In this thesis, the reliability of small wind turbines is studied. Both conventional reliability analysis methods and the novel Bayesian models (Bayesian Hierarchical Modelling (BHM)) are used to analyse the reliability performance of the Gaia-Wind turbines / assemblies and components of the Gaia-Wind turbine. In Chapter 2, a simple failure mode and effect analysis (FMEA) is conducted. An approximated risk priority number (RPN) is calculated for each failure mode and assembly. The assembly that is identified to have the highest RPN is the "Rotor and Blade Assembly". As for the failure modes, "Blade Split" and "Generator Failure" failure modes are identified to have the highest RPNs. In Chapter 3, the conventional methods including the Kaplan-Meier Analysis, Weibull Plot Analysis, Homogeneous Poisson Process (HPP) Analysis, and Crow-AMSAA (Non-Homogeneous Poisson Process (NHPP)) Analysis are used to study the reliability performance of the generic turbine and the critical assemblies based on the approximated RPNs. By using these conventional methods, the L10 life can be approximated (Kaplan-Meier), the main failure modes of an assembly can be identified (Weibull Plot Analysis), the annual failure rate can be estimated (HPP), and the number of future failures can be predicted (NHPP). These methods have been implemented in a novel on-line interactive platform, named ReliaOS (Chapter 7), which effectively facilitates the process of converting the information in the warranty record to the meaningful reliability information. Three novel BHM models are proposed and implemented in WinBUGS (an open source software), namely the repair model, the environmental model, and the informative prior framework, (Chapter 5 and Chapter 6). The repair model is used to quantify the repair effectiveness of a generic repair action. The model is applied on both the turbine level as well as the component level. At the turbine level, the annual failure rate of the generic turbine is predicted to be 0:159 per turbine per year at the first year. Individual turbines can be categorised into different quality levels ("Good", "Good- Normal", "Normal", "Normal-Bad", and "Bad") based on the predicted annual failure rate values. At the component level, "Blade split", "Cracked Frame", and "Generator Failure" failure modes are studied. These are the most critical failure modes for "Rotor and Blade Assembly", "Tower, Foundation, and Nacelle", and "Generator" assemblies respectively. "Cracked Frame" failure mode is predicted to have the lowest characteristic life and a slightly increasing failure rate trend. The repair effectiveness of the "Cracked Frame" failure mode is identified to be slightly ineffective. The environmental model quantifies the influence of three environmental covariates, i.e. AverageWind Speed (AWS), Turbulence Intensity (TI), and Terrain Slope (TS). These environmental covariates are all identified to have negative impact to the reliability of the generic turbine, where TI and AWS have more pronounced impact than TS. The informative prior BHM framework offers a way of quantifying the reliability of the drivetrain frame (which corresponds to the "Cracked Frame" failure mode) in a situation where zero failure instance is recorded for the new drivetrain frame design. This is achieved by jointly considering the simulation results from SOLIDWORKS as the prior information into the BHM model. This thesis strives to understand the reliability performance of the Gaia-Wind small wind turbine from different perspectives, i.e. the generic turbine, individual turbines, and the components, by the use of conventional methods and the proposed BHM models. The novel on-line reliability platform, ReliaOS, mitigates the difficulties in converting the information in the data to the reliability information for the end users. It is believed that the proposed BHM models and the ReliaOS on-line reliability analysis platform will improve the reliability analysis of small-wind turbines.

O presente trabalho apresenta uma metodologia de simulação numérica de perfis aerodinâmicos bidimensionais com foco na utilização para o projeto e otimização de pás e rotores de pequenas turbinas eólicas de eixo horizontal, bem como o emprego desses métodos em simulações nas quais efeitos de alta solidez do rotor e baixos números de Reynolds são avaliados. Essa metodologia inclui geração de malhas, seleção de métodos numéricos e validação, tendo as escolhas sido guiadas pelas práticas mais bem sucedidas na simulação de perfis aerodinâmicos, e foi aplicada na simulação dos aerofólios NACA 0012, S809 e SD7062. O código comercial ANSYS Fluent foi utilizado em todas as simulações. Na simulação de aerofólios isolados a altos números de Reynolds dos perfis NACA 0012 e S809, o modelo Transition SST (γ-Reθ) apresentou resultados mais próximos a dados experimentais do que aqueles apresentados pelo modelo k-ω SST para CL e CD, além de produzir resultados para CP que mostraram boa precisão quando comparados aos mesmos dados experimentais. Resultados de CL, CD, CF e CP são apresentados para 20 diferentes condições de operação às quais o perfil SD7062 foi submetido, com números de Reynolds variando entre 25.000 e 125.000. As distribuições dos dois últimos coeficientes sobre os dorsos do aerofólio evidenciam com clareza a presença e magnitude da bolha de separação laminar. Os coeficientes de sustentação e arrasto mostram o impacto negativo da presença da bolha nessa faixa de números de Reynolds. Além disso, nos casos simulados, o arrasto aumenta em função da diminuição do Re. Um design de pá produzido com o auxílio do código de otimização SWRDC, baseado em algoritmos genéticos, é apresentado. Três seções ao longo da envergadura dessa pá foram simuladas em uma bateria de 45 simulações, sob diversas condições de operação em função de solidez, ângulo de ataque e razão de velocidade de ponta de pá. Esses resultados mostram que a bolha de separação laminar se move na direção do bordo de ataque com o aumento da solidez, do ângulo de ataque e da TSR. Além disso, distribuições do CP mostram aumento de pressão em ambos os dorsos do perfil quando submetido aos efeitos da solidez, embora esses efeitos tenham sido responsáveis por um aumento na relação CL/CD nos casos estudados. / This thesis presents a methodology of two-dimensional airfoil simulation focusing on its application on the design and optimization of blades and rotors of small horizontal axis wind turbines, and its application in a set of numerical simulations involving high rotor solidity and low-Re effects. This methodology includes grid generation, selection of numerical methods and validation, reflecting the most successful practices in airfoil simulation, and was applied in the simulation of the NACA 0012, S809 and SD7062 airfoils. The ANSYS Fluent commercial code was used in all simulations. Results for the isolated NACA 0012 and S809 airfoils at high Reynolds numbers show that the Transition SST (γ-Reθ) turbulence model produces results closer to experimental data than those yielded by the SST k-ω model for CL and CD, having also produced CP plots that show good agreement to the same experimental data. Plots of CL, CD, CF and CP for the SD7062 airfoil are presented, for simulations at 20 different operating conditions. The CF and CP distributions evidence the negative impact of the laminar separation bubble in the range of Reynolds numbers evaluated. Results show that, for Re between 25,000 and 125,000, drag increases with decreasing Re. A blade design generated using the SWRDC optimization code, based on genetic algorithms, is presented. Three sections of the resulting blade shape were selected and were tested in a set of 45 simulations, under an array of operating conditions defined by solidity, angle of attack and TSR. Results show that the laminar separation bubble moves towards the leading edge with increasing solidity, angle of attack and TSR. Furthermore, CP plots show an increase in pressure on both surfaces when the airfoil is subject to solidity effects, although these effects show an increase in the lift-to-drag ratio at the conditions evaluated.

The market of wind power as a sustainable energy source is growing, both on large and small scale. Conventional large scale wind turbines normally operate in uniform areas where expected wind speeds and turbulence characteristics are well investigated and the constructional design of the wind turbines is regulated by standard classes for different external conditions. Small scale wind turbines (SWT), on the other hand, are sometimes placed in more complex environments where the turbulence conditions are rougher. A larger amount of turbulence will generate a larger amount of fatigue loadings on the construction, increasing the risk of breakdown. It is therefore of major concern to perform more measurements and further investigate the turbulence characteristics in complex environments and the effect that these will have on small wind turbine construction. Thus, turbulence is measured with sonic anemometers at two sites with complex environments; at an urban site above a rooftop in a medium sized city (Uppsala, Sweden) and above a forest in Norunda (outside Uppsala) at two heights, near the treetops (z = 33 m) defined as complex and further up (z = 97 m) defined as more uniform. The turbulence data is analyzed and the results are compared to the normal turbulence model (NTM) as it is defined for the standard SWT classes by the International Electrotechnical Commission in the International standard 61400-2: Design requirements for small wind turbines (IEC, 2006). Measurements of  minute standard deviations of longitudinal wind speed (σu) and turbulence intensity (TIu) are reported, as well as the distributions of TIu and of 10  minute mean wind speeds (um) for the different sites and stabilities. The results show that the NTM represents the turbulence at 97 m height above the forest only for light wind speeds, smaller than 10 m/s, but underestimates the turbulence for higher wind speeds.  It should also be noted that the data is scattered and contain a number of occasions with extreme values of σu and TIu. For wind speeds higher than 10 m/s the number of observations is limited but the majority of the observations are more extreme than the NTM. At the complex sites (near the treetops and the rooftop) the NTM clearly underestimates both the magnitude and rate of change of σu with increasing wind speed, although the observed wind speeds close to these rough surfaces are low so the conclusions are limited. Average TIu at 97 m height is 19 %, compared to 41 % close above forest and 43 % above rooftop. Mean values of TIu above forest are generally 10 % lower during stable conditions (z/L > 0.05) while above rooftop, the wind material is sparse and 95 % of the observations had stable stratification so no dependence on stability can be seen. From these results it can be concluded that the turbulence characteristics close above treetops is similar to those above rooftop, but that the NTM, as it is defined for the standard SWT classes, is not valid in these complex and urban terrains and need to be modified to correctly estimate the turbulence intensities, and consequently also the loadings, affecting small wind turbines located at these kinds of sites. / Marknaden för vindkraft som en förnyelsebar energikälla växer snabbt, både stor- och småskaligt. Traditionella storskaliga vindkraftverk placeras normalt på homogena platser där vindklimatet och turbulensens karaktär är ganska väl kartlagda och konstruktionsstandarden regleras av standardklasser utifrån olika externa förhållanden. Små vindkraftverk (SWT) å andra sidan placeras ofta i mer komplex eller urban miljö där turbulensen är mer intensiv. En större andel turbulens genererar större utmattningslaster på konstruktionen vilket ökar risken att vindturbinen går sönder. Det är därför av stor vikt att utföra fler mätningar och ytterligare undersöka turbulensen i komplexa miljöer och vilken effekt den kommer ha på de små vindkraftverkens konstruktion. Med anledning av detta så har turbulensdata analyserats från mätningar med sonicanemometrar. Dels på en urban plats, ovanför ett hustak i en medelstor stad (Uppsala, Sverige). Dels vanför en skog i Norunda (utanför Uppsala) på två höjder, nära trädtopparna (33 m) som anses komplex och högre upp (97 m) som anes mer homogen. Resultaten är jämförda med den normala turbulensmodellen (NTM) så som den definieras för standard SWT klasserna av International Electrotechnical Commission i International standard 61400-2: Design requirements for small wind turbines (IEC, 2006). Mätningar av 10  minuters standardavvikelse av den longitudinella vindhastigheten (σu) och turbulensintensiteten (TIu) redovisas, liksom fördelningen av TIu och 10 minuters medelvinden (um) för olika stabilitet för de olika mätplatserna. Resultaten visar att NTM är representativ på 97 m höjd endast för låga vindhastigheter, under 10 m/s, medan modellen underskattar turbulensen för högre vindhastigheter. Det bör också noteras att spridningen är stor i data och att extrema värden av σu och TIu uppmätts vid flertalet tillfällen. För vindhastigheter över 10 m/s så är antalet mätvärden begränsade, men majoriteten av mätvärdena är högre än NTM. På de komplexa mätplatserna (nära trädtopparna och ovan hustaket) så underskattar NTM avsevärt både storleken av σu och dess förändring med ökad vindhastighet på de komplexa platserna (nära trädtopparna och ovan hustaket). Dock är de observerade vindhastigheterna låga såhär nära de skrovliga ytorna så slutsatserna är begränsade. På 97 m höjd är medelvärdet av TIu 19 %, jämfört med 41 % nära trädtopparna och 43 % ovan hustak. De är generellt 10 % lägre under stabila förhållanden (z/L > 0.05)  över skog, medan ovan hustak där vindmaterialet är begränsat och 95 % av observationerna var stabilt skiktade så ses inte något stabilitetsberoende. Från dessa resultat kan slutsatserna dras att turbulensens karaktär nära trädtoppar liknar den ovan hustak, men att NTM, så som den definieras för standard SWT klasserna, inte gäller vid dessa komplexa och urbana platser och behöver modifieras för att korrekt uppskatta turbulensintensiteterna och därmed också de laster som påverkar små vindkraftverk placerade på den här typen av platser.

O presente trabalho apresenta uma metodologia de simulação numérica de perfis aerodinâmicos bidimensionais com foco na utilização para o projeto e otimização de pás e rotores de pequenas turbinas eólicas de eixo horizontal, bem como o emprego desses métodos em simulações nas quais efeitos de alta solidez do rotor e baixos números de Reynolds são avaliados. Essa metodologia inclui geração de malhas, seleção de métodos numéricos e validação, tendo as escolhas sido guiadas pelas práticas mais bem sucedidas na simulação de perfis aerodinâmicos, e foi aplicada na simulação dos aerofólios NACA 0012, S809 e SD7062. O código comercial ANSYS Fluent foi utilizado em todas as simulações. Na simulação de aerofólios isolados a altos números de Reynolds dos perfis NACA 0012 e S809, o modelo Transition SST (γ-Reθ) apresentou resultados mais próximos a dados experimentais do que aqueles apresentados pelo modelo k-ω SST para CL e CD, além de produzir resultados para CP que mostraram boa precisão quando comparados aos mesmos dados experimentais. Resultados de CL, CD, CF e CP são apresentados para 20 diferentes condições de operação às quais o perfil SD7062 foi submetido, com números de Reynolds variando entre 25.000 e 125.000. As distribuições dos dois últimos coeficientes sobre os dorsos do aerofólio evidenciam com clareza a presença e magnitude da bolha de separação laminar. Os coeficientes de sustentação e arrasto mostram o impacto negativo da presença da bolha nessa faixa de números de Reynolds. Além disso, nos casos simulados, o arrasto aumenta em função da diminuição do Re. Um design de pá produzido com o auxílio do código de otimização SWRDC, baseado em algoritmos genéticos, é apresentado. Três seções ao longo da envergadura dessa pá foram simuladas em uma bateria de 45 simulações, sob diversas condições de operação em função de solidez, ângulo de ataque e razão de velocidade de ponta de pá. Esses resultados mostram que a bolha de separação laminar se move na direção do bordo de ataque com o aumento da solidez, do ângulo de ataque e da TSR. Além disso, distribuições do CP mostram aumento de pressão em ambos os dorsos do perfil quando submetido aos efeitos da solidez, embora esses efeitos tenham sido responsáveis por um aumento na relação CL/CD nos casos estudados. / This thesis presents a methodology of two-dimensional airfoil simulation focusing on its application on the design and optimization of blades and rotors of small horizontal axis wind turbines, and its application in a set of numerical simulations involving high rotor solidity and low-Re effects. This methodology includes grid generation, selection of numerical methods and validation, reflecting the most successful practices in airfoil simulation, and was applied in the simulation of the NACA 0012, S809 and SD7062 airfoils. The ANSYS Fluent commercial code was used in all simulations. Results for the isolated NACA 0012 and S809 airfoils at high Reynolds numbers show that the Transition SST (γ-Reθ) turbulence model produces results closer to experimental data than those yielded by the SST k-ω model for CL and CD, having also produced CP plots that show good agreement to the same experimental data. Plots of CL, CD, CF and CP for the SD7062 airfoil are presented, for simulations at 20 different operating conditions. The CF and CP distributions evidence the negative impact of the laminar separation bubble in the range of Reynolds numbers evaluated. Results show that, for Re between 25,000 and 125,000, drag increases with decreasing Re. A blade design generated using the SWRDC optimization code, based on genetic algorithms, is presented. Three sections of the resulting blade shape were selected and were tested in a set of 45 simulations, under an array of operating conditions defined by solidity, angle of attack and TSR. Results show that the laminar separation bubble moves towards the leading edge with increasing solidity, angle of attack and TSR. Furthermore, CP plots show an increase in pressure on both surfaces when the airfoil is subject to solidity effects, although these effects show an increase in the lift-to-drag ratio at the conditions evaluated.

O presente trabalho apresenta uma metodologia de simulação numérica de perfis aerodinâmicos bidimensionais com foco na utilização para o projeto e otimização de pás e rotores de pequenas turbinas eólicas de eixo horizontal, bem como o emprego desses métodos em simulações nas quais efeitos de alta solidez do rotor e baixos números de Reynolds são avaliados. Essa metodologia inclui geração de malhas, seleção de métodos numéricos e validação, tendo as escolhas sido guiadas pelas práticas mais bem sucedidas na simulação de perfis aerodinâmicos, e foi aplicada na simulação dos aerofólios NACA 0012, S809 e SD7062. O código comercial ANSYS Fluent foi utilizado em todas as simulações. Na simulação de aerofólios isolados a altos números de Reynolds dos perfis NACA 0012 e S809, o modelo Transition SST (γ-Reθ) apresentou resultados mais próximos a dados experimentais do que aqueles apresentados pelo modelo k-ω SST para CL e CD, além de produzir resultados para CP que mostraram boa precisão quando comparados aos mesmos dados experimentais. Resultados de CL, CD, CF e CP são apresentados para 20 diferentes condições de operação às quais o perfil SD7062 foi submetido, com números de Reynolds variando entre 25.000 e 125.000. As distribuições dos dois últimos coeficientes sobre os dorsos do aerofólio evidenciam com clareza a presença e magnitude da bolha de separação laminar. Os coeficientes de sustentação e arrasto mostram o impacto negativo da presença da bolha nessa faixa de números de Reynolds. Além disso, nos casos simulados, o arrasto aumenta em função da diminuição do Re. Um design de pá produzido com o auxílio do código de otimização SWRDC, baseado em algoritmos genéticos, é apresentado. Três seções ao longo da envergadura dessa pá foram simuladas em uma bateria de 45 simulações, sob diversas condições de operação em função de solidez, ângulo de ataque e razão de velocidade de ponta de pá. Esses resultados mostram que a bolha de separação laminar se move na direção do bordo de ataque com o aumento da solidez, do ângulo de ataque e da TSR. Além disso, distribuições do CP mostram aumento de pressão em ambos os dorsos do perfil quando submetido aos efeitos da solidez, embora esses efeitos tenham sido responsáveis por um aumento na relação CL/CD nos casos estudados. / This thesis presents a methodology of two-dimensional airfoil simulation focusing on its application on the design and optimization of blades and rotors of small horizontal axis wind turbines, and its application in a set of numerical simulations involving high rotor solidity and low-Re effects. This methodology includes grid generation, selection of numerical methods and validation, reflecting the most successful practices in airfoil simulation, and was applied in the simulation of the NACA 0012, S809 and SD7062 airfoils. The ANSYS Fluent commercial code was used in all simulations. Results for the isolated NACA 0012 and S809 airfoils at high Reynolds numbers show that the Transition SST (γ-Reθ) turbulence model produces results closer to experimental data than those yielded by the SST k-ω model for CL and CD, having also produced CP plots that show good agreement to the same experimental data. Plots of CL, CD, CF and CP for the SD7062 airfoil are presented, for simulations at 20 different operating conditions. The CF and CP distributions evidence the negative impact of the laminar separation bubble in the range of Reynolds numbers evaluated. Results show that, for Re between 25,000 and 125,000, drag increases with decreasing Re. A blade design generated using the SWRDC optimization code, based on genetic algorithms, is presented. Three sections of the resulting blade shape were selected and were tested in a set of 45 simulations, under an array of operating conditions defined by solidity, angle of attack and TSR. Results show that the laminar separation bubble moves towards the leading edge with increasing solidity, angle of attack and TSR. Furthermore, CP plots show an increase in pressure on both surfaces when the airfoil is subject to solidity effects, although these effects show an increase in the lift-to-drag ratio at the conditions evaluated.

In diesem Beitrag wird eine fertigungsoptimierte Methode der Nutrastmomentreduzierung durch Magnetverschiebung am Beispiel eines Synchrongenerators für Kleinwindenergieanlagen mit in die Läuferoberfläche eingelassenen Permanentmagneten vorgestellt. Auf Grundlage der Drehfeldtheorie wird ein analytisches Berechnungsmodell unter Einbezug der Magnetverschiebung zur Bestimmung des Nutrastmoments entwickelt und mit FEM-Simulationen verifiziert. Durch experimentelle Untersuchung eines gefertigten Prototyps wird der optimierende Effekt der Magnetverschiebung nachgewiesen und mit einem Referenzläufer mit ungeschrägt, symmetrisch verteilten Polen verglichen. Abschließend werden mögliche Ursachen für die verbleibenden Abweichungen identifiziert und Ansätze für die weiteren Betrachtungen abgeleitet. / A production-optimized method of reducing cogging torque by magnet shifting using the example of a synchronous generator for small wind turbines with a rotor with inset surface magnets is proposed in this paper. On the basis of rotating field theory an analytical calculation model is developed taking into account the magnet shifting for determination of cogging torque. The analytical results are verified with FEM simulations. By experimental investigation of a manufactured prototype, the optimizing effect of magnet shifting is verified and compared to a rotor with symmetrically distributed poles. Finally, possible reasons for remaining discrepancies are identified and approaches for further considerations are presented.

Thesis (MScEng (Mechanical and Mechatronic Engineering))--Stellenbosch University, 2008. / The sustainability and economy of the current South African National Antarctic Expedition IV (SANAE IV) base diesel-electric power system are threatened by the current high fuel prices and the environmental pollution reduction obligations. This thesis presents the potential technical, environmental and economical challenges associated with the integration of small wind energy conversion system (WECS) with the current SANAE IV diesel fuelled power system. Criteria derived from technical, environmental and economic assessments are applied in the evaluation of eight commercially available wind turbines as to determine the most technically and economically feasible candidates. Results of the coastal Dronning Maud Land and the local Vesleskarvet cold climate assessments based on long term meteorological data and field data are presented. Field experiments were performed during the 2007-2008 austral summer. These results are applied in the generation of a wind energy resource map and in the derivation of technical wind turbine evaluation criteria. The SANAE IV energy system and the electrical grid assessments performed are based on long term fuel consumption records and 2008 logged data. Assessment results led to the identification of SANAE IV specific avoidable wind turbine grid integration issues. Furthermore, electro-technical criteria derived from these results are applied in the evaluation of the eight selected wind turbines. Conceptual wind turbine integration options and operation modes are also suggested. Wind turbine micro-siting incorporating Vesleskarvet specific climatological, environmental and technical related issues are performed. Issues focusing on wind turbine visual impact, air traffic interference and the spatial Vesleskarvet wind distribution are analysed. Three potential sites suited for the deployment of a single or, in the near future, a cluster of small wind turbines are specified. Economics of the current SANAE IV power system based on the South African economy (May 2008) are analysed. The life cycle economic impact associated with the integration of a small wind turbine with the current SANAE IV power system is quantified. Results of an economic sensitivity analysis are used to predict the performance of the proposed wind-diesel power systems. All wind turbines initially considered will recover their investment costs within 20 years and will yield desirable saving as a result of diesel fuel savings, once integrated with the SANAE IV diesel fuelled power system. Finally, results of the technical and economical evaluation of the selected commercially available wind turbines indicated that the Proven 6 kWrated, Bergey 10 kWrated and Fortis 10 kWrated wind turbines are the most robust and will yield feasible savings.