The capacity credit of wind power

What happens when the wind doesn’t blow? How much of a wind farm’s output can be relied on as “firm” capacity, and how much backup generating capacity is needed? The answer is in a measure called the “capacity credit”.

EIA

In an earlier post, I looked at the capacity factor of wind power. That’s the ratio between a wind farm’s average power output and its maximum or “nameplate” capacity. That ratio is usually between about 20% and 30%. That is, when averaged over a year, a wind farm produces about 20%–30% as much energy as it would if it operated continuously at its maximum power output.

This time around, I’ll look at another key operating parameter for wind power, its “capacity credit”. Whereas the capacity factor is a measure of the average output of a wind farm, the capacity credit is a measure of the worst case minimum output that can be relied on as a part of the total system capacity. The capacity credit is the “firm” capacity of a wind farm that can be counted on as a reliable contribution to the sum of all grid capacity.

Horse Hollow Wind Energy Center, Texas. With 736 MW nameplate capacity, it is the world’s largest wind farm.

Put another way, the capacity credit of a wind farm is the amount by which other generating capacity (such as coal, for example) can be removed from the grid without compromising reliability of supply.

Wind power is not the only energy source that varies with time. Tidal power fluctuates on a 12-hour period, and solar power has a seasonal (and, uh, a day-night) variation, and it also has more random variations caused by cloud cover. In fact all generating technologies have planned maintenance outages, and they have some unplanned outages too. Wind is unusual, however, in the unpredictability of its output. It doesn’t have the fixed periodic variations of tidal or solar. This unpredictability of wind power makes the question of its capacity credit a rather complicated one.

What, then, is the capacity credit of wind power? What is that minimum power capacity that a wind farm can reliably provide?

Since a wind farm’s output can drop all the way to zero, it seems at first sight that the capacity credit of wind power must be zero. In fact that’s not the case. It would be true if the wind farm operated in isolation, but a wind farm is usually connected to a much larger supply grid. Supply and demand across the grid vary all the time, and energy planners have developed detailed statistical calculations to handle this problem.

They plan grid capacity so as to meet a given “loss of load probability”, or LOLP. The LOLP is the probability that generation will be insufficient to meet demand. Energy supply planners must ensure that there is sufficient capacity to keep the loss of load probability below some specified level, but they don’t want to spend money needlessly on surplus capacity beyond that. The issue is one of managing risk:

Wind farms can be treated statistically in exactly the same way as conventional power plant. For any type of power plant it is possible to calculate the probability of it not being able to supply the expected load. As wind is variable, the probability that it will not be available at any particular time is higher.

Can We Rely on the Wind?
BWEA ^[1]

Kentish Flats Offshore Wind Farm, UK. Its nameplate capacity is 90 MW, 5 of the 30 turbines are shown here.       Elsam

Wind power can be factored into the grid reliability statistics in exactly the same way as every other power source. Wind has a lower probability of being available, but that number is simply fed into the calculations. There is nothing qualitatively different about wind. Energy engineers have taken a careful look at the statistics of wind supply, and their conclusion is that wind has a significant capacity credit after all ^[2].

How can this be? After all, the wind speed can drop all the way to zero. To answer that, we have to look at the supply statistics across the entire electricity grid. For example, when wind power is geographically dispersed, it becomes less likely that the wind will stop blowing at all wind farm sites simultaneously. That’s not to say it’s impossible, but it is less likely. Also, when wind strength and electricity demand correlate (for example, in regions where the wind is stronger during the winter) there is again a higher likelihood that wind will contribute to that demand.

Capacity Credit:  The Numbers

Going to the next level of detail, what are the actual numbers for the capacity credit of wind power?

The capacity credit of wind depends on the fraction of total grid capacity that is met by wind power. In the jargon, it depends on the “penetration” of wind power on the grid.

When the amount is wind capacity is a negligible fraction of the total grid capacity, the capacity credit of the wind farm can be treated as being equal to the average power of the wind farm ^[1]. That is, the capacity credit is the same as the capacity factor multiplied by the installed capacity. That’s because at very low levels of wind penetration, the grid can deal with fluctuations in wind output as part of its routine capability.

As wind capacity increases to about 10% of total grid capacity, the capacity credit falls to about 20% of the installed capacity (peak power) of the wind farm. That is, the capacity credit is now lower than the average power of the wind farm.

If still more wind farms are built, so that wind capacity increases to well above 10% of grid capacity, then wind starts to form a very substantial part of total electricity supply. There is now less leeway elsewhere in the system, and the capacity credit falls further still, to about 10% of installed capacity. That is, each 1 GW of installed wind capacity must be treated as only 100 MW of “firm” capacity. Put another way, each 1 GW of installed wind capacity allows 100 MW of conventional (gas or coal) capacity to be removed from the grid, although that wind capacity supplies about 300 MW of power on average (because it still has a 30% capacity factor).

The figures above are for the UK. Figure 1 presents the situation for Germany in greater detail ^[3]:

Figure 1.  Relationship between installed wind capacity and its capacity credit, for Germany.

Graph is from Ref.[3]. The “relative capacity credit” is the capacity credit expressed as a fraction of installed or “nameplate” capacity. The projected capacity credit for 2020 is based on calculations in the DENA 1 study showing the effect on capacity credit of expected improved load factors in Germany, resulting from improved wind power technology (more efficient rotors) and the use of sites with higher wind speeds (offshore).

Germany has 125 GW of generating capacity, of which 20.6 GW, or 16.5%, is wind ^[4]. From Figure 1, Germany’s wind power has an 8% relative capacity credit, which corresponds to a capacity credit of about 1.5 GW.

Backup Requirement

What does the capacity credit of a wind farm tell us about the backup capacity that a wind farm requires? Because a wind farm can’t always be relied on to deliver its average power, it needs an operating reserve to act as back up. Nowadays that operating reserve is usually gas or coal.

On top of that, there is also the question of “spinning reserve”. Some thermal power stations are kept in a “warm” state, with turbines spinning, so they can deliver full power at very short notice. This is the “spinning reserve” on the grid. At the present low level of wind penetration in the UK, no extra “spinning reserve” is needed. Wind power won’t create a need for additional spinning reserve until its share of total grid capacity exceeds 20% ^[1].

Looking further ahead, a more elaborate network of renewable power sources can be set up if we have a large enough grid. Between them the various renewables can provide power reliably, even though each renewable individually has a relatively low probability of being available. (That issue is for another blog post though.)

How much backup capacity is needed?

The backup requirement is usually seen as the amount of capacity needed to make up the difference between a wind farm’s average output and its capacity credit (where the capacity credit is its minimum “firm” output):

Backup capacity needed  =  average power − capacity credit

To illustrate this, here are two examples. The first is for a capacity credit of 20% of “nameplate” installed capacity, as would be the case for low levels of wind penetration. The second example is for a capacity credit of 10% of “nameplate” capacity, the case for a high level of wind penetration. A 30% capacity factor is assumed for both cases:

Backup capacity requirement at 20% capacity credit
Installed capacity (peak power)	1000 MW
Average power at 30% capacity factor	=  1000 MW × 30%  =	300 MW
Capacity credit	=  1000 MW × 20%  =	200 MW
Backup capacity = average power − capacity credit	= 300 MW − 200 MW =	100 MW

In this case, the 1000 MW of installed capacity generates an average power of 300 MW, and requires an additional backup capacity of 100 MW.

Backup capacity requirement at 10% capacity credit
Installed capacity (peak power)	1000 MW
Average power at 30% capacity factor	=  1000 MW × 30%  =	300 MW
Capacity credit	=  1000 MW × 10%  =	100 MW
Backup capacity = average power − capacity credit	= 300 MW − 100 MW =	200 MW

When the capacity credit has fallen to 10% of installed capacity, the 1000 MW of installed capacity still generates the same 300 MW of average wind power, but it now requires 200 MW of backup capacity, more than before, because a lower fraction of the installed capacity can be counted on as “firm” capacity.

Related Posts

• The capacity factor of wind power

References

1. Can We Rely on the Wind?  British Wind Energy Association, 2007  (Webcite cache)
2. Wind Power Has a Capacity Credit: A Catalogue of 50+ Supporting Studies, G. Giebel, e-WindEng Journal, 2005  (Webcite cache)
3. Capacity Credit Values of Wind Power, Wind Energy – The Facts, European Wind Energy Association, 2007  (Webcite cache)
4. Statistical Pocketbook 2009. Part 2:  Energy, European Commission, 2009  (Table 2.4.1)  (Webcite cache)