Electricity is many things. It’s a current, it’s a field, it’s a set fascinating physical phenomena. But more importantly, it’s also a commodity, which can be produced, traded and consumed. Just like natural gas, grain or tulips. However, unlike in the case of grain (but not really tulips), the production and consumption of electricity must happen at exactly the same time. This still tends to be true, despite recent advances in storage technologies. One can, therefore, say that supply and demand in electricity must be in perfect balance at all times. (This must be true at least within the boundaries of a well-defined geographic area, often called a “balancing zone”. Which also often coincides with the state border of European countries.) What would happen if it wasn’t? Well, the short answer to that is: that’s when you have a blackout.
Why do we balance?
Balance is the central key element to the industry. It helps us understanding electricity in the form we encounter it in our homes. If someone flicks on the light switch that creates additional demand, it will need to be met by a power plant (a generator) somewhere. If a thousand people flick their light switch on, that creates an even greater demand. If generators knew all this in advance, they would increase their production accordingly. The problem is: they don’t. Noone does, for sure. That’s why there are electricity demand forecasts which attempt to estimate what that demand is going to look like in the future. Trading of electricity is essentially based on these forecasts. However, these forecasts can never be perfect. If we could forecast our every act and every motion, there would be no need for balancing.
So why do we balance? We balance, because of our imperfections. Because of our deviations from our plans. Because we are not omniscient. Because that’s what balancing essentially is: making up for our forecast errors.
What do we balance?
We live in market economies, where the trading of electricity is considered to be a privatized business, with many market players. More or less. In the majority of current EU countries (including the United Kingdom, for now), the chosen market design for trading electricity is the so-called “net pool”. In a net pool, the buying and selling of electricity may happen directly between consumers and producers and there may be intermediaries of different sorts (including power exchanges, where electricity is traded in a similar manner as any other commodity on a stock exchange). A “gross pool” is a different market design that is less popular in the European Union (utilised in Ireland and to some extent, Spain), but used elsewhere (in some US markets, Ukraine and Australia). In a gross pool, basically all electricity that is produced in the country must be sold to a single business entity (a central counterparty), from which consumers purchase that electricity. As you may have guessed, a gross pool is a gross pool because ALL electricity “flows” into one big pool; a net pool is a net pool because SOME electricity may flow through the pool (such as an exchange), but that’s up to the market players to decide.
There are quite important differences between the two market designs discussed above. In a net pool, it is the market player’s responsibility to find a suitable partner whom she/he would want to sell to or buy from. In a gross pool, market players do not need to worry about finding a buyer or seller, because there can only be one buyer/seller. Both market designs require forecasts of what electricity consumption and production is going to look like in the near future. The process of determining how that forecast is going to be met is called scheduling. In a net pool, market players are responsible for making their own forecasts and sending their own schedules to a central coordinating body, often called the Transmission System Operator (TSO). Besides operating the power poles and wires – the transmission system – , TSOs are usually responsible for keeping the electricity system in balance and in order to do that, they need to know what each market player intends to do (trade, produce or consume) in the near . TSOs also have real-time visibility of what’s going on in the whole system. In a gross pool, there is usually an auction where generators compete to meet forecasted demand at the lowest cost possible. After that, the central coordinating body (usually called a Market Operator) prepares and sends the schedules to the winners of the auction. They must follow those instructions set in the schedules in their production. The schedules tell generators how much electricity they should produce at what time of the day.
Balancing is needed where and when consumption or production does not go according to schedule. As mentioned earlier, forecasts are never perfect and because everyone’s given the right to flick that light switch without prior notice to the TSO, balancing will be necessary in the foreseeable future. What covers up these deviations from plan is called “balancing energy”. It is important to note, that balancing energy can be “negative”, meaning that if for some reason supply exceeds demand, there may be a need of producing less electricity. This “lack of production” helps restoring balance, therefore, it is also considered to be balancing energy. For the avoidance of doubt: scheduling is always a planned process, regardless of the market design in which it happens. Balancing is always a reaction to the unplanned, a correction of the “unforecastable”.
How do we balance?
Balancing is usually done via different, bespoke market arrangements made with TSOs or Market Operators. Continental Europe (the EU being the trendsetter) has a fairly standardized approach to what technical solutions and “products” and TSOs may use to achieve system balance. The rules of balancing are becoming more harmonised across EU countries every year. (Australia, of course does things rather differently.) In summary, these products are usually differentiated based on the reaction time its providers to the TSO (or Market Operator) must comply with the TSO’s instruction to either increase or decrease their electricity input to the grid and the time interval this change must last. These services are usually called “ancillary services”.
How it works is that with the help of specialized equipment, the TSO is able to “sense” when there is a mismatch between planned supply and demand. In order to make up for the difference, and depending on how big the deviation is, the TSO will start using the products offered by the ancillary service providers (mostly generators). In other words, the TSOs will direct (as if they had a remote controller to the generators) the actual production of the power plants – up and down. First, the fastest product is used, then the second fastest and so on. In the EU, these products are called “primary”, “secondary” and “tertiary control reserves”, with primary being the fastest and tertiary the slowest. For smaller anomalies (e.g. flicking on a very large light switch) primary reserves are usually enough. For bigger problems, such as an aluminium smelter, suddenly stopping consumption from the grid because of a faulty machine, the deviation from plan maybe so big that tertiary reserves may be needed to restore the system’s balance. In Australia, the idea of differently timed products is similar, but they are named after the time length they are needed for (6 second, 60 second and 5 minute).
There are typically auction rules and/or agreements behind these activities that set out how these services are called upon and settled at a later stage. You can think of this as if power plants would sell their availability for being directed to produce more or less electricity. Those responsible for the deviations from plan cause the need for procuring ancillary services and usually end up paying for them at some point. In Australia, this principle is often referred to as “causer pays”. Companies providing electricity for households (called universal service providers or suppliers in Europe or retailers in Australia) usually include these costs in their pricing, although this level of detail might not be printed out in their bills . The assumed cost of balancing will ultimately be factored in to the electricity price.
*but were afraid to ask