maanantai 31. elokuuta 2015

Windrush economics

"You miss select one technical parameter
in your subsidy system, and you get a wind rush"

This short essay discusses how economical incentive creates a wind rush. Any rush to invest on something happens when the return of equity is high. Wind rush is induced when a subsidy system has a very good chance to make extra profits. This essay explains which technical parameter creates the incentive, and what it means for return of equity. We use Finland as an example, however the discussion can probably be extended to many countries in the world that has had a wind rush in the recent years.

Case Finland 2005 - 2015

Finland had about 80 MW of installed wind power capacity in 2005 (Technical research center of Finland). Ten years later in 2015 there were planned wind power projects that would reach 11 000 MW (Tuulivoimayhdistys) which is 14 000% increase (Figure 1). What on earth happend?

Figure 1: Wind power projects in preparation vs Subsidy quota

In Finland the feed-in tariff subsidy system was taken into use for renewable energy sources in 2011. The subsidy system included a fixed price for produced wind power: before 2016 the rate was 105,3 €/MWh and after that 83,5 €/MWh. The feed-in tariff for one wind turbine plant was 12 years.

The Scandinavian electricity production mix is depicted in Figure 2. Finland had 37% of renewable energy production and 70% of carbon dioxide free electricity production in 2013. The renewable subsidy system target was to achieve 6% of wind power production corresponding to 6 TWh by 2020.

Figure 2: Scandinavia electricity production 2012 (Source: Energiateollisuus)

The feed-in tariff was calculated using a simple formula that has a couple of financial parameters and one technical parameter describing wind power production capability. The subsidy system was approved by the EU comission in 2010. The reader can check the details from EU Comission letter to the Minister for Foreign Affairs, Alexander Stubb (EU Comission SA.31107). In this letter the Finnish authorities promise that they will recalculate the tariff if cost level changes.


Economical incentive hidden to annual operating hours (AOH)

In short the motivation for any investor is the return of equity (ROI, wikipedia). In the Finnish wind power tariff system the ROI was set to 10% (EU Comission SA.31107). So basically if you invest one million, you get 10% of profit each year. That is pretty good profit when the Finnish state is backing up the equity.

How did the Finnish authorities calculate that 10%? The most important technical parameter is the annual operating hours (AOH), which means how many hours per year does the wind turbine produce electricity in maximum power. In Finland the annual operating hours was set to 2400 when calculating the tariff. In one year we have 24 x 365 = 8760 hours. This means during the decision phase Finland thought that wind power produces 2400/8760 = 27% of the maximum theoretical capacity.

What if wind power produces more? What if the the turbine runs 2800 hours per year? What if it runs 3200 hours per year? What happens to the profit? This is depicted in Figure 3 (Download XLS). As you can see, although the rise in annual operating hours from 2400 to 3600 hours does not seem so significant, it has a dramatic effect on how much money the investor gets. If the wind power plant products 2800 hours electricity the investor gets 21% profit where as the authorities thought they get 10%. But as you can see from Figure 3 it gets even better: if your turbine runs 3200 hours you get 31% profit each year. If the turbine runs 3600 hours you get almost 50%. These would be very
 extreme profits.

Figure 3Annual operating hours and return of equity in feed-in tariff 
system Finland (only variable AOH)

Figure 4: Annual operating hours and return of equity in feed-in tariff
system Finland (external interest rate 5%->3%, investment cost 1,4 k€/kW -> 1,2 k€/kW)

And the essential question is: how long do the turbines run on average? It is not a million dollar question now but it is multi billion euro question. This is because the total capacity that can be fitted into the Finnish feed-in tariff system was 2500 MVA that approximately converts into 2100 MW. If we can get the turbine running over 3000 hours the investors get extremely rich.

Not very many countries have a long history of running wind turbines. Denmark is the country that has probably the longest history with wind turbines and a public register containing all wind turbine installations dating back to 1970 (Energi styrelsen). And it gets better: Denmark has the produced electricity on a yearly basis in the same register. This means we can calculate how much electricity (annual operating hours) the turbines have produced.

The Danish register has around 5400 operating wind turbines. The new turbines installed in Finland are large. Currently the typical size of the turbine is 3.3 MW but there are also 5 MW installations (Energiavirasto). In the Danish register most of the turbines are small compared to the new industrial scale wind turbines that are installed into Finland. Let's select industrial scale turbines that are over 2 MW maximum from the Danish register: we only get less than 400 turbines. This is depicted in Figure 5: the average operating hours are on the vertical axis and the maximum power on the horizontal axis:

Figure 5: Average operating hours for wind turbines > 2MW Denmark H1/2015

Now the average operating hours is 3529 hours in Figure 4. And as you can see, there are not many turbines that are below 3000 hour line. There are even some turbines running 5000 hours. You can argue that half year 2015 is not representative time period. You can check 2014 and it basically produces similar results. So on average the large turbines in Denmark run around 3500 hours per year. But on average all turbines run less than that - the average is on 2600 hour level. 

This means the industrial wind turbines run almost 1000 hours more than turbines on average. The exact level of hours cannot be perhaps compared from a country to another as wind resources are different, however you can use this differential: the large turbines produce more electricity.

What happened in Finland? The large new Vestas, Nordex and Siemens wind turbines produce around 3200 hours of electricity in period of latest 18 months of data (Energiavirasto). The best wind farms have reached the average Danish level, so over around 3500 hours. The new turbines are larger in terms of rotor diameter and hub height. The current turbines in construction phase have commonly over 140 meter hub height and around 140 meter rotor diameter. It seems that the average annual working hours are increasing as diameter of the rotor is increasing.

I would assume the same parameter, average operating hours, is included in all of the worlds feed-in tariff systems. Probably in all renewable subsidy systems in the world. And I assume that it is always set too low if a wind rush has occurred in that country. The subsidy system should not include such fixed parameters but the level should be adjusted based on the development of the technology.