Community Research and Development Information Service - CORDIS


AVATAR Report Summary

Project ID: 608396
Funded under: FP7-ENERGY
Country: Netherlands

Periodic Report Summary 2 - AVATAR (AdVanced Aerodynamic Tools for lArge Rotors)

Project Context and Objectives:
The focus in the AVATAR project lies on the aerodynamic modelling of large wind turbines with a rated power of 10 MW or more (denoted as 10MW+ turbines). The application for such large scale turbines is mainly thought to be off-shore. Because of the significant costs for installation, support structure and grid connection for off-shore wind energy, the cost share of the turbine hardware as percentage of the total investment is roughly half the percentage value that applies to an on-shore turbine which then pleads for large turbines. Moreover, increasing the ratio between rotor diameter and installed generator power, i.e. a lower specific power, corresponds to a higher capacity factor leading to more operating hours in full power. This reduces the variability in wind power and allows more effective use of the power transport cables, which is a major advantage for utilities. Besides these there are other CoE drivers to grow rotor area like the plain economies of scale and increasing energy capture per foundation.
The design of the resulting very large rotor blades is subject to several challenges. These can, at least partly, be overcome by new blade technologies, e.g. active and passive blade control.

The problem is though that rotor designs of such large scale largely fall outside the validated range of current state-of-the-art aerodynamic and aero-elastic tools in various aspects: Very large blades operating at high tip speeds means high Reynolds and Mach numbers for which the effects are uncertain and not enough validated; thick(er) airfoils need to be assessed in terms of aerodynamic performance; increased flexibility will lead to larger deflections and more pronounced non-linear aeroelastic behavior with unknown aerodynamic implications, etc. Further complications enter by the desired implementation of active and/or passive flow devices.
The aim of AVATAR is thereby ‘to deliver aerodynamic and aero-elastic models and tools for a more validated and higher fidelity design modelling of 10MW+ scale wind turbines’.
The AVATAR project has four scientific/technical work packages apart from a communication and dissemination package (WP5) and a coordination Work Package 6:
• WP1: Integration and evaluation 10MW rotor
• WP2: Advanced aerodynamic modelling
• WP3: Models for flow devices and flow control
• WP4: Aero-elastic analysis of large and flexible blades

The specific aim of WP1 is to demonstrate and assess the new generation of aerodynamic design tools for large offshore wind turbines The model demonstration and assessment is carried out on two 10 MW reference turbines (RWT’s), one from the adjacent INNWIND.EU project and one designed in AVATAR. The latter uses the INNWIND.EU RWT as a starting point but it is intended to be more challenging in terms of aerodynamic modelling, i.e. aspects like airfoil thicknesses, Reynolds and Mach numbers etc. are pushed toward the limit of what is still feasible to expect in future commercial applications

WP2 deals with the advanced aerodynamic modelling of all aspects which are expected to play a role in the design of large 10MW+ wind turbine blades. The specific aim of WP2 is to provide cross-validated and reliable aerodynamic tools capable of predicting the aerodynamics of the large scale wind turbine of the future (10-20MW), based on code to code comparison, comparison with wind tunnel data and existing full scale data.
The focus of WP3 lies on the modelling of flow devices. Its specific aim is to generate reliable simulation models and software tools to include flow control concepts (mainly LE/TE flaps and vortex generators, but also root spoilers on large wind turbine blades
The focus of WP4 lies on aero-elastic effects of large and flexible rotor blades. Its aims are:
a. To validate advanced aero-elastic simulations against existing measurements
b. To evaluate the confidence level of the various aero-elastic tools on the AVATAR RWT 10MW (reference wind turbine) in terms of stability, extreme and fatigue loads, using the tools from WP2 and 3 with and without flow devices.
c. To perform a parametric analysis of the effect of solidity, rotational speed and structural tailoring in view of redesigning the AVATAR RWT 10MW rotor and further up-scaling it to 20MW rotor

Project Results:
In the reporting period the project was under full steam and a large number of results have been obtained in all work packages. The focal point lied on the Work Packages 2, 3 and 4. In line with the planning, the main activity of WP1 took place in the first half year of the project and led to the delivery of the AVATAR reference wind turbine as an aerodynamic challenging variant on the INNWIND.EU reference wind turbine. After the delivery of this turbine WP1 progressed on a slower pace but it will become active again in the final year with the evaluation of models. Thereto, within WP1 an inventory was made of the model improvements from the WPs 2 to 4. The deliverable does not only report model improvements but also strategies to reduce calculation time and time to set-up calculations and lessons learned on the proper application of the models.

In WP2 the aerodynamic modelling of large wind turbines is considered without the effects of flow devices and aero-elasticity). In this WP comparisons have been made between results from various aerodynamic models of different complexity on a sectional and rotor level also including turbulent inflow. In terms of airfoil flow modelling, an additional effort was needed to understand the unexpected initial large spread between the results of models which in their fundamentals were largely similar. The improved understanding on model details and associated implementations reduced the spread considerably. The lessons learned have been reported in guidelines for model applications/implementations.

For the first time simulations were carried out regarding the interaction between inflow turbulence and wind rotors in high detail. Results are promising.

Moreover several wind tunnel measurements have been carried out, e.g. measurements on a DU00-W-212 airfoil placed in the industrial wind tunnel from LM and on the same airfoil placed in the Forwind wind tunnel where controlled turbulent inflow conditions could be achieved. This airfoil was also measured in the pressurized DNW-HDG wind tunnel at high Reynolds up to 15 M, i.e. conditions which represent the actual conditions on wind turbine blades. The quality of these measurements was assessed by a cross-check with the LM measurements at moderate Reynolds numbers showing an excellent agreement. The good quality of these measurements then offered a unique opportunity to use these measurements for a ‘blind comparison’ i.e. a model benchmark carried out without knowledge of the measurements which also attracted participants outside AVATAR. The results of this blind test were published in a side event at the EWEA 2015 conferences. An analysis from this blind test was included in the above mentioned lessons learned.

In WP3, models have been developed and assessed for flow devices (mainly vortex generators and leading/trailing edge flaps on airfoils and blades) under static and dynamic conditions, based on an extensive CFD and experimental database. With regard to modelling accuracy it was concluded (very roughly speaking) that flow devices do not add much uncertainty to the already existing uncertainty in the modelling of clean airfoils although this is less true for the modelling of drag from vortex generators. The models have been used to assess several flow device design parameters and eventually the control strategy is studied for the TE flaps. Several recommendations for design purposes have been formulated.

Also WP4 has produced a large amount of results. Thereto calculations on aeroelastic and aeroelastic response have been performed at extreme yaw and shear, in half wake operation and in standstill, also in turbulent conditions where fatigue loads were recorded. The activities include clean blades as well blades equipped with VGs or flaps. Results at standstill showed the sensitivity of the response to the precise airfoil data and to the dynamic stall modelling. Depending on the airfoil data at post-stall conditions and/or the dynamic stall model (which ranged from engineering up to CFD), either a stable or unstable situation were detected. The results have led to several suggestions for model improvement which in the remaining part of the project will be implemented. Very helpful in this respect are the so-called canonical cases, i.e. cases which are very carefully described and which are of a stylistic nature. Thereto some very basic steady axi-symmetric cases are calculated first with prescribed rpm and pitch angle after which complexity (eg. yaw, shear etc) is added one by one so that the modelling effect of all these parameters can be clearly distinguished.

Potential Impact:
As mentioned above several model improvements and lessons learned have been gained in the project. As such aerodynamic and aero-elastic models and tools of a more validated character and of higher fidelity are provided for the design of 10MW+ scale wind turbines.
The wider socio-economic impact and the wider societal implications of the project are the reduced cost of energy for off-shore wind energy through the use of large scale 10MW+ wind turbines with or without flow devices. Understanding of typical aerodynamic phenomena and associated modeling is an important and vital stepping stone for the high fidelity design of such turbines. Offshore wind power is expected to deliver a major new resource of electricity for the European grid.

However, also the data sharing and knowledge exchange amongst scientific and industrial partners is crucial. The open access publication in the project allows all interested parties to follow new results easily. An important example in this respect is the mentioned public blind test. Moreover the deliverables from the technical work packages are all stored on the public AVATAR web site. The measurement data are uploaded to a Web Based validation platform, i.e. the WindBench platform as used by the IRPWIND project too.

List of Websites:

Reported by

Follow us on: RSS Facebook Twitter YouTube Managed by the EU Publications Office Top