Site assessment for a type certification icing class

Introduction

Icing climate has recently received much attention thanks to commonly favorable wind conditions and mostly low populated areas. On the other hand, the structural safety and reliability of wind turbines can be affected as well as the economic efficiency.
Present international guidelines and standards focus on the structural safety under normal climatic conditions, but consider icing effects to a limited extend.
The international research project IcedBlades has started to improve the reliability of wind turbines in icing climate conditions. The results, as the aerodynamic and structural modelling of blade icing with respect to wind turbine loading, have already been included in the new edition of IEC 61400-1 (under development). However, especially the site assessment for icing conditions is described only to a very marginal extend in all guidelines and standards. Site assessment is essential to judge whether a specific site showing icing conditions requires turbines with icing class type certification or not.


Approach

A proposal for a type certification icing class with focus on load calculations has been given in [1], where the blade icing duration (instrumental icing) is the main parameter. The present publication describes the site assessment for icing conditions in order to connect the site to the type certification icing class. Different approaches are described and evaluated for the example site “Rödeser Berg”. The reference site of Fraunhofer IWES is located in an icing relevant area in the middle of Germany where a 200m meteorological mast is installed. The mast is equipped with more than 40 meteorological and wind sensors that provide detailed information about atmospheric conditions.

Main body of abstract

The following approaches are used to estimate the icing duration:
1. Short term direct ice measurements using ice sensors installed on the met mast, as e.g. with ice collector, as well as short term non-direct ice measurements, e.g. with double anemometry technique using heated and unheated wind anemometers at several heights between 60m and 200m.
2. Meteorological measurements from the mast (e.g. air temperature, air humidity, wind data) to determine the meteorological icing and instrumental icing duration.
3. Alternative approaches are discussed as well, e.g. the application of icing atlases, numerical approaches or use of SCADA data of already existing wind farms.
The above listed approaches 1 and 2 will be applied to one example site “Rödeser Berg”, while the different approaches according to 3 are described in general. The approaches are compared; requirements for the necessary data acquisition and its quality are listed.


Conclusion

The following conclusions can be drawn:
- Icing durations can differ significantly between different winters; a long-term correction is required.
- Height dependent icing profiles have been examined, showing different icing probabilities at different heights.
- Different sensors for direct ice measurements (approach 1) can lead to different results; the ice collector shows high failure sensitivity; a recommended setup is the use of double anemometry with one heated ultrasonic anemometer and one unheated cup anemometer.
- The application of approach 2 (meteorological measurements of temperature, wind and relative humidity) shows a lower uncertainty due to reliable sensor techniques. But it results in a higher overall uncertainty due to the application of uncertain models to derive the instrumental icing based on meteorological icing.
- Based on approach 2 some meteorological boundary conditions can be given for instrumental icing at the investigated site “Rödeser Berg”: The lower limit for relative humidity is 85%; above 100% relative humidity (saturation) AND temperature below +3°C a very high probability exists; highest probabilities were observed between +2°C and -5°C, extreme low probabilities below -10°C.
- Estimations for icing probabilities based on SCADA data and on icing atlases are possible.



Learning objectives
Different methods are described to determine the instrumental icing period. Some methods are evaluated for an icing test site in central Germany. These methods can be used as a recipe to estimate turbine icing duration and finally judging whether a specific site requires turbines with icing class type certification or not.