Conference programme

Back to the programme printer.gif Print




Delegates are invited to meet and discuss with the poster presenters during the poster presentation sessions between 10:30-11:30 and 16:00-17:00 on Thursday, 19 November 2015.

Lead Session Chair:
Stephan Barth, ForWind - Center for Wind Energy Research, Germany
Clement Hochart DNV GL, France
Co-authors:
Clement Hochart (1) F
(1) DNV GL, Paris, France (2) DNV GL, Bristol, United Kingdom

Share this poster on:

Printer friendly version: printer.gif Print

Presenter's biography

Biographies are supplied directly by presenters at EWEA 2015 and are published here unedited

Clement has been working in the wind energy industry for 9 years. He is currently a senior consultant engineer at DNV GL. Clément studied mechanical engineering at Ecole Nationale Supérieur des Arts et Metiers (ENSAM) in Lille and then specialised in wind energy at Université du Québec à Rimouski (MSc at UQAR). Over the last 5 years Clément has been responsible within DNV GL France for delivering analysis services for operational wind farms to wind farm owners, banks or investors. In this role Clément has been responsible for the detailed operational monitoring or performance assessment of over 1GW of wind farms.


Poster

Poster Download poster (9.39 MB)

Abstract

Energy Production Assessments in France - How accurate are they, which learnings can be drawn, and how accurate can they get?

Introduction

A robust and sustainable wind energy industry requires a predictable return on investment, which can only be achieved when wind farms perform as expected. DNV GL has been predicting the energy production of wind farms for nearly two decades with more than 100 GW of wind power plants assessed internationally, many of which being now in operation for considerable periods. In order to assess the accuracy of its predictions DNV GL maintains an internal database which allows the actual production of wind farms to be compared with pre-construction projections. This paper presents the most recent results of the validation exercise undertaken by DNV GL in France and is part of a larger validation program undertaken worldwide. Publications are already available for several countries in Europe and North America.

Approach

To overcome issues associated with different periods of data being available from the various wind farms included in the validation database, each year of actual production data has been considered separately and compared against the DNV GL net energy pre-construction central estimate (P50).

The DNV GL validation database for France currently includes 49 wind farms and covers operational years ranging from 2006 to 2014, breaking down to a total of 120 individual wind farm operational years for which both utility production and operational information (at least monthly availability statistics) are available.


Main body of abstract

The distribution of actual annual energy productions relative to DNV GL central estimates for the 120 operational wind farm years in the database has been derived. This raw distribution presents an initial over-prediction bias in the order of 10%.

Further investigations have led to the following results:

• The difference of windiness between the long term reference period retained in pre-construction and each operational year under review explains the main part of the initial bias. This was somehow expected due to the particularly low levels of winds observed in north of France in years 2010 to 2014. South of France has a different wind regime and is less affected by this issue.

• A few wind farms have experienced abnormal operational issues in first years of operation. Other wind farms have also shown low availability or important environmental losses (icing for example) not always appropriately estimated in old DNV GL pre-construction wind studies. All combined these availability and performance issues explain approximately 2% of the initial bias.

• Taking into account the corrections detailed above for (1) windiness and (2) availability and performance, there remains a bias in the order of 1% likely to be due to inaccuracies in the pre-construction assessments or to the validation exercise itself.


Conclusion

In conclusion, the particularly low wind speed levels observed in the north part of France since 2010 have led to several years of wind farm production below pre-construction estimates. While windiness has been confirmed to be the main cause of under-production in the validation exercise, the low availability levels historically observed at some wind farms or the inaccuracies associated to the pre-construction energy assessments methods, may have also slightly widen the gap between pre-construction estimates and actual production.


Learning objectives
As it is currently impossible to predict the level of wind expected in the future (especially whether or not wind speed will increase in the north part of France to reach levels similar to those observed in the past 10 to 15 years) DNV GL considers that the current best solution for wind farm owners to mitigate this windiness risk is to diverse wind farms locations in their portfolios.

Recent pre-construction wind studies which generally benefit from good measurement campaigns compared to old studies (better masts locations, higher hub heights, and longer periods of measurements) have been observed to be relatively accurate. DNV GL therefore highly recommends developers to carefully prepare and manage wind resource measurement campaign at their wind farms.