Project Zero is not intended to be available for use, purchase, or access by U.S. persons, including U.S. citizens, residents, or persons in the United States of America, or companies incorporated, located, or resident in the United States of America, or who have a registered agent in the United States of America.
Part 5 of 5·Published on 7 Jan 2025In an earlier post we looked at how electricity markets are designed, focusing on how suppliers compete against each other and a merit order is used to find the price of electricity at a given time. Now we’re going to explore the demand side of the market.
You might recognise the graph below – it’s a good representation of most markets. When the price goes down, demand increases (people want more of that good). When the price goes up, demand decreases (people want less of that good).
Copyright 2023 Encylopedia Britannica
Historically, wholesale electricity markets have been different: its demand curve has been highly inelastic. In other words, if the price of electricity moves up or down, it doesn’t change most people’s consumption behaviour in the short term.
Therefore the short term demand curve for electricity currently looks more like this:
Retail energy providers handle demand inelasticity by hedging their delivery obligations. Coming up to delivery, they buy electricity over time, smoothing out short-term volatility. This allows your provider to offer a static rate to you for your electricity, instead of one that changes from hour to hour.
Of course, if electricity prices stay persistently high on the wholesale market, this translates to higher retail rates and a decrease in demand. People and businesses use less heating or lighting once they see higher bills.
Despite the proliferation of smart devices and distributed energy resources, demand inelasticity persists. This is mostly due to people having little means to take advantage of electricity price signals throughout the day: information flow is poor and there is no good incentive mechanism to shift usage.
Matching demand yields a price for a specific point in time. Now let’s examine the typical shape of these inelastic demand values as we extend them over the course of a day.
Across an entire country, demand varies throughout any 24 hour period. It rises sharply in the morning and then increases slightly around 5pm, before declining gradually over the course of the evening. It also varies seasonally, as you can see below.
These curves represent typical consumption habits, and are relatively uncorrelated to the hourly price of electricity, hence the price inelasticity described above.
Again, this makes sense given most people’s daily lifestyles. People wake up in the morning, turn on appliances at home, leave for work, then return home in the evening.
You can see Britain’s live electricity usage, and see its demand curve for the past day here. Other countries, like France and the United States, also post their data online.
Given the inflexibility of demand, providers must vary supply to match it. This is done with a combination of base load and peak load.
Some supply sources, such as nuclear, are hard to turn up and down, but once running will generate consistent power for the grid, i.e. base load. On the other hand, natural gas plants ramp up or down quickly, providing peak load to match short term demand needs.
However, in renewable grids, varying supply to match demand is more challenging. While having very low marginal costs of production, we also have less control over when that production happens.
Solar output peaks when the sun is at its brightest – midday – and demand peaks in early evening, both of which put strain on grids that are overdependent on solar. In such systems, the net demand for supply follows the two-peak duck curve.
This can be observed in the demand curve for Queensland in Australia, which increased its reliance on rooftop solar over the period of 2016-2020:
Source: AEMO
Other sources of clean energy also find it difficult to meet demand peaks. It’s tough to turn nuclear output up or down. Hydrogen is promising, but the technology requires more development, which we’re working on. Many forms of long duration storage aren’t scalable (e.g. there are a limited number of sites suitable for pumped hydroelectric power) or are still at the early stages of development. Ultimately gas-fired peaker plants are still being used to meet peak demand.
So what should we do? Create incentives to shift demand.
From the perspective of the grid operator, ramping demand down is equivalent to ramping supply up, and vice versa. A collection of energy resources used to ramp demand up or down can be viewed as a ‘virtual’ generator doing the opposite action.
Such an aggregation of renewable generation, EVs, battery storage, and smart appliances is a Virtual Power Plant (VPP). And we can use them to collectively move our power usage to align with renewable generation profiles. This reduces our reliance on pricey peaker plants and supports grids strained by solar overproduction.
Coordinating energy devices to respond to price signals is enabled by connecting them across the Project Zero network.
For operators, VPPs flatten the duck curve, making renewable generation more useful and economical, and the grid more resilient. For consumers, this means greener energy, a more efficient market, and ultimately lower prices for everyone.
VPPs allow us to deploy renewables faster and to flex demand to their shape, ultimately reducing the costs and time to get to net zero. This is why they are so crucial.
As far as we have seen, no project has been able to unlock the incentives to deploy and coordinate VPPs at scale. We believe Project Zero can make that happen.
