This thesis reports findings from a number of modern machine learning techniques applied to electricity market price forecasting. The techniques evaluated were Support Vector Machines, Boosting, Bayesian networks, neural networks and a weekly average method. All techniques were evaluated on seven day into the future forecasting of the Regional Reference (pool) Prices (RRP) for the New South Wales (NSW) region of the Australian National Electricity Market (NEM). Due to highly volatile and non-repetitive nature of the NSW RRP, all complex machine learning methods provided inferior accuracy forecasts compared to a weekly average method. The weekly average method was computationally less expensive and more transparent to the user than any of the machine learning techniques. The Support Vector Machine (SVM) was chosen for its novel application to electricity price forecasting because it is considered to be the next generation to neural networks. The structured SVM training algorithm proved more consistent and reliable than the neural network algorithm. Bayesian networks offer the adaptability of a neural network with the advantage of providing a price forecast with confidence intervals for each half-hour determined from the actual data. The SVM and Bayesian techniques were found to provide acceptable forecasts for NSW demand. An investigation of international electricity markets found that each market was unique with different market structures, regulations, network topologies and ownership regimes. Price forecasting techniques and results cannot be universally applied without careful consideration of local conditions. For instance, price data for the Spanish and Californian electricity markets were investigated and found to have significantly lower price volatility than the NSW region of the NEM. An extensive examination of the NSW RRP showed that the price exhibited no consistent long-term trend. A stationary data set could not be extracted from the price data. Thus, making forecasting unsuited to techniques using large historical data sets. The strongest pattern found for NSW prices was the weekly cycle, so a weekly average method was developed to utilise this weekly cycle. Over 25 weeks of NSW RRP from February to July 2002, the seven day into the future price forecast mean absolute error (MAE) for the SVM technique was 27.8%. The weekly average method was more accurate with an MAE of 20.6% and with a simple linear price adjustment for demand, the error was reduced to 18.1%. The price spikes and uneven distribution of prices were unsuitable for the Boosting or Bayesian network techniques.

Land use refers to the use of land for intensive human activities aiming at exclusive use of land for certain purposes and adapting the properties of land areas in view of these purposes. Environmental problems are, however, generated as a result of these human activities which modify the shape and properties of large land areas according to the requirements of human activities and thereby excluding wild animals and plants from coexisting on such land areas and in their neighbourhoods. Land use also leads to the degradation of the natural environment. Life Cycle Assessment (LCA) methodology is used for evaluating the environmental burdens associated with products or processes while taking their whole life cycle into consideration. LCA is a comprehensive assessment method which considers all aspects of natural environment, human health, and resources. Land use is regarded as an impact category in Life Cycle Assessment and is treated as such. However, the environmental impacts associated with land use and land use change are not being adequately considered in LCA, if considered at all.Life Cycle Impact Assessment is a part of LCA and is aimed at understanding and evaluating the magnitude and significance of the potential environmental impacts of products or processes and this involves developing characterization factors which link an environmental impact to a category indicator. In the assessment of land use impacts, characterization factors are developed so as to weigh the magnitude of environmental interventions such as land occupation and land transformation on the potentially affected attributes of ecosystem quality such as biodiversity, ecological functions and natural resources.The goal of this study is to review the progress of the implementation of land use and land use change as an impact category in LCA with a particular focus on biodiversity, recognize limitations, and indicate future prospects for the development of land use impact assessment methodologies and subsequent integration into LCA. Land use impacts are not being widely integrated into LCA because they are dependent on the regional or local situation which is not well known in LCA and land use as an environmental intervention is very complex. However, the importance of land use cannot be overemphasized when assessing products or processes which make use of raw materials that originate from land extensive activities. Despite this importance, there have been diverse arguments on how to include land use impacts, for example, on biodiversity in LCA so as to provide a common and acceptable methodology for this assessment.This study focuses on how land use impacts can be included in LCA. With a particular focus on land use impacts on biodiversity, the result of this review shows that only a few studies have been carried out. The problem of non-convergence of the methodology for the assessment of land use in LCA still persists because most of the proposed methodologies deal with different aspects of land use impacts and are therefore conflicting.Most of the studies reviewed stress the importance of biodiversity measured in terms of vascular plant species diversity. However, there are other methodologies which consider other impact pathways such as life support functions. The number of studies thereby correlates with an increase in the interest in the research area. However, it is difficult to identify any trend of convergence. Different methods are being proposed which do not actually agree with one another. Some of these methods are not “closely” related to the use of land in the normal usage sense. Most of the methods being proposed are exemplified in different regions and these have not been found to be applicable to global cases. This could be a limiting factor for the applicability of the proposed methodologies in LCA. In order to overcome these shortcomings, more research work would be needed before these methodologies could be incorporated into LCA which is presumed to be a global assessment methodology. This will enhance the credibility of the results provided by an LCA and the subsequent acceptability of the LCA methodology.

Selected non-renewable and renewable energy processes from the ecoinvent 2.2 life cycle inventory database were analysed using basic contribution analysis, geometric series expansion, and structural path analysis. The hierarchical perspective of the ReCiPe impact assessment method was applied. The sources studied included biomass, wind, solar photovoltaic, hydropower, natural gas combined cycle and hard coal. Several technologies within each energy source were studied for comparison purposes. The processes were compared based on material consumption, land use and emissions for the production of 1 EJ (278 TWh). Results indicate that all of the renewable energy sources studied had a significantly lower impact than the non-renewable sources chosen. With the exception of bioenergies and pumped reservoir hydropower, technologies for the same energy source showed similar behaviour in the analyses performed.The findings from this study confirm previous work stating the environmental and human health superiority of renewable energy technologies over fossil fuel energy.

Waste management is something that affects most people. Thewaste amounts are still increasing, but the waste treatment ischanging towards recycling and integrated solutions. In Swedenproducers’responsibility for different products, a taxand bans on deposition of waste at landfills implicates areorganisation of the municipal solid waste management. Plansare made for new incineration plants, which leads to that wastecombustion comes to play a role in the reorganisation of theSwedish energy system as well. The energy system is supposed toadapt to governmental decisions on decommission of nuclearplants and decreased use of fossil fuels. Waste from private households consists of hazardous waste,scrap waste, waste electronics and wastes that to a largeextent are generated in the kitchen. The latter type has beenstudied in this thesis, except for newsprint, glass- and metalpackages that by source separation haven’t ended up in thewaste bin. Besides the remaining amount of the above mentionedfractions, the waste consists of food waste, paper, cardboard-and plastic packages and inert material. About 80-90 % of thismixed household waste is combustible, and the major part ofthat is also possible to recycle. Several systems analyses of municipalsolid waste managementhave been performed. Deposition at landfill has been comparedto energy recovery, recycling of material (plastic andcardboard) and recycling of nutrients (in food waste).Environmental impact, fuel consumption and costs are calculatedfor the entire lifecycle from the households, until the wasteis treated and the by-products have been taken care of. To stop deposition at landfills is the most importantmeasure to take as to decrease the environmental impact fromlandfills, and instead use the waste as a resource, therebysubstituting production from virgin resources (avoidingresource extraction and emissions). The best alternative tolandfilling is incineration, but also material recycling andbiological treatment are possible. Recycling of plastic has slightly less environmental impactand energy consumption than incineration. The difference issmall due to that plastic is such a small part of the totalwaste amount, and that just a small part of the collectedamount is recycled. Cardboard recycling is comparable toincineration; there are both advantages and disadvantages.Source separation of food waste may lead to higher transportemissions due to intensified collection, but severalenvironmental advantages are observed if the waste is digestedand the produced biogas substitutes diesel in busses.Composting has no environmental advantages compared toincineration, mainly due to lack of energy recovery. Therecycling options are more expensive than incineration. Theincreased cost must be seen in relation to the environmentalbenefits and decreased energy use. If the work with sourceseparation made by the households is included in the analysis,the welfare costs for source separation and recycling becomesnon-profitable. It is however doubted how much time is consumedand how it should be valuated in monetary terms. In systems analyses, several impacts are not measured.Environmental impact has been studied, but not allenvironmental impact. As the parts of the system are underconstant change, the results are not true forever. Recyclingmay not be unambiguously advantageous today, but it can be inthe future. Despite the fact that systems analysis has been developedduring 10 years in Sweden, there are still many decisions takenregarding waste management without support from systemsanalysis and use of computer models. The minority of users ispleased with the results achieved, but the systems analysis isfar from easy to use. The adaptation of tools and models to thedemands from the potential users should consider thatorganisations of different sizes have shifting demands andneeds. The application areas for systems analysis and models arestrategic planning, decisions about larger investments andeducation in universities and within organisations. Systemsanalysis and models may be used in pre-planning procedures. Apotential is a more general application (Technology Assessment)in predominantly waste- and biofuel based energy processes, butalso for assessment of new technical components in a systemsperspective. The methodology and systems approach developedwithin the systems analysis has here been transformed to anassessment of environmental, economic and technical prestandaof technical systems in a broad sense.

Land use refers to the use of land for intensive human activities aiming at exclusive use of land for certain purposes and adapting the properties of land areas in view of these purposes. Environmental problems are, however, generated as a result of these human activities which modify the shape and properties of large land areas according to the requirements of human activities and thereby excluding wild animals and plants from coexisting on such land areas and in their neighbourhoods. Land use also leads to the degradation of the natural environment. Life Cycle Assessment (LCA) methodology is used for evaluating the environmental burdens associated with products or processes while taking their whole life cycle into consideration. LCA is a comprehensive assessment method which considers all aspects of natural environment, human health, and resources. Land use is regarded as an impact category in Life Cycle Assessment and is treated as such. However, the environmental impacts associated with land use and land use change are not being adequately considered in LCA, if considered at all.Life Cycle Impact Assessment is a part of LCA and is aimed at understanding and evaluating the magnitude and significance of the potential environmental impacts of products or processes and this involves developing characterization factors which link an environmental impact to a category indicator. In the assessment of land use impacts, characterization factors are developed so as to weigh the magnitude of environmental interventions such as land occupation and land transformation on the potentially affected attributes of ecosystem quality such as biodiversity, ecological functions and natural resources.The goal of this study is to review the progress of the implementation of land use and land use change as an impact category in LCA with a particular focus on biodiversity, recognize limitations, and indicate future prospects for the development of land use impact assessment methodologies and subsequent integration into LCA. Land use impacts are not being widely integrated into LCA because they are dependent on the regional or local situation which is not well known in LCA and land use as an environmental intervention is very complex. However, the importance of land use cannot be overemphasized when assessing products or processes which make use of raw materials that originate from land extensive activities. Despite this importance, there have been diverse arguments on how to include land use impacts, for example, on biodiversity in LCA so as to provide a common and acceptable methodology for this assessment.This study focuses on how land use impacts can be included in LCA. With a particular focus on land use impacts on biodiversity, the result of this review shows that only a few studies have been carried out. The problem of non-convergence of the methodology for the assessment of land use in LCA still persists because most of the proposed methodologies deal with different aspects of land use impacts and are therefore conflicting.Most of the studies reviewed stress the importance of biodiversity measured in terms of vascular plant species diversity. However, there are other methodologies which consider other impact pathways such as life support functions. The number of studies thereby correlates with an increase in the interest in the research area. However, it is difficult to identify any trend of convergence. Different methods are being proposed which do not actually agree with one another. Some of these methods are not “closely” related to the use of land in the normal usage sense. Most of the methods being proposed are exemplified in different regions and these have not been found to be applicable to global cases. This could be a limiting factor for the applicability of the proposed methodologies in LCA. In order to overcome these shortcomings, more research work would be needed before these methodologies could be incorporated into LCA which is presumed to be a global assessment methodology. This will enhance the credibility of the results provided by an LCA and the subsequent acceptability of the LCA methodology.

Selected non-renewable and renewable energy processes from the ecoinvent 2.2 life cycle inventory database were analysed using basic contribution analysis, geometric series expansion, and structural path analysis. The hierarchical perspective of the ReCiPe impact assessment method was applied. The sources studied included biomass, wind, solar photovoltaic, hydropower, natural gas combined cycle and hard coal. Several technologies within each energy source were studied for comparison purposes. The processes were compared based on material consumption, land use and emissions for the production of 1 EJ (278 TWh). Results indicate that all of the renewable energy sources studied had a significantly lower impact than the non-renewable sources chosen. With the exception of bioenergies and pumped reservoir hydropower, technologies for the same energy source showed similar behaviour in the analyses performed.The findings from this study confirm previous work stating the environmental and human health superiority of renewable energy technologies over fossil fuel energy.

An emerging interest in constructing ultra low energy buildings, with low impact materials and maximizing the potential of using renewable energy reflects the potential in building industry to significantly contribute towards reducing environmental impacts. Life cycle assessments of the different green building prototypes provide a means to estimate the impacts of such buildings as well as provide suggestive improvements. The Active house in Stjørdal, Norway is one such prototype of a green building. This is a single family residence which is built with a concept of solar architecture in ultra low-energy buildings. It is challenging to harness solar energy at high latitudes. The Active house uses the fundamental construction details for a Passive house as mentioned in Norwegian regulatory standard, with specific changes in increasing the glazed surface to promote passive solar heat gain as well as increase daylighting , but also making it vulnerable to heat loss. The house is based on timber framework. Apart from electricity the house uses solar collectors which are connected to the hot water storage and hydronic floor heating. Space heating is also compensated by use of wood stoves. In the LCA results suggest that, based on the construction the Active house requires ten percent more energy than an equivalent Passive house which uses only electricity and wood. However, due to the effectivity of the solar collectors, it compensates for the need of the extra energy and in a lifetime of 60 years, it performs 15 % better , contributing to lesser environmental impacts than an equivalent Passive house. It is understood that extra embodied energy does not affect the environmental performance of a building if it results in better environmental performance (1). However, it is important to create demonstrable value of the building for the end user. Lifecycle assessment results from simulated operational use carries considerable error with respect to how the building actually performs. The results in this study have also been estimated with an approximate error factor derived from previous studies (2). There is a necessity to make every stakeholder of the building participative in the functioning of the building, inclusive of the end user, and maintaining the well-being. The case has also been scored in the basic categories of a sustainibility certification, with the results available from the lifecycle assessment and energy simulation.

A systems thinking based approach was used to define and investigate the current state of knowledge in the academic disciplines related to sustainable residential dwellings via literature review. Semi-structured interviews were conducted with respondents connected to the system definition. The goal of the study was to determine the extent to which the academic disciplines, as well as the respondents, incorporated the social, economic, environmental and intergenerational aspects of sustainability. Significant variation was found in both the disciplines as well as in the responses of interviewees. Life-cycle cost analysis of dwellings built to the Norwegian passive house standard was used to investigate the implications of using alternative discount factors for extended assumed life spans. The results indicated that alternative discount factors have the potential to significantly reduce rental costs while fulfilling the potential Pareto optimality criterion. The alternative discount factors used in the life-cycle cost analysis were later shown to have flaws which limit their usefulness; a methodology for deriving a representative multigenerational discount rate was proposed.

In the coming years, organic food waste will be collected in the region of Gästrikland, Sweden. The collection is planned to cover the entire region by the end of year 2007. To start with, smaller amounts are being collected and transported to a central composting plant in Sala. However, a central composting plant in the region of Gästrikland is projected. The objective of this thesis is to evaluate a number of possible methods for composting of organic food wastes regarding their environmental impacts and localization. There are four scenarios for future composting of organic waste that have been evaluated. There are two that consist of membrane composting, either in Sala or in the region of Gästrikland. The other two scenarios consist of tunnel composting in the region of Gästrikland with two possible placements. The result from the systems analysis shows a lower environmental impact when using tunnel composting, compared to membrane composting. As the tunnel compost uses technologies for treating the compost gas, the amount of substances that can contribute to acidification and eutrophication is considerably lowered. However, the use of electricity is higher since the automatic process demands a greater electricity input. At a membrane composting plant, vehicles are being used to move compost material. The combustion of diesel oil gives rise to gases that increase the global warming. According to future legislations, tunnel composting gives an easier control of the emissions and optimization of the compost process. / Under de kommande åren kommer organiskt avfall samlas in från Gästrikeregionen för att komposteras. Insamlingen beräknas täcka hela regionen vid årsskiftet 2007/2008. Till en början samlas mindre mängder in och transporteras till en komposterings-anläggning i Sala. I framtiden planeras dock en komposteringsanläggning i Gästrikeregionen. Syftet med examensarbetet var att utvärdera ett antal tänkbara komposterings-anläggningar för matavfallet med avseende på deras miljöpåverkan, samt deras lokalisering. Fyra framtidsscenarier för kompostering av det organiska avfallet har utvärderats. Två innefattar membrankompostering, antingen i Sala eller i regionen. De följande två scenarierna innebär tunnelkompostering i Gästrikeregionen, vid Forsbacka avfallsdeponi eller vid ett område söder om Forsbacka. Resultaten från systemanalysen påvisar en lägre miljöpåverkan från en tunnelkompost än från en membrankompost. På grund av en högre rening av kompostgasen förhindras övergödande och försurande ämnen att släppas ut. Tunnelkompostering innebär dock en högre elförbrukning jämfört med membrankompostering eftersom stora delar av anläggningen är automatiserad. Membrankompostering förbrukar däremot mer diesel än tunnelkompostering vid förflyttning av kompostmaterialet, vilket ger högre utsläpp av framförallt klimatpåverkande gaser. Generellt kan sägas att tunnelkompostering innebär att anläggningens emissioner lättare kan kontrolleras. Dessutom kan styrningen av kompostprocessen lättare förändras i enlighet med framtida krav och bestämmelser.

The aim of this study is to provide an overview of the core factors for the environmental performance of future Norwegian high-speed rail (HSR) and to study their likely development up to 2050 in a life-cycle assessment (LCA) perspective. The analysis included the infrastructure, rolling stock and operations. This work was conducted with MiSA, an environmental consulting company based in Trondheim, Norway. MiSA recently completed a life-cycle inventory (LCI) for HSR in Norway. To start with, core factors were chosen through a literature review. The corridor Oslo-Trondheim was then modeled using the new LCI in order to establish a set of the core factors to analyze. The LCA was performed with SimaPro. LCA literature is the preferred source for emissions data. First because results show that emissions must cover life-cycle emissions from fuel, electricity, materials and processing (source-to-wheel). Second, LCA provides guidelines for good practice for environmental accounting and benchmarking of transport alternatives. Chapter 4 is an investigation of the core factors. Through the study of technical writings for current and future use of HSR in Norway, as well as sensitivity analyses, certain core factors were earmarked to produce detailed scenarios for future use up to 2050.Cement, steel, XPS, infrastructure, deforestation and the number of passengers per day are core factors. Cement, steel and XPS are the materials that have the most impact. The impact of the infrastructure of future Norwegian HSR is high because the number of passengers and the carbon footprint (CF) of the electricity mix used for operation are low. Norwegian HSR is lacking passengers. A high number of passengers in the Norwegian context constitutes a low number of passengers in other European countries. A high potential for change is to abstract passengers from air travel, which is the most used mode of transport in Norway in 2010. The energy used for operation and the energy per seat-km are not core factors because the electricity mix used for operation has a low CF (166 g CO2/kWh). The impact of HSR is reduced on average by 17% by updating the database (scenario updated 2010). The impact is reduced by 50% in a likely future (scenario 2050) by improving the production technology of the materials for the infrastructure and by having more passengers. Finally, the impact is reduced by 60% by, in addition to changes from scenario 2050, setting specific requirement to the suppliers and by having an active yield management (scenario 2050+).