Here are some brief considerations about the performance of German wind energy facilities. Over the past two decades the number of wind turbines installed in Germany has increased dramatically. The surge in capacity amounted to 350 % between 2000 and 2010. The growing capacity lead in turn to a growing production of electricity, 300 % up over the same period.
One question of particular interest to us was how much energy one might expect to be generated per MW installed on average. As the production of this kind of energy is highly sensitive to the availability of wind we were seeking for a model which is able to reflect the variations of the latter. The source data for our investigation were taken from both Eurostat and Bundesverband WindEnergie, the German association of wind energy producers.
We started by looking at the amount of energy produced per installed capacity (Prod/cap) measured in MWh/MW between 2001 and 2010. We may call this quantity the specific production. The result is the blue curve in the figure below showing a clearly increasing trend. It must be noted here that this graph needs to be interpreted with special care, since it is distorted by two major sources of uncertainty. The first source is the varying wind availability. The second source of uncertainty is based on the fact that the number of wind mills is constantly growing over the year as more and more turbines are constructed. The total installed capacity may go up by some 30 % and more during the course of a year. Some of the newly built turbines may take up operation in spring and others may be commissioned towards the end of the year. In each case their contribution to the entire production will be very different. Thus our model has to account for the extra capacity erected during one particular year.
Another factor coming into play is the location of the turbines. Although each of them is supposedly optimized in terms of output, there may be significant differences between various wind farms. However, as more and more turbines come into existence, the influence of individual outliers should diminish compared to the total average. We always have the big picture in mind, thus neglecting the performance of particular locations.
In theory, the specific production should exactly correspond to the amount of wind available. The latter quantitiy is represented by the red curve as a percentage over the long-term average (r.h.s. scale). As can be seen, the variations may be tremendous.
We note that the red and the blue curve do not coincide as expected. This lack of coincidence is mainly caused by the addition of extra capacity which, in relative terms, was very large during the first part of the period in question (44 % and 37 % in 2001 and 2002, respectively, compared to the previous year).
In order to find a better agreement between the wind availablity curve and the specific production we developed a statistical model which enabled us to eliminate the distortions caused by the newly built wind mills. The result of our model calculations is shown in the green curve which nicely matches with the availability curve (red). What we got is a new quantity Prod/cap* which allows us to draw meaningful conclusions about the mean productivity of each MW of installed capacity.
Depending on the availability of wind Prod/cap* may vary considerably. At a value of 100 % availability each MW installed should produce slightly more than 1700 MWh annually. During our reference period Prod/cap* varied between a maximum of 1800 MWh and a minimum of 1250 MWh. This is a massive variation which must be taken into account when considering the energy supply stemming from wind farms.