the first phase of Mexnext started on June 1st 2008 and lasted for 3 years until June 1st 2011. The project was divided into 5 Work Packages where the Work Packages 3 and 4 were subdivided in tasks.

WP1: Processing/presentation of data, uncertainties.

The Mexico experiment produced 100 Gbyte of data. After the Mexico project ended the data were stored on external hard discs and distributed between the Mexico participants. On the disc, the time series (of pressures and strain gauges) are given in raw form. Furthermore some processed data are stored in ASCII format. The disc also contains a data point overview and the basic software with which the raw time series can be processed. The PIV data are given in processed form (i.e. ASCII data of the velocity vectors at different x-y-z positions). Within this WP the data will be made available to all Task participants, including those which were not involved in the Mexico project. Thereto the content of the external hard disc will be stored on a password protected Internet site. This Internet site will be developed in WP0 (see below).

In principle the data are organised in a self explanatory way but it will be investigated whether some further processing, explanations, corrections and descriptions are needed. Furthermore an uncertainty analysis will be performed in the form of consistency checks and an investigation of the reproducibility of data. The WP also includes an assessment of the blade manufacturing (note that the actual blade shape has been measured. This shape will be compared with the specified geometry. Under reservation, the 3D Mexico blades will be mounted into the DUT-LST wind tunnel to validate the pressure distributions at non-rotating conditions. These pressure distributions will then be compared with the pressure distributions as measured in standstill conditions during the Mexico experiment (task 4.1). This experiment is offered under the condition that funding can be organised.
The descriptions and the reprocessed data will be added to the (confidential part of the) Internet site.

WP2: Analysis of tunnel effects

This WP is an extension of the tasks on tunnel effects which were already performed in the Mexico project. In the Mexico project, tunnel effects were studied using CFD and simple engineering models. The complexity of the configuration (open tunnel, slits at the collector) may make it necessary to apply advanced CFD models. These codes will then be used to study tunnel effects at different conditions (ranging from low C_DAX to high C_DAX, including yaw misalignment). Supporting information on tunnel effects can also be obtained from 8 pressures, which were measured with taps in the collector entrance. These pressures measure the speedup in the outer flow (outside the wake) needed for the mass conservation of the tunnel flow.

WP3: Comparison of calculational results from different types of codes with Mexico measurement data.

In this WP the calculational results from the codes which are used by the participants are compared with the data from the Mexico experiment. It is meant to be a thorough validation of different codes and it provides insights into the phenomena which need further investigation (see WP4).

The following quantities will be compared:

• Pressure data
• Normal force coefficients
• Tangential force coefficients
• Balance data
• PIV data

 Note that the data to be compared depend on the type of code (e.g. a BEM code cannot produce pressure data).

The measurement campaigns to be reproduced will be agreed upon in the beginning of the project but the campaigns during which PIV mapping was carried out (i.e. axi-symmetric campaigns at V_tun =10, 15 and 24 m/s and yawed campaigns at V_tun = 15 m/s), will anyhow be calculated.  The comparison will mainly be based on time averaged data, but some interesting time traces at high angles of attack will also be considered.

 The WP is divided in 5 tasks.  

Task 3.1: Preparation

In this task an inventory will be made of the codes in the Task. Furthermore a selection will be made of the measurement campaigns to be simulated. A description will be made of the input and the calculation program. Results from task 3.2 (2D airfoil data) will be used, where corrections due to 3D effects will be added.

Finally the comparison procedure will be defined.

Task 3.2 2D airfoil data

In this task the 2D airfoil data will be analysed. Three different airfoils were used in the design of the Mexico rotor (The DU91-W2-250, RISOE-A1-21 and NACA 64-418 airfoils). Within the Mexico project, the 2D pressure distributions of the DU91-W2-250 and NACA-64-418 airfoil have been measured at the appropriate Reynolds number and tripped conditions. These measurements will be compared with results from airfoil and CFD codes. LM-Glasfiber has offered to measure the aerodynamic characteristics of the RISOE-A1-21 airfoil in the LM-Glasfiber wind tunnel

Task 3.3: Calculations

In this task the calculations, as defined in task 3.1, will be performed.

Task 3.4 Comparison between calculations and measurements

In this task the comparison between calculations and measurements is carried out. The task will be based on the comparison procedure as defined in task 3.1

Task 3.5: Evaluation and refinement of WP4 description

In this task the comparison between calculations and measurements is evaluated. On basis of this evaluation, subjects may be established which need special attention (as an example the measurements at high axial force coefficient may be reproduced poorly. It will then be decided to pay special attention to the turbulent wake state). On basis of the evaluation it may be necessary to refine the description of WP4.

WP4: Deeper investigation into phenomena

In this Work package a deeper investigation of different phenomena will take place.
The phenomena will be investigated with isolated submodels, simple analytical tools or by physical rules. As an example, the velocity difference due to the passage of the blades is measured in a fixed sheet in the rotor plane. It can be compared with the difference which is calculated from the bound vorticity which in turn is determined from the measured pressure distribution.

Task 4.1: Standstill

Parked conditions are amongst the design driving load conditions (storm loads). Within the Mexico project, pressure distributions have been carried out at standstill and a large number of pitch angles. The resulting loads will be analyzed and compared with common design standards. It is furthermore expected that these data form a reference for the rotational measurements in order to distinguish rotational effects (see task 4.6).

It is noted that the measurements are carried out at the highest possible tunnel speed. Nevertheless, the resulting pressures are still in the lower end of the measurement range.

Task 4.2: Sensitivity of results on Reynolds number

A number of measurement campaigns have been performed at two rotational speeds where the remaining conditions are more or less similar. As such a comparison of the results at the two rotor speeds may offer insight on the Reynolds number sensitivities of the various data.

Task 4.3: Angle of attack

For many aerodynamic investigations, the angle of attack is a crucial quantity. However the definition, determination and measurement of the angle of attack for a rotating wind turbine is far from straightforward (see e.g. [4]).

Task 4.4: Near wake aerodynamics, including tip vortex trajectories and the turbulent wake state

An important feature of the Mexico project are the PIV measurements which map the flow field around the rotor. Different types of PIV traverses have been made: Radial traverses near the rotor plane at different blade azimuth angles, axial traverses (i.e. traverses at 1 azimuth angle and 2 radial positions between 1D upstream of the rotor plane until 1D downstream of the rotor plane) and tip vortex tracking. As such these measurements provide direct information on the near wake, and the tip vortex trajectory. The near wake is an important flow region since it determines the inflow in the rotor. The near wake also affects the far wake which is the relevant flow region when assessing wind farm effects. 

Within this task several activities will be carried out.

• One of the activities will consider the axial PIV traverse (i.e. the u,v,w velocity components as function of the axial coordinate). The results of these measurements will be compared with the results obtained from different types of models (eg BEM from vortex wake methods, CFD methods, Note that this activity has some overlap with WP3).
  A distinction needs to be made between results from 1) methods in which the flow field is calculated from prescribed rotor loads (prescribed either by means of airfoil characteristics or by means of the axial force measured with the DNW balance) and 2) methods which calculate both the rotor and the flow field.
  Some limited analyses have already been carried out on this subject. It was found that the axial traverse at one radial position behaves strange, possibly as a result of strong shed vorticity. This strange behaviour obviously deserves special attention.
• The tip vortex trajectory and the stability of the tip vortices at different conditions will be investigated.
• PIV mapping has been carried out simultaneously with the measurements of pressures and loads. As such the measured tip vortex strength can be matched with the blade loads, utilising pressure distributions and bending moment data.
• PIV measurements show an up-flow just after blade passage. This up-flow is caused by the viscous wake behind the blade. It will be investigated to what extent viscous profile drag values can be estimated from these data.
• Attention will be paid to the turbulent wake state. This state occurs at high tip speed ratios and it cannot be modelled with standard BEM theory. Within the Mexico project an axial traverse has been made at a tip speed ratio of 10. This campaign still needs to be studied in detail but a first investigation shows a very strong deceleration in the wake where the velocities calculated with common turbulent wake models deviate from the measured values.
  
  

Task 4.5: Non-uniformity of flow between the blades

In this task the non-uniformity of the flow between the blades will be studied, explained and evaluated with common theory.

The task makes use of the radial PIV traverses in the rotor plane. As a matter of fact PIV sheets were located just upstream and just downstream of the rotor plane with  a slight overlap in the rotorplane.

The radial traverses are performed at different tunnel speeds and at 6 blade azimuth positions. This feature allows to map the flow between the blades, and verify the effects of non-uniformity of the flow in the tip region, which is usually corrected for by Prandtl’s tip correction model in Blade Element Momentum (BEM) methods.

Task 4.6: 3D effects

In this task the 3D effects on the rotating airfoil data are being analysed. Sectional data as measured on the 3D blade are compared with 2D airfoil data in order to distinguish the rotational effects on the airfoil data. A comparison with standstill data will also be made.

Task 4.7: Yawed flow

Measurements have been taken at several yaw angles: 15, 30 and 45 degrees.  Both pressure measurements and PIV data have been carried out, although the latter are only taken at a yaw angle of +/- 30 degrees and a tunnel speed of 15 m/s. In this task special attention will be paid to the PIV measurements. Due to the homogeneous wind tunnel conditions, the measurements at opposite yaw angles (i.e. at +/- 30 degrees) can be considered as the measurement at one yaw angle (eg + 30 degrees) and two opposite azimuth angles (note that the PIV sheet was always located the ‘9 o’clock position’). Radial traverses just upstream and just downstream of the rotor plane have been made at different blade azimuth angles. These measurements are expected to give insight on the non-uniformity of the flow between the blades at yaw. Furthermore the variation of velocities between the two opposite azimuth positions will be studied to find the so-called skewed wake effects on the induced velocities in the rotor plane.

Task 4.8: Instationary airfoil aerodynamics

The Mexico measurements are effectively sampled with a high rate of 5 kHz.

Some limited analysis on instationary effects on the pressure distributions have already been carried out. The pressure distributions at non-yawed conditions are generally found to behave very stable, i.e. the variation in pressure distribution over the time is limited. The exception lies in the measurement at the highest tunnel speed of 30 m/s which, even at non-yawed conditions and at the low turbulence level of the tunnel, shows a very instable pressure distribution. Furthermore instationary effects on the airfoil aerodynamics appear at yawed flow. As such, the task has some overlap with task 4.7. However task 4.7 mainly focuses on PIV data at yawed flow (and moderate angles of attack) where the present task focuses on the pressure measurements at high angles of attack.
In this task existing models for instationary airfoil aerodynamics will be compared with the Mexico measurements and if necessary the models will be adjusted.

Task 4.9: Dynamic Inflow

Measurements have been taken at fast pitching steps and rotor speed steps. In such cases the wake behind the rotor and subsequently the induction responds with a certain delay (dynamic inflow). These dynamic inflow effects lead to an overshoot in the load response. A first investigation on dynamic inflow effects showed clear dynamic inflow effects in the pressure data (note that no PIV measurements have been done at dynamic inflow conditions). Within task 4.9 the load transients will be analysed and the associated time constant will be determined for the different radial positions.

 

WP5: Comparison with results from other (mainly NASA-Ames) measurements.

The results from WP3 and WP4 are expected to provide many insights on the accuracy of different codes and their underlying sub-models. Within WP5 it will be investigated whether these findings are consistent with results from other detailed aerodynamic measurements. In particular, one can think of the data as provided within IEA Wind Task XX  by NREL (i.e. the NASA-Ames experiment).Within the NASA-Ames experiment the emphasis lied on the measurement of pressure distributions at different conditions (stall, yaw, dynamic inflow). As such the information is complementary to the activities which are carried out in the tasks 4.6 to 4.9. Apart from the IEA Wind Task XX measurements, data from IEA Wind Task XVIII can be used as reference material. Within this Task, aerodynamic measurement have been taken in the field. They turned out to be useful for validating models on time averaged aerodynamic effects, in particular 3D effects at stalled conditions. Furthermore, detailed near wake measurements have been taken on a 1.2 m diameter rotor by DUT. These measurements have been taken at non-yawed and yawed flow. Part of these measurements will also be brought into the Task and compared with the PIV measurements from the Mexico project.

The workplan for the Mexnext extension will be posted shortly. The extension has been approved until June 1st 2014.