Intermittent Renewables and Electricity Markets
- Aug 14, 2013 2:00 am GMT
- 1777 views
The fact that wind energy still requires generous subsidies after having reached grid-parity with conventional power is a very direct real-world example of the added costs of intermittency even at low penetration levels. This article will explore the market mechanisms responsible for this often ignored added cost.
In principle, electricity markets are fairly straightforward. Offers made by sellers are arranged in ascending order, bids made by buyers are arranged in descending order and the market clearing price and capacity is determined by the point where the two lines cross. In this way, no sellers are forced to sell power below the price they offered and no buyer has to buy power at a higher price than they bid.
When working with dispatchable generating capacity, the offers being put forward will mostly start with baseload coal or nuclear plants which are optimized to run at very high efficiencies and capacity factors. Load-following coal and gas plants will then bid in at a somewhat higher price because they have to work at lower capacity factors and efficiencies. And finally, peaking plants capable of rapid ramping will bid in at very high prices in case of very high demand.
Since it is very difficult to store electricity, this arrangement is necessary to closely match supply with demand as it varies over various timescales ranging from minutes to months. An example of how this arrangement will impact electricity prices is shown below for three different times of day. It is clear that electricity prices will be low in the early morning hours because very little of the more expensive load-following capacity is required. In the afternoon peak, however, electricity prices can increase substantially as more expensive supply is brought online.
The demand lines are drawn diagonally in order to visualize the limited effect that price can have on demand. For example, if prices fall for whichever reason (i.e. the blue curve shifts downwards) more capacity will be bought at this lower price perhaps due to some energy-intensive industries upping production.
Practical difficulties posed by intermittent sources
This system of matching supply and demand operates very well for dispatchable sources, but things become a lot trickier when intermittent sources are included. Because of the practical challenges discussed in this section, intermittent renewables generally enjoy dispatch priority – a priority of selling electricity to the grid regardless of supply/demand fundamentals. Without such legislation, intermittent energy surges might not be accepted even if they made competitively low price offers.
The most important practical problem arises when intermittent renewable energy surges enjoying priority dispatch displace the demand for baseload power. Due to the potentially very large renewable energy spikes, this can already happen at fairly low penetrations. For example, German wind power shown below achieved a capacity factor of about 18% in 2012, but intermittent spikes regularly exceeded 70% of capacity – quadruple the mean output. As a result of such intermittent spikes occurring during times of low demand (e.g. at 04:00 in the graph shown above), wind will regularly eat into baseload demand even at penetration levels as low as 5%.
Cutting supply from a baseload plant to compensate for an hour or two of very high wind generation is not practically possible. This means that both the wind and baseload capacity must be cleared by the market through very large price reductions. In recent years, this has at times resulted in negative electricity prices.
In addition, renewable energy fluctuations often happen quite rapidly, thereby requiring very rapid ramping of load-following plants. The long neck of California’s “duck graph” (the time when solar PV supply falls at the same time as demand picks up towards the afternoon/evening peak) offers a good example of this issue. In practice, this situation will require a much higher percentage of the expensive peaking plants capable of such rapid ramping.
The effect of intermittency on price
Free market systems are very adept at correctly valuing commodities. It is therefore no coincidence that the electricity price is low (sometimes negative) during times of high production of intermittent renewable energy. As shown below, this relationship between prices in red (€/MWh) and wind energy output in blue (MWh) is clearly visible for German wind power even when the data is smoothed to daily averages.
Naturally, this significantly hurts the business case for intermittent renewables simply because wind and solar farm operators must sell the bulk of their product at below-average prices. From this point of view, it is only natural that wind power will continue requiring substantial subsidies even though grid parity has already been reached many years ago.
At this point, solar advocates normally point out that solar PV production is much better correlated with demand and therefore deserves to be sold at a premium. Yes, this is true for the first few percentage points of penetration, but, due to the very pronounced intermittency (low capacity factor) of solar PV, it is not long before the same problem is encountered. As an example, the projected electricity supply profile for Germany in the year 2020 is shown below with solar PV at 10% penetration.
It is clear that solar PV supply surges will eat into baseline supply and force very deep and rapid ramping of load-following plants on many days during the summer. Under such circumstances, Germany will have to pay its neighbours to take this unwanted solar power off her hands, thereby severely hurting the business case for solar PV and forcing unnatural shifts in economic activity in the core of the already fragile European economy.
Added costs of the transition from baseload to load-following
This incompatibility between baseload capacity such as nuclear and intermittent renewables such as wind and solar is part of the reason why Germany is retiring her nuclear fleet and building more flexible coal plants. Naturally, this is a tremendously expensive endeavour and struggling German utility companies are now claiming €15 billion in damages. Without this compensation, German utilities will not be able to meet the great challenges posed by rapidly fluctuating loads such as the example shown above and the Energiewende will fail. It can therefore be expected that utility bailouts will contribute significantly to the rapidly rising costs of the Energiewende in coming years (the numbers below are in billions of Euros).
As this brief analysis has shown, intermittent renewables will sell at below-average prices even at relatively low levels of penetration, implying that these technologies will require generous subsidies even after grid parity is reached. This effect has shown itself in real-world markets even at fairly low penetration levels and will escalate rapidly as more intermittent capacity is added.
In addition, substantial added costs can be expected from the premature transition from a dispatchable power fleet predominantly running baseload plants to one running only load-following plants. Baseload plants must be retired early, new load-following plants must be constructed and the overall capacity factor and efficiency of dispatchable power generation will drop substantially.