How does imbalance work

Energy Trading · 12 May, 2024 · 10 Min Read

Grid balancing

All interconnected European power networks are said to be balanced when they operate at a frequency of 50Hz. Inherent uncertainties in power consumption and generation cause the network frequency to permanently fluctuate. When more power is produced than consumed, the nominal frequency of the grid increases, whereas in a power deficit it decreases. Maintaining the frequency within a 0.05 Hz tolerance range is essential to the system stability (see European decision eur-lex.europa.eu ANNEX III, Table I). To ensure steadiness, TSOs can activate adjustment mechanisms and ancillary services when overall system imbalance is observed. Once the physical equilibrium has been guaranteed, the costs generated by the balancing operations are passed on to the balance responsibility parties (BSP) through the ISP as we will see later in the article.

Balance responsibility party

Let’s start by recalling some important notions. A balance responsibility party is an actor of the electricity market who enters into a contractual agreement with the TSO to ensure that they will cover the costs associated with the difference between injections and extractions on its balancing perimeter. The balancing perimeter is made up of injections (physical production sites, purchases on the stock exchange, downward flexibility…) and withdrawals (physical consumer sites, sales on the stock exchange or to other market participants, exports, …) contracted by the actor.

When the actor is not balanced, it is charged the amount of energy corresponding to its remaining open position at Imbalance Settlement Price, depending on the global trend of the system and its open position.

Imbalance Settlement Price Definition

The ISP is designed to limit the scope for arbitration between the financial settlement for the certification perimeter manager’s deviation and the financial settlement for supplier rebalancing (Code de l’énergie, Article R335-32). In other words, a BSP is penalized at the ISP price if it delivers less energy than it sells. Conversely, the BSP can earn revenue from the ISP if it produces more energy than it sells. However, the ISP price differs between overproduction and underproduction to incentivize balanced production by the BSP. The difference comes from a coefficient that is added to the calculation of the ISP and that depends on the balancing trend, the volume weighted average price (VWAP) of balancing costs, and the imbalance sign of the market actor. It is explained in more details later

ISP+ corresponds to the price at which the extra volume V+ of the producer is sold. The producer then earns ISP+ x V+. Beware that ISP+ is not signed and can be negative in some rare occasions, in which case this turns into a penalty.
ISP- corresponds to the price at which the missing volume V- of the producer is bought. The producer is then penalized at ISP- x V-. Beware that ISP- is not signed and can be positive in some rare occasions, in which case this turns into an earning.

Currently the ISP is computed for each half-hour, but will switch to 15 min before 2025 to comply with the European guideline : 2017/2195 « Guideline on Electricity Balancing » (EBGL).

VWAP Definition

VWAP definition The VWAP component of the ISP aims at reflecting the balancing operation costs. RTE (the French TSO) takes into account the following types of balancing energies (E_i)_{i ∈ I} and the associated prices (P_{E_i})_{i ∈ I} for both upward U and downward D operations (www.services-rte.com section 0.O.2.3.):

Specific balancing bids (only defined for balancing needs in mainland France)
Standards balancing bids (shared on European level)
Backup reserve exchange contracts
Standard RR energy
Automatic frequency restoration reserves (aFRR)
Manual frequency restoration reserves (mFRR)

VWAP are then computed using the following formulas:

VWAP_U = ^{∑_i(E_i,U × P_{E_i,U})} / _{∑_iE_i,U}

VWAP_D = ^{∑_i(E_i,D × P_{E_i,D})} / _{∑_iE_i,D}

Coefficient

TFactor $k$ is a parameter of the imbalance settlement price. It is defined ex-ante with the aim of balancing income and expenditure of the energy component of the balancing-imbalances account based on historical data. – RTE

Since the 1st of January 2023 the parameter $k$ is computed every month based on the cumulative balancing imbalances account (BIA) denoted $S$ which regroups the income and expenses generated by the imbalance of different market actors. Each month $M$, $k$ is calculated from the BIA up to the end of the previous month, and will be applicable for 3 months later. It evolves according to the following piecewise-linear function.

The goal of this new definition is to encourage not only the actors to be balance, but also the global system to move towards long-term stability by penalizing less imbalances that are opposed to the system imbalance. For example, if the system is globally in an energy deficit (generation < production), then $S$ increases, and $k$ decreases. Looking at the definition of the ISP in the table

What’s in there for me ?

Because of the uncertainty around battery activation, most aggregators don’t take imbalance costs into account when they optimize the asset on frequency restoration reserve markets, but these can be huge ! For a 2h battery optimized on the cross market it amount to 25% of the battery revenue ! Moreover, because of RTE’s invoicing system you won’t see the associated costs before the end of the month, and the consolidated data only a year later, leaving you with an unpleasant surprise. At StackEase, not only do we provide you with a vision of the imbalance costs, but we also ajust our bidding strategy on the aFRR, FCR, Day-Ahead and Intraday market to maximize the asset profits after balancing correction. If you want to know more about how we can optimize your battery, and help you maximize profit with cross market optimization get in touch with us here.