Glossary · Energy industry & energy transition
Frequency control
Frequency control comprises the technical and organisational measures used to keep the frequency of an alternating current grid close to its setpoint. For all synchronous areas of the EU, the nominal frequency is 50 hertz. The frequency directly indicates whether generation and consumption in the synchronously coupled grid are in balance. If more electrical power is drawn than is fed in, the frequency falls. If more is fed in than is drawn, it rises.
What matters here is not the quantity of energy over a longer period, but the instantaneous power. A kilowatt hour indicates how much energy was delivered or drawn over a given time. The frequency, by contrast, reacts at every moment to the ratio of feed-in to withdrawal. A power system can therefore generate enough energy over the course of a day and still run into short-term frequency problems. This happens when power is lacking at the wrong time or too much power is fed in at the same time.
Frequency as a balance signal
In an alternating current grid, large synchronous generators, motors and other rotating machines are electrically coupled to one another. Their rotational speed follows the grid frequency. In Continental Europe, many countries share a common synchronous area. Germany therefore does not hold the frequency alone, but together with the other transmission system operators of this interconnection. If a large power plant fails in France, Poland or Italy, the same frequency falls as it would after a failure in Germany. However, the load flows are distributed differently from region to region.
The frequency is not a local grid signal like the voltage. Voltage can differ greatly from place to place depending on grid level, line, transformer and feed-in situation. A frequency deviation, by contrast, fundamentally affects the entire synchronous area. Physically, however, it does not propagate completely without delay. This distinction is important because frequency control is often mixed up with voltage control or grid congestion management. A local voltage rise in a distribution grid with a lot of photovoltaics is not a frequency problem. An overloaded line is likewise not relieved by frequency control. That requires grid expansion, operational switching measures or redispatch.
Permissible deviations: what is normal?
The grid frequency does not have to be exactly 50.000 hertz at all times. The EU guideline on electricity transmission system operation, Regulation (EU) 2017/1485, sets the following parameters for the Continental Europe synchronous area:
-
Standard frequency range: plus/minus 50 millihertz. Small fluctuations within this band are routine.
-
Maximum steady-state frequency deviation: 200 millihertz. This is how far the frequency may still deviate after a disturbance once it has settled.
-
Maximum instantaneous frequency deviation: 800 millihertz. This value applies to the brief lowest or highest point directly after a disturbance.
-
Frequency restoration time: 15 minutes. As a quality target, a maximum of 15,000 minutes per year outside the standard frequency range is provided for.
These values are standard values. The transmission system operators of a synchronous area can jointly propose other values. Other synchronous areas such as Great Britain or Northern Europe in some cases have significantly wider bands. It becomes critical when the frequency changes too quickly, the reserves are insufficient or protection systems disconnect plants. A disturbance can then spread. As a last stage, graded automatic load shedding is provided for in the event of underfrequency. In Continental Europe it usually begins at around 49 hertz.
Time-staggered frequency maintenance
After a disturbance, several mechanisms take effect one after another. They can be compared to a relay race: each stage bridges the time until the next one takes over.
-
Inertia (instantaneous reserve): Directly after a disturbance, the inertia of rotating masses slows the change in frequency, because kinetic energy is stored in generators and motors. This response sets in without delay and is not an actively controlled regulation. However, it limits the rate of change of the frequency and buys time for the subsequent reserves. Until now it came mainly from synchronous machines in conventional power plants. In future, grid-forming inverters are also intended to contribute. In Germany, the transmission system operators have procured instantaneous reserve on a market basis under § 12h EnWG since 22 January 2026.
-
Frequency Containment Reserve (FCR, formerly primary control): Providers measure the frequency themselves on site and react automatically. The FCR must be delivered in full within 30 seconds and is designed for deployment of no more than 15 minutes. It is held symmetrically and activated throughout the synchronous area according to the solidarity principle, regardless of where the disturbance arose. The FCR arrests the frequency drop but does not bring the frequency back to 50 hertz.
-
Automatic Frequency Restoration Reserve (aFRR, formerly secondary control): Each transmission system operator determines the need roughly every four seconds from the balance of its control area. A power-frequency controller gives providers setpoints. The power must be delivered in full after five minutes at the latest. The aFRR is activated in the control area in which the imbalance arose. It brings the frequency back and replaces the FCR.
-
Manual Frequency Restoration Reserve (mFRR, minute reserve): It compensates for larger or longer-lasting imbalances and relieves the aFRR. It is called up via an electronic merit order list system, not by telephone. Full power must be available within 12.5 minutes.
How much FCR is needed is determined jointly for the whole of Continental Europe. For a long time, the FCR was designed for a reference incident of 3,000 megawatts, i.e. roughly the simultaneous failure of several large power plant units. With 3,000 megawatts of FCR and full activation from a deviation of 200 millihertz, the calculation yields a power-frequency characteristic of 15,000 megawatts per hertz. A probabilistic approach now supplements the dimensioning. It is intended to limit the probability that the FCR is insufficient to at most once in 20 years. This can lead to values above 3,000 megawatts.
Balancing capacity and balancing energy
In everyday usage, these products are often grouped together under the term balancing energy. Strictly speaking, two things must be distinguished. Balancing capacity is the reserved ability to increase or decrease power when needed. Balancing energy is the energy actually delivered or absorbed when called upon. For frequency stability, what counts first is that enough power is quickly available. For costs and settlement, what counts afterwards is how much energy was used. In Germany, for example, the FCR is remunerated only via a capacity price. The energy that flows into a control area through FCR activation is settled among the transmission system operators.
Distinction from balancing group, redispatch and security of supply
Frequency control does not replace careful schedule management. Electricity suppliers, direct marketers and larger market participants plan their expected feed-ins and withdrawals in a balancing group. If they deviate from this, a need for balancing energy arises. The balancing reserves compensate the physical deviation immediately. The later settlement assigns economic responsibility to the party that caused it. In the longer term, the affected balancing group must eliminate the imbalance itself.
Redispatch serves a different purpose. Here the feed-in of power plants or installations is changed in order to avoid or eliminate line congestion. The overall balance usually remains the same: less is fed in at one location, more at another. Frequency control, by contrast, reacts to a power balance. It does not ask which line is overloaded, but whether there is too much or too little power in the grid overall.
Frequency control must also be distinguished from security of supply. Security of supply means reliably supplying consumers even in the event of outages, weather changes, load peaks and market uncertainties. Frequency control is part of this, but not all of it. A system can have sufficient frequency reserves and still have regional grid congestion, too little firm capacity or problems with restoration after a large-scale outage.
Renewable energies and new providers
With more wind power and photovoltaics, the physical principle does not change: generation and consumption must still be balanced at every moment. What changes, however, are the sources of controllability. Conventional power plants with synchronous generators brought rotating mass, short-circuit power and controllable turbine output along with them, so to speak, as a by-product. Wind farms, photovoltaic installations and battery storage are mostly connected via power electronics. They can respond very quickly. However, they provide system services only if they are technically designed for it, suitably configured and integrated in market or regulatory terms. The separate procurement of instantaneous reserve is an example of how properties that were once taken for granted are now purchased deliberately.
It does not follow that renewable energies make frequency control more difficult in principle. Modern inverters, battery storage, flexible industrial processes, electrolysers and controllable consumers can provide balancing capacity. If the frequency falls, a storage unit can feed in or a consumer can reduce its load. If it rises, a storage unit can charge or a consumer can draw more power. What matters for the grid is that power changes in the right direction, and with the required speed and reliability. Flexibility thereby becomes a system-relevant property. Frequency control thus depends more heavily on connection rules, prequalification, data availability and market products.
Costs and common misinterpretations
Frequency control costs money. Installations must reserve capacity, provide measurement technology and communication links, and in operation deviate from a purely market-driven mode of operation. The mere readiness to react within seconds or minutes therefore has a value. The German transmission system operators tender the reserves daily, the FCR together with other European grid operators. In a prequalification, providers must demonstrate that they deliver the required power quickly enough and reliably.
In political debates, frequency stability sometimes serves as an argument for or against particular technologies. Such interpretations remain imprecise if they do not name the control mechanisms. A coal or gas power plant can support the frequency if it is connected to the grid and operated accordingly. A battery storage unit reacts even faster, but is limited by energy content, operating strategy and market rules. A photovoltaic installation can provide positive reserve only if it is operated with headroom. Every option has technical prerequisites and economic opportunity costs.
With increasing electrification, load profiles and residual load change. Heat pumps, electromobility, data centres and electrolysers not only increase electricity demand but also change when it occurs. For frequency control, it matters how quickly these loads can react and whether they can be controlled in a bundled way. High consumption is not in itself a frequency problem. What is problematic is an imbalance that is faster or larger than the available control mechanisms.
Frequency control shows that a power system is more than an energy balance. At every moment it must balance power, absorb disturbances, settle responsibility and provide technical properties that were often previously supplied as a by-product of power plant operation. The term therefore describes not a peripheral function but a central link between physics, market rules and institutional responsibility.
Sources
- Verordnung (EU) 2017/1485 der Kommission vom 2. August 2017 zur Festlegung einer Leitlinie für den Übertragungsnetzbetrieb — Europäische Union (EUR-Lex); 2017-08-02. Art. 127 (qualitätsbestimmende Frequenzparameter, Frequenzqualitäts-Zielparameter), Verweis auf Anhang III
- Commission Regulation (EU) 2017/1485, Annex III (adopted version) — The National Archives (legislation.gov.uk); 2017-08-02. Annex III, Table 1 und Table 2, Spalte CE
- Welche Arten der Regelreserve gibt es? — Deutsche Übertragungsnetzbetreiber (regelleistung.net); unbekannt. Abschnitte FCR, aFRR, mFRR und Tabelle mit technischen Eigenschaften
- Momentanreserve — Deutsche Übertragungsnetzbetreiber (regelleistung.net); unbekannt. Abschnitte „Was ist Momentanreserve?“ und „Marktstart und Umsetzung“
- Genehmigung zum Vorschlag der kontinentaleuropäischen ÜNB zur FCR-Dimensionierung (Az. 622-23-008) — Bundesnetzagentur; unbekannt. Abschnitte zur FCR-Dimensionierung nach Art. 153 SO-VO und zur MW/Hz-Kennlinie
Related glossary entries
- power
- redispatch
- Frequency Containment Reserve
- Balancing capacity
- balancing group
- security of supply
- Flexibility
- residual load
Edited by eCMS on 12.11.2026 05:05
© 2026 Erik Schmidt. All rights reserved. Quotations with a link to the source are welcome.