Spontaneous balancing of wind power in Western Europe: origin and limits

Introduction

“Wind is always blowing, somewhere”. Although it seldom refers to quantized information, the statement sounds like solid common sense. It might even be true as far as the fuzziness associated with “somewhere” could apply to any terrestrial area as large as one might conceive. Now recently, this actually vague and consequently pointless sentence has been promoted to the rank of first principle by such official organizations as ADEME (French national) or the European Union (international). “Spontaneous wind power balancing” is thus endowed with the virtue of efficiently smoothing the electric production from wind mills over an entire country or a continent. Unfortunately, a major drawback of wind power, i.e. its intermittency, is thus underestimated. Actually, high level wind power is poorly predictable[1] and hardly manageable as shown nowadays in Germany. According to the natural balancing “theory”, an erratic production in a given region would be compensated by the production of another region whose erraticism would undergo opposite fluctuations. Adding productions would result in a smoother behaviour. Such a smoothing would be badly needed since another major disadvantage of wind power is the extreme sensitivity to wind speed. Indeed, on a national scale, wind power may vary from 3% to 70% of the rated capacity within a few hours.

Targeting the general public, the natural balancing concept is aimed at building up a positive image of wind power, both onshore and offshore. Actually it justifies ambitious and costly plans aiming at extending the high voltage grid, clearing borders over the whole of Western Europe. An assessment of its validity is therefore useful. Surprisingly, the same organizations that invoke balancing in order to support huge grid extension projects did not publish so far any analysis of its amplitude nor of its efficiency at curing the specific drawbacks of wind power production.

The present paper is intended at providing the missing analysis[2]. In order to do so, data describing wind power production for year 2012 will be used. They refer either to onshore and offshore wind farms in Denmark or to the cumulated wind power from six countries of Western Europe[3]. In section II amplitudes of balancing effects if any are reviewed. A closer look is devoted to the transition in production that occurs considering larger and larger geographical areas. The next step, section III, is a modelling of the data using purely stochastic methods. It is then shown through both observation and simulation that wind power production is random however large is the size of the considered European area. Furthermore the subsequent balancing appears limited.

The complete document (English version - pdf) is downloadable through this link.

[1]         This is confirmed comparing 24h wind forecast with effective data for France as given on the site of the national grid manager RTE. For Germany see the site of the electricity exchange EEX.

[2]         A similar study dealing with a 6 months period prior to year 2012 was proposed on the site of « Sauvons le climat » : intermittence et foisonnement

[3]         Data analysed in the present document are extracted from the site of P.F. Bach, former technical manager of the Danish national power company. Mr Bach has collected these informations during several years from different European networks. Since whenever they are open to the public, the data are cast into different formats, Mr Bach put them on his site in a single standard format. Surprisingly, there is no such achievement at the European Commission level.