estimation. Least squares requires a functional model that describes the mathematical relationship between the observations and the unknown parameters and a stochastic model which describes the statistical behaviour of the observations. The estimation process yields both the parameters and their precision estimates. Functional models for GNSS positioning are well known and have remained essentially unchanged for the last two decades. On the other hand, stochastic models are less well understood. Providing a realistic stochastic model in support of GNSS processing is a major challenge. A clear solution is yet to be found and as a result rudimentary models continue to be used in practice. The stochastic model is represented by the variance-covariance matrix in the least squares algorithm. The diagonal terms of the matrix are variances which describe the precision of individual observations. The off-diagonal terms are the covariances which arise from physical correlations between the observations. Three types of physical correlation have been identified: spatial, temporal, and observation-type. It has been shown by a number of studies that ignoring these correlations will produce unreliable precision estimates of the unknown parameters. However estimating physical correlations, especially for in real-time, has proved an elusive goal. Approaches to model these correlations developed to date are suited only for post-processing applications. This study proposes a new stochastic model for real-time GNSS processing based on a quantity known as Time Differenced Range Residual (TDRR) that enables empirical noise estimation from raw observations in real-time. The advantages of this approach are that it is based on raw observations, it is computationally efficient and it allows modelling of the main physical correlations. The TDRR is investigated in this study for its usefulness in empirical noise estimation. It is shown that the TDRR can be used as a tool to investigate the noise characteristics of various GNSS receivers. The stochastic model based on the TDRR is developed in full. This model includes variances for the individual satellite observations as well the spatial correlations. The stochastic model is compared to conventional approaches for processing three short baselines of 3, 9 and 12 km long. Short baselines are chosen for the analysis to minimise the effect of the atmospheric biases on the solution. It is shown that the TDRR-based stochastic model provides more realistic precision estimates for the parameters compared to the conventional approaches, however some limitations in the development are also identified, which require further refinement before the newly developed model can be applied in practical application.
-
ItemPhytocaps as biotic systems to mitigate landfill methane emissions( 2013)Landfill gas is a significant source of anthropogenic methane emissions and accounts for more than half of greenhouse gas emissions from waste sectors. While harvesting landfill gas for energy is the best mitigation option, methane oxidation by landfill cover soils is considered an important secondary measure to reduce landfill methane emissions. In recent years, regulatory control has evolved to allow consideration of alternative options for final covers. An evapotranspiration cover, also commonly known as phytocap in Australia, is one of the alternative cover options that has been widely considered and investigated. A phytocap presents a soil-plant alternative to the traditional barrier cap approach. It relies on the capacity of a porous layer of soil to store water, and the combination of evaporation and transpiration of vegetation to control the percolation of water into a landfill. When planted with native vegetation, it also improves the ecology and sustainability of a closed landfill. While the hydrological performance of phytocap has been investigated by a number of studies resulting in positive outcomes, its ability to serve also as a “biocover” for effective methane oxidation to mitigate emissions has received little attention. The main aim of this thesis was to assess phytocap performance in terms of enhancing methane oxidation activity in the cover soil and mitigating methane emissions. The research methodology included a full-scale field comparison between phytocaps and conventional compacted clay covers in terms of methane oxidation and emissions. A supplementary glasshouse experiment with both blank and planted soil columns was also conducted to investigate vegetation-methane interactions, and to identify plant influenced soil properties that would affect methane oxidation and emissions. This research forms a part of the 5-year Australian Alternative Cover Assessment Project (A-ACAP), co-funded by the Australian Research Council and Waste Management Association of Australia. In the full-scale field comparison, trial sites located at five landfills under a broad range of Australian climatic conditions have been monitored. The 5 A-ACAP trial sites with side-by-side phytocap and conventional cover test pads were built directly on top of active landfills with an aim to study their hydrological performance as well as methane mitigation efficiency. This thesis related to the methane mitigation component focused on the trial site located in Melbourne where more frequent monitoring campaigns have been conducted. The results of the field trial indicated that phytocaps could mitigate methane emissions more effectively compared to conventional covers. Emission rates detected from the Melbourne phytocap averaged at 1.45 gCH4/m2/day (out of the 17% measurements that resulted in significant positive fluxes), compared to the conventional cover which averaged at 5.57 gCH4/m2/day (out of the 65% measurements that resulted in significant positive fluxes). This positive finding is supported by the gas concentration profile data obtained from both types of covers. The field trial also concluded that the effectiveness of methane oxidation in phytocaps can be significantly enhanced with methane emission reduced to a negligible level when used in combination with gas extraction systems. In contrast, only a marginal gas extraction influence was observed on conventional covers. In addition to the overall reduction in emissions, phytocaps can also significantly reduce the amount of hot spots in surface emissions. For the glasshouse experiment, at both high and low gas influx rates, the planted soil columns showed high oxidation fractions (mostly higher than 0.5), which are comparable to the performance of some biocovers reported in the literature. Rather unexpectedly, the blank soil columns exhibited an even higher average CH4 oxidation fraction (average 0.89 under 36.5-73 gCH4/m2/day load) compared to the planted soil columns (average 0.67 under 36.5-73 gCH4/m2/day load). This finding appeared to be contradictory to the positive methane oxidation enhancement effects of vegetation in soil covers commonly reported in previous studies. With a closer examination, it was observed that the plant roots brought in a significant increase in soil gas diffusivity of the planted columns, which significantly shortened the methane retention time in the soil and subsequently reduced the methane oxidation capacity of the planted columns. The high oxidation fraction of the blank columns was attributed to the organic rich soil. Combining the research of this thesis with the findings of a concurrent A-ACAP hydrological study, it can be concluded that phytocaps provide an economical and sustainable option for new and old landfills to minimise water percolation and to mitigate methane emissions. As a result of achieving the objective of minimising percolation, the soil moisture profile of a phytocap may not be at its optimum for methane oxidation during certain periods of the year. Maintaining a balance between minimising water infiltration and promoting methane oxidation has to be addressed in a phytocap design in order to achieve optimum performance in both functionalities.