When I read in the newspaper that more than half of the EU’s total coal-fired power stations are losing money, I get the feeling that we are doing something right. Air pollution and climate change policies have been pushing coal-fired electricity stations down the drain, and Europe is increasingly setting stricter rules and higher carbon prices as part of its efforts to tackle climate change.
Partly, the reason for the coal power bankruptcy has been the changing economy of renewables. Particularly, wind and solar power are rapidly earning a significant share in electricity markets around the world. This brings several benefits for the environment, but also poses new challenges. Renewables require change in the energy supply chain, as well as in our current energy infrastructure and institutions.
The question now is not whether variable renewable energy (VRE) is viable, but how far it can go?
Electricity is difficult to store. To satisfy our needs, grid operators around the world have developed throughout history accurate models of the total electric loads (i.e. how much energy will be consumed on a given day) in order to generate that energy at the exact moment it is demanded. But as utilities began to produce more energy from renewable sources, the models have started to shift as well.
For many decades, demand for electricity has always followed a fairly predictable daily course, allowing utility grid managers to become experts at predicting and satisfying it. If plotted in a graph, a typical load curve would look something like this:
Source: U.S. Information Administration based on data from Independent System Operator in New England
(This curve is based on an autumn day in New England. Certainly, the curve changes slightly across regions and seasons. In some times and places the humps are more pronounced; in temperate climates, with less heating and cooling demand, they are slightly flatter. But for the sake of simplifying the question we will stick to this example.)
As shown in the graph, demand spikes in early morning and evening, when we consume most energy. However, something different happens with solar energy. It is not possible to schedule solar energy like it is to schedule a power plant, since it just comes and goes with the sun. For grid operators, this means that when the sun is out and a customer’s solar panel is generating energy, this customer is using less of the energy put on the grid by the utility. As more solar photovoltaic (PV) is integrated into the grid, the net load (= total energy – renewable energy) falls during midday and later swoops back up when the sun goes down. This results in the following curve:
However, the problem arises because the increase in supply of renewable energy added to the grid does not increase the actual demand for electricity. This excess of real-time demand is called oversupply. If the market is unable to manage oversupply automatically, this can lead to overgeneration, which requires the manual intervention of the market to maintain reliability. Although in almost all cases oversupply is manageable, overtime is not a sustainable condition – that’s the risk.
The Duck Curve
If we plot the oversupply that is expected to occur during midday, the curve will show steep ramping needs and overgeneration risk. The result? An adorable duck-shaped curve:
Source: California Independent System Operator (not the “quack quack” though)
This curve, named the “Duck Curve” by the California Independent System Operator Corporation, refers to the effect that solar power has on demand for utility electricity, and is one of the greatest challenges facing renewable energy. Unfortunately, as cute as this graph may look to me, the duck curve is a problem. As it can be seen, the potential for oversupply becomes more pronounced over the years as more renewable energy is added to the grid, but demand for electricity does not increase.
Grid operators must continuously balance supply and demand, and thus adapt their plants and utilities to the new curve. This forces them to switch power plants offline and online rapidly every time the sun goes up and down. And this sudden switching off and on, apart from being highly polluting, makes voltage and frequency management much more difficult.
The overgeneration brings along an economic problem as well. Coal plants are only economically profitable when they are running on a continuous basis – basically all the time. At the time when there is an overgeneration risk, solar panels have to be turned off to avoid overloading or damaging the power grids. This ultimately means that we are wasting a clean and renewable source of energy.
Looking back to our Duck Curve, the solution relies on softening the peaks and ramps and hence, making the curve more flexible –flattening the duck. But how do we do this? There are many approaches but two main ones should be highlighted: The first main solution is interconnection. The more grids can be connected with one another, over larger areas, the more spread out the potential for solar power and the more spread out load will be. The second solution is energy storage. A sufficient amount of energy storage would ideally flatten the curve and allow a complete integration of wind and solar. But at least for now, this method remains still costly.
Renewable power is still underused around the globe. In tackling pollution and finding a way of powering our society with clean energy, there will be different political and economic interests involved. There is still a long way to go. However, it is not too bad to start with a graph that looks like a duck.