Delegates are invited to meet and discuss with the poster presenters during the poster presentation sessions between 10:30-11:30 on Thursday, 29 September 2016.

Régis Decoret 3E, Belgium
Co-authors:
Nicolaz Guidon (1) ^F Baris Adiloglu (1)
(1) 3E, Brussels, Belgium

Benefits of doing MCP per stability class

Introduction

Long-term extrapolation methodologies rely on the assumption that MCP relationships between a site and a reference remain constant over time.
But in reality, these vary as a function of atmospheric stability at the site and reference.
This is an issue when the predicted long-term time series needs to represent the actual wind speed variations over time, such as when it is used to assess the monthly production profile of a wind farm, or curtailment losses that apply at given times of the year or day.

Approach

3E adapted MCP methods (linear regression and matrix), so that relationships between site and reference wind speeds are established per stability class, or as a function of the month and time of day.
Such adaptations are intended to make the generated long-term time series more representative of the actual wind speed variations over time.
3E then compared monthly production profiles and curtailment losses obtained on a number of test cases, using these adapted MCP methods versus standard ones.

Main body of abstract

3E will demonstrate that the long-term wind speed variations over time can be better reproduced using advanced MCP methods, where relationships between the site and reference are established per stability class, or as a function of the month and time of day

Conclusion

The proposed advanced MCP methods allows a wind farm's monthly production profile, or curtailment losses that apply at given times of the year or day, to be better evaluated.


Learning objectives
Delegates will learn that standard MCP methods, although generally suitable for AEP calculations, do not produce a long-term time series that reproduces the actual wind speed variations over time.
This is an issue when the predicted long-term time series needs to represent the actual wind speed variations over time, such as when it is used to assess the monthly production profile of a wind farm, or curtailment losses that apply at given times of the year or day.
In these cases, It is advisable to use advanced MCP methods, where relationships between the site and reference are established per stability class, or as a function of the month and time of day.