Matthias Kinzel

Energy exchange in an array of vertical-axis wind turbines

We analyze the flow field within an array of 18 counter-rotating, vertical-axis wind turbines (VAWTs), with an emphasis on the fluxes of mean and turbulence kinetic energy. The turbine wakes and the recovery of the mean wind speed between the turbine rows are derived from measurements of the velocity field using a portable meteorological tower with seven, vertically-staggered, three-component ultrasonic anemometers. The data provide insight to the blockage effect of both the individual turbine pairs within the array and the turbine array as a whole. The horizontal and planform kinetic energy fluxes into the turbine array are analyzed, and various models for the roughness length of the turbine array are compared. A high planform kinetic energy flux is measured for the VAWT array, which facilitates rapid flow recovery in the wake region behind the turbine pairs. Flow velocities return to 95% of the upwind value within six rotor diameters downwind from each turbine pair. This is less than half the recovery distance behind a typical horizontal-axis wind turbine (HAWT). The observed high level of the planform kinetic energy flux is correlated with higher relative roughness lengths for the VAWT array as compared to HAWT farms. This result is especially relevant for large wind farms with horizontal dimensions comparable to the height of the atmospheric boundary layer. As shown in recent work and confirmed here, the planform kinetic energy flux can be the dominant source of energy in such large-scale wind farms.

Fluid–Structure Interaction Modeling of Vertical-Axis Wind Turbines

Full-scale, 3D, time-dependent aerodynamics and fluid‐structure interaction (FSI) simulations of a Darrieus-type vertical-axis wind turbine (VAWT) are presented. A structural model of the Windspire VAWT (Windspire energy, http://www.windspireenergy.com/ )i s developed, which makes use of the recently proposed rotation-free Kirchhoff‐Love shell and beam/cable formulations. A moving-domain finite-element-based ALE-VMS (arbitrary Lagrangian‐Eulerian-variational-multiscale) formulation is employed for the aerodynamics in combination with the sliding-interface formulation to handle the VAWT mechanical components in relative motion. The sliding-interface formulation is augmented to handle nonstationary cylindrical sliding interfaces, which are needed for the FSI modeling of VAWTs. The computational results presented show good agreement with the field-test data. Additionally, several scenarios are considered to investigate the transient VAWT response and the issues related to self-starting. [DOI: 10.1115/1.4027466]

Low-order modeling of wind farm aerodynamics using leaky Rankine bodies

We develop and characterize a low-order model of the mean flow through an array of vertical-axis wind turbines (VAWTs), consisting of a uniform flow and pairs of potential sources and sinks to represent each VAWT. The source and sink in each pair are of unequal strength, thereby forming a “leaky Rankine body” (LRB). In contrast to a classical Rankine body, which forms closed streamlines around a bluff body in potential flow, the LRB streamlines have a qualitatively similar appearance to a separated bluff body wake; hence, the LRB concept is used presently to model the VAWT wake. The relative strengths of the source and sink are determined from first principles analysis of an actuator disk model of the VAWTs. The LRB model is compared with field measurements of various VAWT array configurations measured over a 3-yr campaign. It is found that the LRB model correctly predicts the ranking of array performances to within statistical certainty. Furthermore, by using the LRB model to predict the flow around two-turbine and three-turbine arrays, we show that there are two competing fluid dynamic mechanisms that contribute to the overall array performance: turbine blockage, which locally accelerates the flow; and turbine wake formation, which locally decelerates the flow as energy is extracted. A key advantage of the LRB model is that optimal turbine array configurations can be found with significantly less computational expense than higher fidelity numerical simulations of the flow and much more rapidly than in experiments.

Turbulence in vertical axis wind turbine canopies

Experimental results from three different full scale arrays of vertical-axis wind turbines (VAWTs) under natural wind conditions are presented. The wind velocities throughout the turbine arrays are measured using a portable meteorological tower with seven, vertically staggered, three-component ultrasonic anemometers. The power output of each turbine is recorded simultaneously. The comparison between the horizontal and vertical energy transport for the different turbine array sizes shows the importance of vertical transport for large array configurations. Quadrant-hole analysis is employed to gain a better understanding of the vertical energy transport at the top of the VAWT arrays. The results show a striking similarity between the flows in the VAWT arrays and the adjustment region of canopies. Namely, an increase in ejections and sweeps and decrease in inward and outward interactions occur inside the turbine array. Ejections are the strongest contributor, which is in agreement with the literature on evolving and sparse canopy flows. The influence of the turbine array size on the power output of the downstream turbines is examined by comparing a streamwise row of four single turbines with square arrays of nine turbine pairs. The results suggest that a new boundary layer forms on top of the larger turbine arrays as the flow adjusts to the new roughness length. This increases the turbulent energy transport over the whole planform area of the turbine array. By contrast, for the four single turbines, the vertical energy transport due to turbulent fluctuations is only increased in the near wake of the turbines. These findings add to the knowledge of energy transport in turbine arrays and therefore the optimization of the turbine spacing in wind farms.

Short Length-Scale Rotating Stall Inception in a Transonic Axial Compressor: Experimental Investigation

To improve the understanding of spike-type stall inception of a transonic axial compressor, measurements of the unsteady static pressure in the rotor endwall region are analyzed. At design speed, a detailed experimental investigation of the unsteadiness of the pressure field at the rotor endwall at near stall condition shows a strong fluctuation of the tip clearance flow. Both vortex strength and -trajectory oscillate randomly. Analysis of the wall pressure time histories during stall inception suggests that spike-type disturbances of the flow field correlate with an upstream motion of one blade passages shock front. In addition, the evolution of a stall cell could be visualized by means of static wall-pressure contour plots.Copyright

Performance enhancement of downstream vertical-axis wind turbines

Increased power production is observed in downstream vertical-axis wind turbines (VAWTs) when positioned offset from the wake of upstream turbines. This effect is found to exist in both laboratory and field environments with pairs of co- and counter-rotating turbines, respectively. It is hypothesized that the observed production enhancement is due to flow acceleration adjacent to the upstream turbine due to bluff body blockage, which would increase the incident freestream velocity on appropriately positioned downstream turbines. A low-order model combining potential flow and actuator disk theory captures this effect. Additional laboratory and field experiments further validate the predictive capabilities of the model. Finally, an evolutionary algorithm reveals patterns in optimized VAWT arrays with various numbers of turbines. A “truss-shaped” array is identified as a promising configuration to optimize energy extraction in VAWT wind farms by maximizing the performance enhancement of downstream turbines.

Measurement of Lagrangian acceleration in turbulent flows using the laser Doppler technique

The use of the laser Doppler technique for measuring Lagrangian acceleration with a high spatial resolution is introduced. The requirements in system alignment and accuracy of signal processing is estimated. Specifications of the optical design and the signal processing are given.

Increasing the Power Production of Vertical-Axis Wind-Turbine Farms Using Synergistic Clustering

Vertical-axis wind turbines (VAWTs) are being reconsidered as a complementary technology to the more widely used horizontal-axis wind turbines (HAWTs) due to their unique suitability for offshore deployments. In addition, field experiments have confirmed that vertical-axis wind turbines can interact synergistically to enhance the total power production when placed in close proximity. Here, we use an actuator line model in a large-eddy simulation to test novel VAWT farm configurations that exploit these synergistic interactions. We first design clusters with three turbines each that preserve the omni-directionality of vertical-axis wind turbines, and optimize the distance between the clustered turbines. We then configure farms based on clusters, rather than individual turbines. The simulations confirm that vertical-axis wind turbines have a positive influence on each other when packed in well-designed clusters: such configurations increase the power generation of a single turbine by about 10 percent. In addition, the cluster designs allow for closer turbine spacing resulting in about three times the number of turbines for a given land area compared to conventional configurations. Therefore, both the turbine and wind-farm efficiencies are improved, leading to a significant increase in the density of power production per unit land area.

Measurement of Particle Accelerations with the Laser Doppler Technique

An extension of the laser Doppler technique for measuring particle acceleration is presented. The basic principles of the technique follow closely those introduced in [11], although numerous improvements have been implemented in the signal processing for increasing the reliability of individual estimates of particle acceleration. The main contribution of this study is to identify and quantify the errors due to optical fringe divergence in the detection volume of the present laser Doppler system, to introduce an appropriate experiment involving a falling wire and to compare the acceleration measurements of the laser Doppler system to the results of a particle tracking system with high-speed cameras in a highly turbulent flow. Noteworthy is the fact that all measurements were performed with a commercial off-the-shelf laser Doppler system.

Two Scale Experiments via Particle Tracking Velocimetry: A Feasibility Study

In preparation of simultaneous large-scale / small-scale 3D Particle Tracking Velocimetry (3D-PTV) experiments in a developing turbulent flow we performed two types of measurements separately: (i) the velocity and coarse-grained velocity derivatives were measured in a large observation volume with focus on the large-scale flow features and (ii) spatially resolved velocity derivatives were measured in a small observation volume with the goal to obtain small-scale quantities associated with vorticity and strain. In this contribution we demonstrate that the characteristic flow structures were captured and velocity derivatives were accessed with sufficient accuracy. The problem of measuring velocity derivatives both in the Lagrangian and Eulerian frame of reference is also addressed. Although comparable accuracies in both settings could be achieved with our method, only statistics obtained from the spatially resolved measurement were found to be practically the same in both settings.