Search

Wind energy is a field with a large number of uncertainties. The random nature of the weather conditions, including wind speed, wind direction, and turbulence intensity, influences the energy output and the structural safety of a wind farm, making its performance fluctuate and difficult to predict. The uncertainties presented in the energy output and structure lifetime lead to increased investment risk. There are possibilities to reduce the risk associated with these uncertainties by optimizing the design of the farm or the wind turbine, with respect to the stochastic parameters. The goal of this project is to improve the wind farm optimization problem by providing accurate and computationally efficient annual energy production (AEP) estimates, which is a uncertainty quantification that is required in every optimization step. Uncertainty quantification has been recognized as a challenge in the wind energy industry, as the chaotic nature of the weather condition complicates the prediction of energy production. High-fidelity wind farm models usually employ advanced models like Large Eddy Simulation or Reynolds averaged Navier-Stokes equation for better accuracy. However, the prolonged computation time of these high-fidelity models make the traditional uncertainty quantification approach like the Monte-Carlo simulation or other integration techniques infeasible for larger wind farms. To overcome this limitation, the report proposes the use of generalized polynomial chaos expansion (PCE) to characterize the AEP as a function of wind speed and wind direction. PCE is a technique that approximates a random variable using a series of orthogonal polynomials, the polynomials are chosen based on the target distribution. This report explains how a surrogate model of the AEP can be constructed using PCE, which can be used in optimization or model analysis. The objective of the thesis work is to minimize the number of model evaluations required for obtaining an accurate energy response surface. Different ideas of non-intrusive PCE are implemented and explored in this project. The report demonstrates that, the multi-element polynomial chaos fitted by point collocation, with a dependent polynomial basis, is not only able to make accurate and stable (with respect to the placement of the measurements) energy predictions, but also produces realistic energy response surface. Vindkraft är en bransch med många osäkerheter, där väderförhållandena påverkar energiproduktionen och strukturens livslängd. Denna osäkerhet ökar investeringsrisken, men kan minskas genom optimering av vindkraftverkets design med hänsyn till de stokastiska parametrarna. Syftet med denna rapport är att förbättra optimeringsproblemet för vindkraftverk genom att ge noggranna och effektiva årliga energiproduktionsberäkningar (AEP), vilket krävs vid varje optimeringssteg. I rapporten används polynomial chaos expansion (PCE) för att approximera AEP och minska antalet nödvändiga modellutvärderingar. Resultaten visar att PCE är en effektiv metod för att göra energiprognoser.

In this study, the local similarity theory functions were calculated with two different approaches and on the other hand the mean hub height wind speed was compared with the rotor equivalent wind speed. Both calculations are based on two independent data-sets from measurement campaigns Hornamossen and Ryningsnäs which were conducted in the south of Sweden between May 2015 and June 2017, and November 2010 and February 2012. The first campaign includes measurements between 100 and 173 m and the second 98 and 138 m. In general, the aims were to validate if the results with reference functions and to compare the results from both approaches. The local similarity theory was used, because well above the ground, the assumption of a constant flux layer typically does not hold especially in the stable boundary layer. The used approaches are the flux-gradient and Richardson number formulation. Based on those, the non-dimensional universal functions for momentum and heat could be calculated and those could be presented as functions of the stability parameter. As shown in this study, the scatter the Richardson number formulation results are significant smaller compared to the flux-gradient formulation. One reason can be that the stability parameter and the universal functions for momentum and heat depend solely on the Richardson number. Despite the higher scatter, the medians of the universal function for momentum based on the flux-gradient formulations for both data-sets agree also with the references. Furthermore, for the results of the universal function for heat based on the flux-gradient formulation agree with the references if the minimum limit for the kinetic heat flux is significantly higher than for the universal function for momentum. Furthermore, in the publication from England & McNider 1995, who derived the Richardson number formulation, includes two erroneous equations for stable stratification. One of them has a tipping error and the other was incorrectly derived. Thus, the corrected equations are presented in this work. This work also presents new equations which are not based on the assumption that the constants of the empirical formulation for the universal function for momentum and heat with the same value. A comparison of the old and new equations show for a generated Richardson number vales a agreement of the results over the defined Richardson number range. Finally, in the wind industry it is a common practice to use the mean wind speed at the hub height as the representative mean wind speed over the entire rotor swept area. However, this assumption differs increasingly from the reality, because turbine sizes increase constantly. Thus, in this study, this common method is compared with another averaging concept. Hence, the work focuses on a area-weighted mean wind speed which is called the rotor equivalent wind speed. This average gives a better estimation of the existing wind field because it is based on multiple measurements at various heights. Since the wind gradient changes with height, those two velocities are plotted as functions of the same stability parameter as above. The main results in unstable stratification are that the hub height wind speed underestimates the rotor equivalent wind speed by about 1 to 1.5 %. In stable stratification the results vary: Two calculations show a overestimates by about 1 % and another shows no difference between those averages. Hence, the conclusion based on those findings are that the hub height wind speed is a source for a higher modelling uncertainty. On the contrary, the rotor equivalent wind speed gives more accurate modelling results. I denna studie beräknades de lokala likhetsteorifunktionerna med två olika tillvägagångssätt och å andra sidan jämförs genomsnittliga vindhastigheten vid navhöjden med rotorekvivalent vindhastigheten. Beräkningarna är baserad på två oberoende datamängder från mätningskampanjer Hornamossen och Ryningsnäs som genomfördes i södra Sverige mellan maj 2015 och juni 2017 och november 2010 och februari 2012. Första kampanj innehåller mätningar mellan 100 och 173 m och den andra 98 och 138 m. Generellt var målet att validera resultaten med referensfunktioner och jämföra både tillvägagångssätt med varandra. Lokala likhetsteorinen används eftersom för mätningar långt över marken håller antagandet om ett konstant flödesskikt vanligtvis inte. Det gäller särskilt i det stabila gränsskiktet. De två tillvägagångssätten är flödesgradientförhållandet och Richardson-talformuleringen. Baserade på de formuleringarna kan de icke-dimensionella universella funktionerna för momentum och värme beräknas och de visas som en funktion av stabilitetsparametern. I denna studie visas att spridningen av Richardsons talformuleringsresultat är signifikant mindre jämfört med andra metoden. En anledning är att stabilitetsparameter och både universella funktioner beror endast på Richardson tal. Trots den högre spridningen överensstämmer medianerna för den universella funktionen för momentum baserat på flödesgradientformuleringarna med referenserna. Detsamma gäller för resultaten av den universella funktionen för värme baserat på flödesgradientformuleringen om minimigränsen om kinetiska värmeflödet är betydligt högre än för den universella funktionen för momentum. Dessutom innehåller publikationen från England & McNider 1995, som innehåller härledning av Richardson talformulering, två felaktiga ekvationer för stabila gränsskiktet. En av dem har ett tippfel och den andra var felaktigt härledd. Detta arbete presenteras de korrigerade ekvationerna. Dessutom presenteras en uppsättning nya ekvationer där de konstanterna av den empiriska formuleringen för den universella funktionen för momentum och värme inte antas att har samma värde. Slutligen är det i vindindustrin en vanlig praxis att använda den genomsnittliga vindhastigheten vid navhöjden som den representativa medelvindhastigheten för hela "rotor swept area". Turbinstorlekarna ökar dock ständigt och därför får skillnaden mellan realitet och beräkningen alltid större. Således fokusera denna studien en areaviktad medelvindhastighet som heter rotorekvivalent vindhastighet. Den beräknar medelvindhastigheten med ett mindre osäkerhet eftersom den är baserad på flera vindmätningar på olika höjder. På grund av ett ojämt vindgradient i gränsskiktet visas resulten som funktion av densamma stailitetsparameter från likhetsteorien. Huvudresultaten för instabil gränskiktet är att navhöjdens vindhastigheten underskattar rotorekvivalent vindhastigheten med cirka 1 till 1,5 %. För det stabila gränskiktet finns olika resultaten: Två beräkningar visar att navhöjdens vindhastigheten överskattningar rotorekvivalent vindhastigheten med ungefär 1 % och en beräkning visa inget skillnad mellan medelvärdarna. Slutsatsen är att navhöjdens vindhastigheten är ett källa till ett högre modelleringsosäkerhet. Däremot visades att användningen av rotorekvivalent vindhastigheten leda till ett bättre prognosresultat. In dieser Studie wurde einerseits die lokale ähnlichkeitstheorie mit zwei unterschiedlichen Ansätzen berechnet und andererseits die durchschnittliche Nabenhöhen- mit der rotor-äquivalenten Windgeschwin- digkeit verglichen. Dafür standen zwei unabhängige Datensätze zur Verfügung, welche Messwerte zwischen 98 und 173 m beinhalteten. Die Messungen wurden in Südschweden durch die Messtürme Hornamossen und Ryningsnäs in den Zeiträumen von Mai 2015 bis Juni 2017 und von November 2012 bis Februar 2012 erhoben. Das Ziel dieser Studie war es, die Ergebnisse aus den Berechnungen mit Referenzfunktionen zu validieren und den ausgewählten Ansätzen zu vergleichen. Für Messungen mit mehr als 100 m über der Erdoberfläche ist die lokale ähnlichkeitstheorie anstelle der ähnlichkeitstheorie basierend auf der konstanten Flussschicht besser geeignet, da letztere von einem konstanten Wert ausgeht. Die Ansätze, die hierfür verwendet wurden, sind die „Flux-Gradient Formulation“ und die „Gradient Richardson Number Formulation“. Mit beiden kann die universelle Impuls- und Wärmefunktion berechnet und als Funktion des Stabilitätsparameters dargestellt werden. Wie diese Studie zeigt, ist die Streuung um die Referenzkurven sehr klein für die „Gradient Richardson Number“ Ergebnisse im Vergleich zur „Flux-Gradient Formulation“. Dies liegt daran, dass sowohl der Stabilitätsparameter und die universellen Impuls- und Wärmefunktion nur von der Richardson-Zahl abhängen. Trotz der höheren Streuung für die „Flux-Gradient Formulation“ stimmen die Mediane der universellen Impulsfunktionen von beiden Datensätzen und die Referenzkurven überein. Das Gleiche gilt für die Ergebnisse der universellen Wärmefunktion basierend auf der „Flux-Gradient Formulation“, wenn der Mindestwert für den kinetischen Wärmefluss signifikant größer ist als für die universelle Impulsfunktion. Ausgangspunkt ist die Veröffentlichung von England & McNider 1995, welche die „Gradient Richardson Number Formulation“ dargestellt haben. Diese enthält zwei fehlerhafte Gleichungen, wovon eine einen Tippfehler aufweist und die andere auf einer fehlerhaften Herleitung basiert. Diese Arbeit stellt daher ebenso die korrigierten Gleichungen dar. Zusätzlich werden neue Gleichungen vorgestellt, bei denen nicht angenommen wird, dass die von England & McNider angenommenen Konstanten der empirischen Funktionen für die universelle Impuls- und Wärmefunktion den gleichen Wert haben. In der Windindustrie ist es üblich, dass die durchschnittliche Nabenhöhen-Windgeschwindigkeit als Durchschnittsgeschwindigkeit für die gesamte „rotor swept area“ angenommen wird. Diese Annahme weicht immer mehr von der Realität ab, weil Windkraftwerke kontinuierlich größer werden. Daher wird in dieser Arbeit auf eine flächengewichtete mittlere Windgeschwindigkeit mit der Bezeichnung Rotor-äquivalente Windgeschwindigkeit gearbeitet. Diese produziert realistischere Durchschnittswerte, weil sie Messungen von mehreren vertikal verteilten Messpunkten einbezieht. Aufgrund des sich verändernden Windgradientens in der Grenzschicht wird die Abweichung zwischen diesen beiden Geschwindig- keiten als Funktion der Schichtungsstabilität dargestellt. Die zentralen Ergebnisse dieser Studie zeigen zusammenfassend, dass bei labiler Schichtung die Nabenhöhen-Windgeschwindigkeit die Rotor-äquivalente Windgeschwindigkeit um etwa 1 % bis 1.5 % unterschätzt. Für die stabile Schichtung unterscheiden sich die Ergebnisse: Zwei Berechnungen zeigen, dass die Nabenhöhen-Windgeschwindigkeit die Rotor-äquivalente Windgeschwindigkeit um ca. 1 % überschätzt und eine andere Berechnung zeigt keinen Unterschied zwischen den Mittelwerten. Daraus kann gefolgert werden, dass die Nabenhöhen-Windgeschwindigkeit eine höhere Fehlerquelle aufweist. Im Gegensatz dazu liefert die rotor-äquivalente Windgeschwindigkeit genauere Prognosewerte.

This chapter examines how evolutionary and institutional economic theory can improve our understanding of the politico-economic dynamics of the green transition. The authors focus on how path dependency in technological systems perpetuate and reward unsustainable business practices, while mission-oriented policy programs can break the feedback loops that keep the trajectory in place. Significantly, policy programs creating sizable markets for renewable energy have generated sufficient learning curves to move renewable energy sources within the cost range of fossil fuels, thereby enabling private profit incentives to accelerate the transition in our immediate future. These gains have depended on managerial decisions to co-evolve their businesses with the institutional environment shaped by policymaking despite uncertainty of technological progress, market evolution, and political credibility. We explore the case of how the Danish energy company Dong Energy was transformed from a fossil fuel company to the world’s leading offshore wind developer. These dynamics point to the need to create markets through political institutions for the coming generations of sustainable technologies to enable potential gains from learning, network effects, and economies of scale to be identified and exploited.

This study uses a unique set of hourly wind speed data observed over a period of 16 years to quantify the potential of collective offshore wind power production. We address the well-known intermittency problem of wind power for five locations along the Norwegian continental shelf. Mitigation of wind power intermittency is investigated using a hypothetical electricity grid. The degree of mitigation is examined by connecting different configurations of the sites. Along with the wind power smoothing effect, we explore the risk probability of the occurrence and duration of wind power shutdown due to too low or high winds. Typical large-scale atmospheric situations resulting in long term shutdown periods are identified. We find that both the wind power variability and the risk of not producing any wind power decrease significantly with an increasing array of connected sites. The risk of no wind power production for a given hour is reduced from the interval 8.0 %–11.2 % for a single site to under 4 % for two sites. Increasing the array size further reduces the risk, but to a lesser extent. The average atmospheric weather pattern resulting in wind speed that is too low (too high) to produce wind power is associated with a high-pressure (low-pressure) system near the production sites.

This paper is a result of the WOW project (Wind power On Wikipedia) which forms part of the SAPIENS (Scientometric Analyses of the Productivity and Impact of Eco-economy of Spain) project (Sanz-Casado et al. in Scientometrics 95(1):197–224, 2013). WOW is designed to observe the relationship between scholarly publications and societal impact or visibility through the mentions of scholarly papers (journal articles, books and conference proceedings papers) in the Wikipedia, English version. We determine (1) the share of scientific papers from a specific set defined by Wind Power research in Web of Science (WoS) 2006–2015 that are included in Wikipedia entries, named data set A; (2) the distribution of scientific papers in Wikipedia entries on Wind Power, named data set B, captured via the three categories for the topic Wind Power in the Wikipedia Portal: Wind Power, Wind turbines and Wind farms; (3) the distributions of document types in the two wiki entry data sets’ reference lists. In parallel the paper aims at designing and test indicators that measure societal impact and R&D properties of the Wikipedia, such as, a wiki reference focus measure; and a density measure of those types in wiki entries. The study is based on Web mining techniques and a developed software that extracts a range of different types of Wikipedia references from the data sets A and B. Findings show that in data set A 25.4% of the wiki references are academic, with a density of 17.62 academic records detected per wiki entry. However, only 0.62% of the original WoS records on Wind Power are also found as wiki references, implying that the direct societal impact through the Wikipedia is extremely small for Wind Power research. In the second Wikipedia set on Wind Power (data set B), the presence of scientific papers is even more insignificant (10.6%; density: 3.08; WoS paper percentage: 0.26%). Notwithstanding, the Wikipedia can be used as a tool informing about the transfer from scholarly publications to popular and non-peer reviewed publications, such as Web pages (news, blogs), popular magazines (science/technology) and research reports. Non-scholarly wiki reference types stand for 74.6% of the wiki references (data set A) and almost 90% in data set B. Interestingly, the few WoS articles in wiki entries on Wind Power present on average 34.3 citations received during the same period (2006–2015) as WoS Wind Power publications not mentioned in wiki entries only receives on average 5.9 citations. Owing to the scarcity of Wind Power research papers in the Wikipedia, it cannot be applied as a direct source in evaluation of Wind Power research. This is in line with other recent studies regarding other subject areas. However, our analysis presents and discusses six supplementary indirect indicators for research evaluation, based on publication types found in the wiki entry reference lists: share of (WoS) records; density; and reference focus, plus popular science knowledge export, non-scholarly knowledge export and academic knowledge export. The same indicators are direct measures of the Wikipedia reference properties. This paper is a result of the WOW project (Wind power On Wikipedia) which forms part of the SAPIENS (Scientometric Analyses of the Productivity and Impact of Eco-economy of Spain) project (Sanz-Casado et al. in Scientometrics 95(1):197–224, 2013). WOW is designed to observe the relationship between scholarly publications and societal impact or visibility through the mentions of scholarly papers (journal articles, books and conference proceedings papers) in the Wikipedia, English version. We determine (1) the share of scientific papers from a specific set defined by Wind Power research in Web of Science (WoS) 2006–2015 that are included in Wikipedia entries, named data set A; (2) the distribution of scientific papers in Wikipedia entries on Wind Power, named data set B, captured via the three categories for the topic Wind Power in the Wikipedia Portal: Wind Power, Wind turbines and Wind farms; (3) the distributions of document types in the two wiki entry data sets’ reference lists. In parallel the paper aims at designing and test indicators that measure societal impact and R&D properties of the Wikipedia, such as, a wiki reference focus measure; and a density measure of those types in wiki entries. The study is based on Web mining techniques and a developed software that extracts a range of different types of Wikipedia references from the data sets A and B. Findings show that in data set A 25.4% of the wiki references are academic, with a density of 17.62 academic records detected per wiki entry. However, only 0.62% of the original WoS records on Wind Power are also found as wiki references, implying that the direct societal impact through the Wikipedia is extremely small for Wind Power research. In the second Wikipedia set on Wind Power (data set B), the presence of scientific papers is even more insignificant (10.6%; density: 3.08; WoS paper percentage: 0.26%). Notwithstanding, the Wikipedia can be used as a tool informing about the transfer from scholarly publications to popular and non-peer reviewed publications, such as Web pages (news, blogs), popular magazines (science/technology) and research reports. Non-scholarly wiki reference types stand for 74.6% of the wiki references (data set A) and almost 90% in data set B. Interestingly, the few WoS articles in wiki entries on Wind Power present on average 34.3 citations received during the same period (2006–2015) as WoS Wind Power publications not mentioned in wiki entries only receives on average 5.9 citations. Owing to the scarcity of Wind Power research papers in the Wikipedia, it cannot be applied as a direct source in evaluation of Wind Power research. This is in line with other recent studies regarding other subject areas. However, our analysis presents and discusses six supplementary indirect indicators for research evaluation, based on publication types found in the wiki entry reference lists: share of (WoS) records; density; and reference focus, plus popular science knowledge export, non-scholarly knowledge export and academic knowledge export. The same indicators are direct measures of the Wikipedia reference properties.

A single pulsed Doppler wind lidar, deployed at the FINO1 platform in the North Sea, has been used to monitor the lateral and vertical coherence of the along-wind component. To maximize the sampling frequency of the monitoring system, a particular configuration based on small sweeping angles around the mean wind direction is used. The set-up provides wind velocity measurement with an increasing cross-flow separation with increasing distances from the lidar. We present hereby preliminary results from the analysis of the Plan Position Indicator scans carried out with a sampling frequency of 0.13 Hz, and the Range Height Indicator scans sampled at 0.19 Hz. The mean velocity and the turbulence intensity indicated a sufficiently uniform flow, which was verified by the calculation of the along and crosswind turbulence length scales. The lateral and vertical coherence was estimated based on multiple 10-minutes samples, for a mean wind velocity ranging from 10 to 14 ms􀀀1. For the conditions examined, the measured coherence showed a good agreement with the IEC model for low and medium cross-flow separations, and a fairly good agreement with the Frøya model for large separations. A single pulsed Doppler wind lidar, deployed at the FINO1 platform in the North Sea, has been used to monitor the lateral and vertical coherence of the along-wind component. To maximize the sampling frequency of the monitoring system, a particular configuration based on small sweeping angles around the mean wind direction is used. The set-up provides wind velocity measurement with an increasing cross-flow separation with increasing distances from the lidar. We present hereby preliminary results from the analysis of the Plan Position Indicator scans carried out with a sampling frequency of 0.13 Hz, and the Range Height Indicator scans sampled at 0.19 Hz. The mean velocity and the turbulence intensity indicated a sufficiently uniform flow, which was verified by the calculation of the along and crosswind turbulence length scales. The lateral and vertical coherence was estimated based on multiple 10-minutes samples, for a mean wind velocity ranging from 10 to 14 m s -1. For the conditions examined, the measured coherence showed a good agreement with the IEC model for low and medium cross-flow separations, and a fairly good agreement with the Frøya model for large separations.

Ocean winds in the Baltic Sea are expected to power many wind farms in the coming years. This study examines satellite Synthetic Aperture Radar (SAR) images from Envisat ASAR for mapping wind resources with high spatial resolution. Around 900 collocated pairs of wind speed from SAR wind maps and from 10 meteorological masts, established specifically for wind energy in the study area, are compared. The statistical results comparing in situ wind speed and SAR-based wind speed show a root mean square error of 1.17 m s−1, bias of −0.25 m s−1, standard deviation of 1.88 m s−1 and correlation coefficient of R2 0.783. Wind directions from a global atmospheric model, interpolated in time and space, are used as input to the geophysical model function CMOD-5 for SAR wind retrieval. Wind directions compared to mast observations show a root mean square error of 6.29° with a bias of 7.75°, standard deviation of 20.11° and R2 of 0.950. The scale and shape parameters, A and k, respectively, from the Weibull probability density function are compared at only one available mast and the results deviate ~2% for A but ~16% for k. Maps of A and k, and wind power density based on more than 1000 satellite images show wind power density values to range from 300 to 800 W m−2 for the 14 existing and 42 planned wind farms.

Increasing levels of wind power penetration in modern power systems has set intensively high standards with respect to wind turbine technology during the last years. Security issues have become rather critical and operation of wind farms as conventional power plants is becoming a necessity as wind turbines replace conventional units on the production side. This article includes a review of the basic control issues regarding the capability of the Doubly Fed Induction Generator (DFIG) wind turbine configuration to fulfill the basic technical requirements set by the system operators and contribute to power system security. An overview of ancillary services provided by wind turbine technology nowadays is provided, i.e. , fault ride-through capability, reactive power supply and frequency-active power control.

We present a study on the characteristics of the sea breeze flow at a coastal site located in the centre of the Mediterranean basin at the southern tip of Italy. This study is finalized to add new data on breeze circulations over a narrow peninsula and present a unique experimental coastal site at about 600 m from the coastline in a flat open area at the foot of a mountain chain located in a region of complex orography. We study the seasonal behaviour of the sea-land breeze circulation by analysing two years of hourly data of wind speed and direction, temperature, radiation and relative humidity from a surface meteorological station, eighteen-months data from a wind profiler, and two-year data from the ECMWF analysis. Results show that breezes dominate the local circulation and play a major role for the local climate. They are modulated by the season, through the sea-land temperature difference and the large-scale flow. The large-scale forcing acts in phase with the diurnal breeze and opposes the nocturnal breeze. In summer, the daytime difference between the land surface temperature and the SST (Sea Surface Temperature) reaches its maximum, while the nigh-time difference has its minimum. This causes a strong, frequent and intense diurnal breeze and a weak nocturnal breeze. In winter and fall the nocturnal difference between the sea and land surface temperature reaches a maximum value, while the diurnal difference is at its minimum value. This causes a strong, frequent and intense nocturnal breeze despite of the large-scale forcing that is usually opposed to local-scale flow.

The goals are to compare lidar volume averaged wind measurement with point measurement reference sensors and to demonstrate the feasibility of performing 3D turbulence measurements with lidars. For that purpose three pulsed lidars were used in staring mode, placed so that their beams crossed close to a 3D sonic anemometer mounted at 78 m above the ground. The results show generally very good correlation between the lidar and the sonic times series, except that the variance of the velocity measured by the lidar is attenuated due to spatial filtering. The amount of attenuation can however be predicted theoretically by use of a spectral tensor model of the atmospheric surface-layer turbulence.

In seeking to advance the understanding of the production, diffusion and negotiation of meaning in and between organisations, we introduce below the notion of mise-en-sens. Mise-en-sens refers to the performing art term of mise-en-sce` ne and builds on the French term sens denoting meaning as well as direction. Starting from a qualitative analysis of how Swedish wind farm developers manage the permit-application process for their projects, we suggest in a first-order analysis that the activities of the developers consist in contextualising the project, ontologising it and neutralising criticism addressed to it. In a second-order analysis, we conceptualise these activities as a mise-en-sens. Miseen- sens spells out the way in which developers, as meaning managers, stage the project and provide it with direction. Having positioned mise-en-sens in relation to the notions of sensemaking and sensegiving, we suggest in conclusion that mise-en-sens could serve to describe not only the activity of wind farm developers but also, for example, that of project managers or entrepreneurs in general since they too are engaged in the management of mean (Less)

PhD thesis in Offshore technology

A NACA0015 airfoil section instrumented with an array of high frequency microphones flush-mounted beneath its surface was measured in the wind tunnel at LM Wind Power in Lunderskov. Various inflow speeds and angles of attack were investigated. In addition, a hot-wire device system was used to measure the velocity profiles and turbulence characteristics in the boundary layer near the trailing edge of the airfoil. The measured boundary layer data are presented in this report and compared with CFD results. A relative good agreement is observed, though a few discrepancies also appear. Comparisons of surface pressure fluctuations spectra are used to analyze and improve trailing edge noise modeling by the so-called TNO model. Finally, a pair of hot-wires were placed on each side of the trailing edge in order to measure the radiated trailing edge noise. However, there is no strong evidence that such noise could be measured in the higher frequency range. Nevertheless, low-frequency noise could be measured and related to the presence of the airfoil but its origin is unclear.

Wind turbines play a major role in the transformation from a fossil fuel based energy production to a more sustainable production of energy. Total-cost-of-ownership is an important parameter when investors decide in which energy technology they should place their capital. Modern wind turbines are controlled by pitching the blades and by controlling the electro-magnetic torque of the generator, thus slowing the rotation of the blades. Improved control of wind turbines, leading to reduced fatigue loads, can be exploited by using less materials in the construction of the wind turbine or by reducing the need for maintenance of the wind turbine. Either way, better total-cost-of-ownership for wind turbine operators can be achieved by improved control of the wind turbines. Wind turbine control can be improved in two ways, by improving the model on which the controller bases its design or by improving the actual control algorithm. Both possibilities have been investigated in this thesis. The level of modeling detail has been expanded as dynamic in ow has been incorporated into the control design model where state-of-the-art controllers usually assume quasi-steady aerodynamics. Floating wind turbines have been suggested as an alternative to ground-fixed wind turbines as they can be placed at water depths usually thought outside the realm of wind turbine placement. The special challenges posed by controlling a floating wind turbine have been addressed in this thesis. Model predictive control (MPC) has been the foundation on which the control algorithms have been build. Three controllers are presented in the thesis. The first is based on four different linear model predictive controllers where appropriate switching conditions determine which controller is active. Constraint handling of actuator states such as pitch angle, pitch rate and pitch acceleration is the primary focus of this controller. The wind turbine is a highly nonlinear plant and a gain scheduling or relinearizing model predictive controller forms the next step to improve performance compared to a linear controller. Finally, a nonlinear model predictive controller has been devised and tested under simplified conditions. At present, the nonlinear model predictive controller is however not expected to be an realistic option for real world application as the computation burden is to heavy to achieve real-time performance. This thesis is comprised of a collection scientific papers dealing with the various topics presented in this summary.

Taylor’s frozen turbulence hypothesis is tested in its applicability for wind turbine applications. In this research full field measurements are performed at a test site for multi-megawatt wind turbines by means of a pulsed LIDAR with a scanning device. The system is installed at the top of the nacelle of a 5MW wind turbine. It provides simultaneous wind speed, with a maximum sampling rate of 5 Hz, at different stations parallel to the mean wind. Measurements in a range between 0.4 and 1.6 rotor diameter are performed following several two and three dimensional trajectories. The spectral characteristics of measurements taken simultaneously at different separation distances are studied. The scanning strategy which maximizes the wavenumber region where results are consistent with Taylor’s hypothesis is assessed. The best results are achieved by a horizontal sliding trajectory with valid wavenumbers up to 0.125 rad/m.

The power curve of a wind turbine is the primary characteristic of the machine as it is the basis of the warranty for it power production. The current IEC standard for power performance measurement only requires the measurement of the wind speed at hub height and the air density to characterise the wind field in front of the turbine. However, with the growing size of the turbine rotors during the last years, the effect of the variations of the wind speed within the swept rotor area, and therefore of the power output, cannot be ignored any longer. Primary effects on the power performance are from the vertical wind shear and the turbulence intensity. The work presented in this thesis consists of the description and the investigation of a simple method to account for the wind speed shear in the power performance measurement. Ignoring this effect was shown to result in a power curve dependant on the shear condition, therefore on the season and the site. It was then proposed to use an equivalent wind speed accounting for the whole speed profile in front of the turbine. The method was first tested with aerodynamic simulations of a multi-megawatt wind turbine which demonstrated the decrease of the scatter in the power curve. A power curve defined in terms of this equivalent wind speed would be less dependant on the shear than the standard power curve. The equivalent wind speed method was then experimentally validated with lidar measurements. Two equivalent wind speed definitions were considered both resulting in the reduction of the scatter in the power curve. As a lidar wind profiler can measure the wind speed at several heights within the rotor span, the wind speed profile is described with more accuracy than with the power law model. The equivalent wind speed derived from measurements, including at least one measurement above hub height, resulted in a smaller scatter in the power curve than the equivalent wind speed derived from profiles extrapolated from measurements at hub height and below only. It is well established that the turbulence intensity also influences the power performance of a wind turbine. Two ways of accounting for the turbulence were tested with the experimental data: an adaptation of the equivalent wind speed so that it also accounts for the turbulence intensity and the combination of the equivalent wind speed accounting for the wind shear only with the turbulence normalising method for turbulence intensity suggested by Albers. The second method was found to be more suitable for normalising the power curve for the turbulence intensity. Using the equivalent wind speed accounting for the wind shear in the power performance measurement was shown to result in a more repeatable power curve than the standard power curve and hence, in a better annual energy production estimation. Furthermore, the decrease of the scatter in the power curve corresponds to a decrease of the category A uncertainty in power, resulting in a smaller uncertainty in estimated AEP.

Results are summarised for the project "Improved design for large wind turbine blades (Phase 4)", partially supported by the Danish Energy Agency under the Ministry of Climate and Energy through the EUDP journal no.: 33033-0267. The aim of the project was to develop new and better design methods for wind turbine blades, so that uncertainties associated with damage and defects can be reduced. The topics that are studied include buckling-driven delamination of flat load-carrying laminates, cracking along interfaces in material joints (fracture mechanical characterisation and modelling), cyclic crack growth with large scale bridging and the use of cohesive laws in finite element programmes for simulating wind turbine blade failure. An overview is given of the methods and the major research results of the project. The implementation of the knowledge in the industry is discussed. Finally, some ideas for future research activities are considered.

The purpose of wind turbines and their predecessors the windmill, is to convert the energy in the wind to usable energy forms. Whereas windmills of the past focused on the conversion of wind power to torque for grinding, pumping and winching, modern wind turbines convert the wind energy into electric power. They do so through incorporation of generators, which convert mechanical torque into electricity. Wind turbines are designed to keep the overall cost per produced Kilo Watt hour as low as possible. One way of improving the performance and lifetime of the wind turbine is through active flow control. Active control is often considered costly but if the lifespan of the components can be increased it could be justifiable. This thesis covers various aspects of ‘smart control’ such as control theory, sensoring, optimization, experiments and numerical modeling.

This thesis includes the results of a PhD study about complex terrain and wind lidars. The study mostly focuses on hilly and forested areas. Lidars have been used in combination with cups, sonics and vanes, to reach the desired vertical measurement heights. Several experiments are performed in complex terrain sites and the measurements are compared with two different flow models; a linearised flow model LINCOM and specialised forest model SCADIS. In respect to the lidar performance in complex terrain, the results showed that horizontal wind speed errors measured by a conically scanning lidar can be of the order of 3-4% in moderately-complex terrain and up to 10% in complex terrain. The findings were based on experiments involving collocated lidars and meteorological masts, together with flow calculations over the same terrains. The lidar performance was also simulated with the commercial software WAsP Engineering 2.0 and was well predicted except for some sectors where the terrain is particularly steep. Subsequently, two experiments were performed in forested areas; where the measurements are recorded at a location deep-in forest and at the forest edge. Both sites were modelled with flow models and the comparison of the measurement data with the flow model outputs showed that the mean wind speed calculated by LINCOM model was only reliable between 1 and 2 tree height (h) above canopy. The SCADIS model reported better correlation with the measurements in forest up to ∼6h. At the forest edge, LINCOM model was used by allocating a slope half-in half out of the forest based on the suggestions of previous studies. The optimum slope angle was reported as 17º. Thus, a suggestion was made to use WAsP Engineering 2.0 for forest edge modelling with known limitations and the applied method. The SCADIS model worked better than the LINCOM model at the forest edge but the model reported closer results to the measurements at upwind than the downwind and this should be noted as a limitation of the model. As the general conclusion of the study, it was stated that the lidars can be used in complex terrain with the known limitations and the support of flow models.

We carried out column model simulations to study particle fluxes and deposition and to evaluate different particle formation mechanisms at a boreal forest site in Finland. We show that kinetic nucleation of sulphuric acid cannot be responsible for new particle formation alone as the simulated vertical profile of particle number concentration does not correspond to observations. Instead organic induced nucleation leads to good agreement confirming the relevance of the aerosol formation mechanism including organic compounds emitted by the biosphere. The simulation of aerosol concentration within the atmospheric boundary layer during nucleation event days shows a highly dynamical picture, where particle formation is coupled with chemistry and turbulent transport. We have demonstrated the suitability of our turbulent mixing scheme in reproducing the most important characteristics of particle dynamics within the boundary layer. Deposition and particle flux simulations show that deposition affects noticeably only the smallest particles in the lowest part of the atmospheric boundary layer.

This is the final report of a Danish research project “Grid fault and design-basis for wind turbines”. The objective of this project has been to assess and analyze the consequences of the new grid connection requirements for the fatigue and ultimate structural loads of wind turbines. The fulfillment of the grid connection requirements poses challenges for the design of both the electrical system and the mechanical structure of wind turbines. The development of wind turbine models and novel control strategies to fulfill the TSO’s requirements are of vital importance in this design. Dynamic models and different fault ride-through control strategies have been developed and assessed in this project for three different wind turbine concepts (active stall wind turbine, variable speed doublyfed induction generator wind turbine, variable speed multipole permanent magnet wind turbine). A computer approach for the quantification of the wind turbines structural loads caused by the fault ride-through grid requirement, has been proposed and exemplified for the case of an active stall wind turbine. This approach relies on the combination of knowledge from complimentary simulation tools, which have expertise in different specialized design areas for wind turbines. In order to quantify the impact of the grid faults and grid requirements fulfillment on wind turbines structural loads and thus on their lifetime, a rainflow and a statistical analysis for fatigue and ultimate structural loads, respectively, have been performed and compared for two cases, i.e. one when the turbine is immediately disconnected from the grid when a grid fault occurs and one when the turbine is equipped with a fault ride-through controller and therefore it is able to remain connected to the grid during the grid fault. Different storm control strategies, that enable variable speed wind turbines to produce power at wind speeds higher than 25m/s and up to 50m/s without substantially increasing the structural loads, have also been proposed and investigated during the project. Statistics in terms of mean value and standard deviation have been analysed and rainflow calculations have been performed to estimate the impact over the lifetime of a variable speed wind turbine.

This report was prepared as documentation for the activities in the Certification Secretariat of Wind Turbines in 2009. The secretariat has performed duties under a contract with the DEA, where the con-tents of the tasks are focused on administration of the Danish Certi-fication Scheme for Wind Turbines. The report describes the current state of deployment of wind tur-bines in Denmark and the latest developments in wind turbine in-dustry. Furthermore, the development of test areas is described. The tasks of the Secretariat has included daily administration, including ensuring the secretariat for the Advisory Committee, implementing the system of maintenance and service, development of new licens-ing rules for small wind turbines, and the approval and reviews of small wind turbines. In addition, the Secretariat and related experts from Risø participated in Danish and international standards com-mittees.

In the context of the increasing application of remote sensing techniques in wind energy, the feasibility of upwind observations via a spinner-mounted wind lidar was tested during the SpinnerEx 2009 experiment. The objective was to install a QinetiQ (Natural Power) ZephIR lidar in the rotating spinner of a MW-sized wind turbine, and investigate the approaching wind fields from this vantage point. Time series of wind speed measurements from the lidar with 50 Hz sampling rate were successfully obtained for approximately 60 days, during the measurement campaign lasting from April to August 2009. In this report, information is given regarding the experimental setup and the lidar’s operation parameters. The geometrical model used for the reconstruction of the scanning pattern of the lidar is described. This model takes into account the lidar’s pointing direction, the spinner axis’s vertical tilt and the wind turbine’s yaw relative to the mean wind speed direction. The data analysis processes are documented. A methodology for the calculation of the yaw misalignment of the wind turbine relative to the wind direction, as a function of various averaging times, is proposed, using the lidar’s instantaneous line-of-sight radial wind speed measurements. Two different setups have been investigated in which the approaching wind field was measured at distances of 0.58 Ø and 1.24 Ø rotor diameters upwind, respectively. For both setups, the instantaneous yaw misalignment of the turbine has been estimated from the lidar measurements. Data from an adjacent meteorological mast as well as data logged within the wind turbine’s control system were used to evaluate the results.

The generic 5 MW NREL wind turbine model is used in Risø’s aeroelastic simulator HAWC2 to investigate 120 different swept blade configurations (forward and backward sweep). Sensitivity for 2 different controllers is considered as well. Backward sweep results in a pitch to feather torsional moment of the blade, effectively reducing blade twist angles under increased loading. This behaviour results in decreased flap-wise fatigue and extreme loads, an increase for edge-wise fatigue loading and status quo or slight decrease in extreme loads (depending on the controller). Tower base and shaft-end bending moments are reduced as well. Forward sweep leads to an increase in angle of attack under loading. For a pitch controlled turbine this leads to an increase in fatigue and extreme loading in all cases. An controller inflicted instability is present for the more extreme forward swept cases. Due to the shape of considered sweep curves, an inherent and significant increase in torsional blade root bending moment is noted. A boomerang shaped sweep curve is proposed to counteract this problematic increased loading. Controller sensitivity shows that adding sweep affects some loadings differently. Power output is reduced for backward sweep since the blade twist is optimized as a rigid structure, ignoring the torsional deformations which for a swept blade can be significant.

Probabilistic design of wind turbines requires definition of the structural elements to be included in the probabilistic basis: e.g., blades, tower, foundation; identification of important failure modes; careful stochastic modeling of the uncertain parameters; recommendations for target reliability levels and recommendation for consideration of system aspects. The uncertainties are characterized as aleatoric (physical uncertainty) or epistemic (statistical, measurement and model uncertainties). Methods for uncertainty modeling consistent with methods for estimating the reliability are described. It is described how uncertainties in wind turbine design related to computational models, statistical data from test specimens, results from a few full-scale tests and from prototype wind turbines can be accounted for using the Maximum Likelihood Method and a Bayesian approach. Assessment of the optimal reliability level by cost-benefit optimization is illustrated by an offshore wind turbine example. Uncertainty modeling is illustrated by an example where physical, statistical and model uncertainties are estimated. Probabilistic design of wind turbines requires definition of the structural elements to be included in the probabilistic basis: e.g., blades, tower, foundation; identification of important failure modes; careful stochastic modeling of the uncertain parameters; recommendations for target reliability levels and recommendation for consideration of system aspects. The uncertainties are characterized as aleatoric (physical uncertainty) or epistemic (statistical, measurement and model uncertainties). Methods for uncertainty modeling consistent with methods for estimating the reliability are described. It is described how uncertainties in wind turbine design related to computational models, statistical data from test specimens, results from a few full-scale tests and from prototype wind turbines can be accounted for using the Maximum Likelihood Method and a Bayesian approach. Assessment of the optimal reliability level by cost-benefit optimization is illustrated by an offshore wind turbine example. Uncertainty modeling is illustrated by an example where physical, statistical and model uncertainties are estimated.

This technical note documents the equations for primal-dual interior-point quadratic programming problem solver used for MPC. The algorithm exploits the special structure of the MPC problem and is able to reduce the computational burden such that the computational burden scales with prediction horizon length in a linear way rather than cubic, which would be the case if the structure was not exploited. It is also shown how models used for design of model-based controllers, e.g. linear quadratic and model predictive, can be linearized both at equilibrium and non-equilibrium points, making the presented exstension of the controller formulation equivalent to that of the extended Kalman filter compared to an ordinary Kalman filter.

This technical note documents the trust-region-based sequential quadratic programming algorithm used in other works by the authors. The algorithm seeks to minimize a convex nonlinear cost function subject to linear inequalty constraints and nonlinear equality constraints.

This report investigates the impact of wind power in large power systems. The motivation for this investigation is the ever-increasing wind energy penetration into the power systems throughout the world. A generic large power system model delivered by the Danish Transmission System Operator EnergiNet.dk is used in the presented study cases.

From August 2008 to August 2009, the project ‘Off-Shore Wind Energy Resource Assessment and Feasibility Study of Off-Shore Wind Farm Development in China’ was carried out by China Meteorological Administration (CMA), which was funded by the EU-China Energy and Environment Programme (EEP). As one part of the project, off-shore wind energy resource in China was assessed with QuikSCAT Satellite data and SAR Satellite Images. In this paper, the results from these two ways were introduced.

Reliability of offshore wind production under extreme wind conditions was investigated in this report. The wind power variability from existing and future large offshore wind farms in Western Denmark were simulated using the Correlated Wind model developed at Risø. The analysis was done for five years, with each year simulated with five random seeds, leading to a total of 25 annual wind power time series for six large offshore wind farms, summing up to a little over 330 wind turbines. Two storm control strategies were used. The analysis involved several aspects inspired from reliability studies. The aspects investigated are storm events occurrences and durations, storm control strategy impact on the capacity factor (lost production), the loss of production (power produced from wind drops below a certain threshold due to high wind speeds and storm controller) and finally, the wind power production ramp rates and reserves requirements.