Search

A control scheme for drag power kites, also known as airborne wind turbines, for the entire wind speed range is proposed, including 1) a temperature controller allowing for temporary overloading of the powertrain; 2) a limitation controller ensuring that power, force, speed, and actuator constraints are satisfied; 3) a tangential flight speed controller; and 4) a tangential force control allocation, which inverts the nonline arities of the plant and allocates the flight speed controller’s tangential force demand to the available actuators. The drag power kite plant model is based on a point-mass model and a simple aerodynamics model with various drag contributions. Simulations are conducted with the parameters of the 20 kW Wing 7 developed by Makani Power, Inc. The proper working of the control scheme is indicated by the good match of the simulation results with independent simulation results and measurements published by Makani. A temporary overloading of the power train with about twice the nominal power can be concluded as a requirement; otherwise the mean power would be significantly lower. Because of the reduction of the lift and thus reduction of the centripetal force at high wind speeds, the inside-down figure eight can be concluded as the best pattern.

This paper has the goal of comparing one and two-sided wake steering, in terms of plant-level power capture and loading on the upstream and downstream turbines. Wind tunnel tests were conducted in a boundary layer wind tunnel in typical offshore conditions, using three scaled controlled and sensorized wind turbines, governed by an open-loop plant controller. Dynamic wind direction changes were obtained by rotating the wind tunnel turntable. Results indicate that one-sided steering achieves less than half the power gain of double-sided steering with respect to the uncontrolled greedy case. Additionally, although loading on the steering turbine is reduced, loading downstream is also significantly higher than for double-sided steering. On the other hand, actuator duty cycle is much reduced for one-sided steering, thanks to a reduced activity around the wind-aligned condition.

This paper compares the accuracy of four different FLORIS model implementations characterized by increasing levels of accounted physics. The predictions of each model were tested against experimental observations obtained from wind tunnel tests of a cluster of three scaled wind turbines. The experiments were conducted for a wide range of environmental conditions, and mainly with yawed turbines. The results showed a significant improvement in the accuracy of the base model by adding more details to the physical effects of the simulation, such as secondary steering and a heterogeneous flow field. It is also interesting to note that the FLORIS predictions become particularly accurate when considering inflow-dependent power losses in yaw-misaligned conditions.

In this study a new yaw mechanism for a small existing wind turbine model is designed and tested. The special requirement for this yaw device is its compact design due to the small length scaling factor of the model turbine. Such a dense design is crucial in order for the wake of the model turbine not to be influenced by the yaw mechanism. Furthermore, it has to have a fast system response and high accuracy to take into account the time scaling. Different concepts are investigated and a design located at the tower base is realized. To limit its drag, the structure is being concealed in an aerodynamically shaped cover. Wind tunnel measurements confirmed that the wake of the turbine is not influenced by the structure. A system analysis, using a Simulink model of the turbine helps finding the right motor and shows that the device has a fast system response with high accuracy. Consequently, the designed yaw mechanism is suited for the small wind turbine model, which can be used for experimental wake and wind farm control studies in the future.

This paper describes a method to improve and correct an engineering wind farm flow model by using operational data. Wind farm models represent an approximation of reality and therefore often lack accuracy and suffer from unmodeled physical effects. It is shown here that, by surgically inserting error terms in the model equations and learning the associated parameters from operational data, the performance of a baseline model can be improved significantly. Compared to a purely data-driven approach, the resulting model encapsulates prior knowledge beyond the one contained in the training data set, which has a number of advantages. To assure a wide applicability of the method – including also to existing assets – learning is here purely driven by standard operational (SCADA) data. The proposed method is demonstrated first using a cluster of three scaled wind turbines operated in a boundary layer wind tunnel. Given that inflow, wakes and operational conditions can be precisely measured in the repeatable and controllable environment of the wind tunnel, this first application serves the purpose of showing that the correct error terms can indeed be identified. Next, the method is applied to a real wind farm situated in a complex terrain environment. Here again learning from operational data is shown to improve the prediction capabilities of the baseline model.

Simple demanded power point tracking (DPPT) control algorithms with a mode switch are proposed and experimentally validated in this study. One is a torque-based control and uses the generator torque with fixed blade-pitch angle. The other is blade-pitch based and uses both the generator torque and blade pitch for DPPT control. Both control algorithms receive power demand from a higherlevel controller and use their control strategies to track it. The two DPPT control algorithms were integrated with the basic torque and pitch control algorithms, and simulations using an in-house code and a high fidelity multi-body dynamic aeroelastic code were performed for steady and dynamic conditions. To verify the algorithms experimentally, wind tunnel tests with a scaled wind turbine having active pitch control capability were performed. From the simulation and the test, the proposed DPPT algorithms integrated into the conventional control algorithm were found to work well with the basic wind turbine power control algorithm. The torquebased control showed better performance in terms of power tracking, but the pitch-based control showed better performance in terms of loading.

This paper combines the research methodologies of scaled wind turbine model experiments in wind tunnels with short-range WindScanner lidar measurement technology. The wind tunnel at the Politecnico di Milano was equipped with three wind turbine models and two short-range WindScanner lidars to demonstrate the benefits of synchronised scanning lidars in such experimental surroundings for the first time. The dual-lidar system can provide fully synchronised trajectory scans with sampling timescales ranging from seconds to minutes. First, staring mode measurements were compared to hot-wire probe measurements commonly used in wind tunnels. This yielded goodness of fit coefficients of 0.969 and 0.902 for the 1 Hz averaged u and v components of the wind speed, respectively, validating the 2-D measurement capability of the lidar scanners. Subsequently, the measurement of wake profiles on a line as well as wake area scans were executed to illustrate the applicability of lidar scanning to the measurement of small-scale wind flow effects. An extensive uncertainty analysis was executed to assess the accuracy of the method. The downsides of lidar with respect to the hot-wire probes are the larger measurement probe volume, which compromises the ability to measure turbulence, and the possible loss of a small part of the measurements due to hard target beam reflection. In contrast, the benefits are the high flexibility in conducting both point measurements and area scanning and the fact that remote sensing techniques do not disturb the flow during measuring. The research campaign revealed a high potential for using short-range synchronised scanning lidars to measure the flow around wind turbines in a wind tunnel and increased the knowledge about the corresponding uncertainties.

While it has long been a practice to place spires near the inlet of a wind tunnel to quickly develop a turbulent boundary layer with similarities to an atmospheric boundary layer, this has not been the case for creating turbulent boundary layer inflow in large eddy simulations (LESs) of turbulent flows. We carry out LES with the curvilinear immersed boundary method to simulate the flow in a wind tunnel with a series of spires in order to investigate the feasibility of numerically developing inflow conditions from a precursory spire LES and assessing the similarities of the turbulence statistics to those of an atmospheric boundary layer. The simulated mean velocity field demonstrates that a turbulent boundary layer with height equal to the spire height develops very quickly, within five spire heights downstream. The major attribute of using spires for precursory simulations is the spatially evolving coherent structures that form downstream of the spires offering a range of length scales at both the...

Dynamic Induction Control (DIC) has the potential of boosting wind farm power by enhancing wake recovery, whereby periodic pitch motions are used to exploit the natural instabilities of wind turbine wakes. This work studies DIC both experimentally and numerically. A thorough validation of an LES-ALM (Actuator Line Method) simulation tool is first conducted against experimental measurements. This shows that the CFD model is able to accurately simulate the power, loads and wake behavior of a wind turbine operating with DIC. The validated CFD model is then employed to study the effects of some DIC parameters. Results indicate an increase in the fatigue loads caused by the pitch activity that enhances wake recovery.

This paper aims at quantifying and explaining the effects of Individual Pitch Control (IPC) on the wake of a wake-steering wind turbine. As the machine is intentionally misaligned with respect to the wind, IPC can be used to mitigate the resulting extra loading. However, while IPC reduces loads, it also affects the wake, which influences the power of downstream turbines. The question is therefore whether the IPC activity has any other appreciable effect at the wind farm level, in addition to its original turbine-level load reduction goal. In this work, experiments and CFD simulations of scaled wind turbines in a boundary layer wind tunnel are considered. The CFD model is first validated with measurements, and then used to show subtle changes in the wake, power and loads caused by IPC. It is observed that IPC does indeed have some non-negligible effects, and that these effects differ for positive and negative yawing.

The impact of wake steering control strategy on power production and loads for a cluster of two wind turbines is analysed in this work. The power and loads data base are generated by means of high resolution simulations using an ALM-LES model along with an aeroelastic code for multibody analysis, while the optimal yaw misalignment values are obtained, for different wind directions and performance requirements, using a weighted cost function that accounts for both power production and fatigue damage on main wind turbine components. The sensitivity of the results regarding the inflow conditions, in terms of turbulence intensity, is also analysed. The overall results show that when accounting for both power production and fatigue damage, greater values of optimal yaw misalignment are obtained. This indicates that standard yaw misalignment strategies that accounts only for power production reduce at the same time the loading damage on the wind turbine, compared to greedy configuration.

In this paper a comparison between the wake velocity, wake deflection and turbulence intensity predictions of several wake models was carried out against wind tunnel data obtained with a state-of-the-art scaled wind turbine model. In order to achieve a fair comparison, the models’ parameters were all tuned with respect to the same experimental dataset using the Maximum Likelihood Estimation (MLE) method. A quantitative assessment of all models’ predictions highlighted that the Porté-Agel model seems to provide, for a wide range of inflow and wind turbine operating conditions, the most accurate estimation of the wake flow field. Further improvements to the model are also suggested in the conclusions.

This paper first describes a large-eddy simulation approach, and then verifies it with respect to single-turbine wind tunnel experiments. Various aspects of the numerical approach are considered, to try to reduce its need for tuning, improve its accuracy and limit its computational cost. Simulation results are compared to measurements, including rotor and wake quantities. The study includes normal operating conditions, as well as wake manipulation by derating, yaw misalignment and cyclic pitching of the blades. Results indicate a good overall matching of simulations with experiments. Low turbulence test cases appear to be more challenging than moderate and high turbulence ones, due to the need for denser grids to limit numerical diffusion and accurately resolve tip-shed vortices in the near wake region.

We present wind tunnel and numerical experiments of the flow behind a model wind turbine in optimal and sub-optimal regimes. Spatio-temporal filtering and dynamic mode decomposition reveal that the onset and amplitude of wake meandering are affected by both the operating regime and the nacelle.

This paper describes the methodology that was followed for designing a scaled wind turbine model. This model is intended to be used in complex terrain studies as well as deep array wind farm control tests. Therefore, emphasis was given on making it as compact as possible while keeping a high level of instrumentation. The result is a three blade wind turbine with rotor diameter of 0.6 m equipped with active pitch, yaw and torque control. After a discussion on the design procedure, we present preliminary performance characteristics of the model obtained experimentally.

The present work describes a method for the structural optimization of wind turbine rotor blades for given prescribed aerodynamic shape. The proposed approach operates at various description levels producing cost-minimizing solutions that satisfy desired design constraints at the finest modeling level. At first, a “coarse”-level constrained design optimization is performed by using a 1D spatial geometrically exact beam model for aero-servo-elastic multibody analysis and load calculation, integrated with a 2D FEM cross sectional model for stress/strain analysis, and the evaluation of the 1D model fully-populated cross sectional stiffness matrices. Next, a “fine”-level 3D FEM model is used for the refinement of the coarse-level solution. Improved results obtained at the level of the 3D model are utilized at the following coarse-level iteration through a heuristic modification of the design constraints. In addition, a buckling analysis is performed at the fine description level, which in turn affects the nonstructural blade mass. The updated constraint bounds and mass make their effects felt at the next coarse-level constrained design optimization, thereby closing the loop between the coarse and fine description levels. The multilevel optimization procedure is implemented in a computer program and it is demonstrated on the design of a multi-MW horizontal axis wind turbine rotor blade.

A method to detect wake impingement on a wind turbine is described. The wake detector, which uses rotor loads as measured by on-board sensors, can then be used for implementing wind farm control strategies, for example in the form of wake redirection. The method is verified and validated by using field test data, aeroservoelastic simulations and by experimental data obtained with scaled wind turbine models tested in a large boundary layer wind tunnel.

A new method to estimate bending moments of a blade specimen having a constant cross section throughout was developed using fiber Bragg grating sensors. Based on theories of solid mechanics, strains of the simplified model of the specimen were evaluated for various pitching angles and the forces exerted. It was found by both theory and experiment that at least two FBG sensors are necessary to find out the magnitude and direction of the load acting on the specimen from measured longitudinal strain information. A method was also found to estimate the flatwise and edgewise bending moments from the measured strains. An experimental setup that enables pitching the specimen from 0 ° to 180 ° was used to experimentally verify the theoretical findings. The theory and the experimental results showed a fair agreement.

The synergistic use of passive (by bend-twist coupling) and active (by individual blade pitch control) load mitigation is considered in this work. It is found that while both contribute to load reduction, they have opposite effects on the actuator duty cycle, and this induces an additional unexpected pay-off from the combination of these two technologies.

This paper presents results from wind tunnel tests aimed at evaluating a closed- loop wind farm controller for wind farm power maximization by wake deflection. Experiments are conducted in a large boundary layer wind tunnel, using three servo-actuated and sensorized wind turbine scaled models. First, we characterize the impact on steady-state power output of wake deflection, achieved by yawing the upstream wind turbines. Next, we illustrate the capability of the proposed wind farm controller to dynamically driving the upstream wind turbines to the optimal yaw misalignment setting.

This paper combines the currently relevant research methodologies of scaled wind turbine model experiments in wind tunnels with remote-sensing short-range WindScanner Lidar measurement technology. The wind tunnel of the Politecnico di Milano was equipped with three wind turbine models and two short-range WindScanner Lidars to demonstrate the benefits of synchronised scanning Lidars in such experimental surroundings for the first time. The dual- Lidar system can provide fully synchronised trajectory scans with sampling time scales ranging from seconds to minutes. First, staring mode measurements were compared to hot wire probe measurements commonly used in wind tunnels. This yielded goodness of fit coefficients of 0.969 and 0.902 for the 1 Hz averaged u- and v-components of the wind speed, respectively, validating the 2D measurement capability of the Lidar scanners. Subsequently, the measurement of wake profiles on a line as well as wake area scans were executed to illustrate the applicability of Lidar scanning to measuring small scale wind flow effects. The downsides of Lidar with respect to the hot wire probes are the larger measurement probe volume and the loss of some measurements due to moving blades. In contrast, the benefits are the high flexibility in conducting both point measurements and area scanning, and the fact that remote sensing techniques do not disturb the flow while measuring. The research campaign revealed a high potential for using short-range WindScanner Lidar for accurately measuring small scale flow structures in a wind tunnel.

The paper describes a simple method for detecting presence and location of a wake affecting a downstream wind turbine operating in a wind power plant. First, the local wind speed and shear experienced by the wind turbine are estimated by the use of rotor loads and other standard wind turbine response data. Then, a simple wake deficit model is used to determine the lateral position of the wake with respect to the affected rotor. The method is verified in a boundary layer wind tunnel using two instrumented scaled wind turbine models, demonstrating its effectiveness.

In this study, the effect of active induction control of upstream wind turbines is investigated. Two scaled wind turbines having a rotor diameter of 1 m with a spacing of four times of the rotor diameter were used to experimentally validate the concept. Also, an in-house c code was used to simulate the same two wind turbines and see if the experimental observations can be obtained. From the experiment, approximately 0.81% increase of total power could be observed. Although the simulation results were not exactly the same as the experimental results but the shape was similar and the maximum power increase of 0.27% was predicted. Also from further simulation using NREL 5MW wind turbines instead of scaled wind turbines with realistic ambient turbulence intensity, it was found that the power increase could become more than 1%.

The paper describes a wind tunnel scaled wind farm made of two wind turbine models, each featuring active blade pitch and torque control. The models, governed by supervision and control systems similar to those of a real wind turbine, are capable of simulating steady conditions and transient maneuvers in the boundary layer test section of the wind tunnel of the Politecnico di Milano. Expanding the classical scope of wind tunnel models, the present experimental facility enables the testing of wind farm controllers, as well as the study of interaction phenomena within wind farms.