Real-time data · Glossary

Real-time data are data that are collected, transmitted and processed quickly enough to describe the current state of a system with sufficient accuracy for a specific technical or economic decision. In the power system, they include, for example, frequency, voltage, power flows, switching states, feed-in, consumption, storage levels, the availability of plants, grid congestion or weather data. The term does not denote a uniform time span. What counts as real time depends on what the data are needed for.

In protection technology, milliseconds count, because a fault current must be detected and disconnected very quickly. In grid operation, seconds or a few minutes are sufficient depending on the process, for example to monitor load flows or to prepare switching operations. In intraday trading, quarter-hour values and forecasts updated every minute already count as very timely. In the German market area, continuous trading is possible until about five minutes before the start of delivery. Household customers who want to understand their electricity consumption better usually manage with hourly or quarter-hourly values. Anyone speaking of real-time data should therefore always state at what level and for what purpose the data are used.

Real time in the European regulatory framework

European law also uses the term functionally. The guideline on electricity transmission system operation, Regulation (EU) 2017/1485 (System Operation Guideline, SO GL for short), distinguishes three types of data: structural data, scheduled and forecast data, and real-time data. It governs how these data are exchanged between transmission system operators, distribution system operators and significant grid users, and at what time intervals this takes place. The transmission system operators of a synchronous area exchange their grid state data, including frequency, via a Europe-wide IT tool. Distribution system operators must provide the responsible transmission system operator with real-time information from its observability area, unless the latter determines otherwise. Typical items are the position of circuit breakers and active and reactive power.

A definition in the Regulation shows how fine the time resolution can be. Instantaneous frequency data there are measurements of the system frequency with a measurement period of at most one second, used to assess frequency quality. The Regulation has applied directly in all EU Member States since 2017. The exact scope of the data exchange is determined by the transmission system operators together with distribution system operators and grid users. In Germany and Austria, national implementing documents therefore also apply.

Timeliness, measurement quality and system function

Fast data are not automatically accurate data. A measured value can be available very quickly and still be erroneous, incomplete or poorly calibrated. Conversely, a settled meter value can be very accurate and still arrive too late to influence grid operation. For the power system, what matters is the interplay of timestamp, measurement interval, transmission time, data quality and processability. The SO GL expressly requires grid operators to provide and use high-quality data.

The location of the measurement also plays a role. A measured value at a household's grid connection point says something different from a measured value in a substation, at a wind farm, in a battery storage system or in a control room. The closer the measurement is to a technical bottleneck, a controllable plant or a critical operating point, the more directly it can be used for control. Aggregated data may suffice for market processes, balancing or planning. However, they do not replace measurements at points that are important for the grid in every case.

Real-time data must also be machine-readable, checked for plausibility and integrated into the existing control technology, for example into SCADA systems for monitoring and remote control. A single current value is of little use if it cannot be assigned, if the timestamp is missing or if it is unclear which plant, voltage level or balancing zone is affected. The data become technically valuable only when measurement, communication, data model, responsibility and an option to intervene come together.

Distinction from forecasts, structural data and settlement data

Real-time data are often confused with other types of data. Forecasts describe expected future states, such as feed-in from wind and solar plants or electricity consumption the next day. They can be updated at very short notice, but remain model calculations with forecast errors. Real-time data, by contrast, describe measured or directly observed states. Estimated state data from grid state estimation lie in between: they are current, but partly calculated and not directly measured.

Structural data contain relatively permanent information about plants, grid connections or market roles, for example capacity, location, voltage level, metering point or operator. Without correct structural data, real-time data often cannot be interpreted meaningfully. The current measured value of a storage system is only usable if its maximum charging power, its grid connection capacity and its controllability are known. The SO GL therefore also links real-time exchange with a regular exchange of structural data.

Settlement data serve to allocate energy quantities for balancing and financial purposes. They can have a high time resolution, but are usually not processed as quickly as grid protection or operational grid control require. A quarter-hour measurement from a smart metering system is therefore not automatically a real-time data set in the sense of grid operation. It can be very valuable for variable tariffs, consumption transparency or balancing, but it does not replace control technology.

A further distinction concerns power and energy. Many real-time data in the power system describe instantaneous or very short-term power in kilowatts or megawatts, not the quantity of energy consumed in kilowatt-hours. For grid operation, the current loading of a line is often more important than the energy quantity summed over a month. For settlement and statistics, the reverse is often true.

Grid state data from smart metering systems in Germany

In Germany, the Metering Point Operation Act (Messstellenbetriebsgesetz, MsbG) governs how data are collected via smart meter gateways. Grid state data are defined there as voltage and current values, phase angles and values derivable from them with which the grid state can be determined. Metering point operators may collect such data only on behalf of the grid operator and only in justified cases. A justified case exists if the data are necessary for safe and reliable grid operation and the distribution system operator collects them at specific points. These include installations under the EEG and KWKG, controllable consumption devices and grid connections under § 14a EnWG, and metering points with a smart metering system. Grid operators must document the collections.

How promptly the data arrive differs considerably. The services of the metering point operator described in § 34 MsbG include, on the one hand, collecting grid state data with quarter-hour accuracy and transmitting them to the grid operator once a day. On the other hand, they include transmission every minute, but only for up to 25 percent of the grid connections equipped with smart metering systems in a grid area. This illustrates well how relative the term is: a quarter-hour series delivered daily helps with analysis and planning in the distribution grid. For ongoing monitoring, transmission every minute is more likely to be suitable.

Significance for grid operation and flexibility

With the expansion of photovoltaics, wind energy, heat pumps, electric cars and battery storage, conditions in the power grid are changing. More and more plants feed in depending on the weather or draw electricity at times that depend on prices, comfort preferences, charging states or control signals. As a result, the current state of the distribution grid becomes more important. In the past, many grid areas could be operated with fairly rough load assumptions. In grids with a lot of decentralised feed-in and new consumers, historical standard load profiles are increasingly insufficient.

Real-time data help to detect grid congestion, monitor voltage and deploy controllable plants more precisely. They are a basis for flexibility, because flexibility only helps the system if it is known when and where it is needed. An electric car can postpone charging, a heat pump can buffer heat, a storage system can charge or discharge. Whether this is useful depends on whether price, grid or balancing signals arrive in time and whether the plant is allowed to respond to them.

In the transmission grid, real-time data have long been part of system operation. Frequency control, redispatch, balancing energy and congestion management rely on continuous observation and control. In the distribution grid, digitalisation was for a long time less advanced, because many low-voltage grids were planned and operated passively. With growing electrification, part of the operational attention is shifting to these lower voltage levels. New requirements arise there for metering concepts, communication paths and the division of tasks between grid operators, metering point operators, suppliers, aggregators and plant operators.

Market, balancing and institutional limits

Real-time data are also important for the electricity market, but market and grid follow different rules. Market participants need timely information on prices, schedules, generation, demand and balancing energy. Grid operators need information on physical loading, voltage, switching states and available equipment. A market price can indicate scarcity, but does not reveal whether a particular line in the distribution grid is overloaded. A grid congestion can occur locally even though the wholesale price shows no problem.

An important limit follows from this: more real-time data alone do not resolve a conflict between market incentive and grid constraint. If many plants respond simultaneously to the same price incentive, this can even amplify load peaks. If grid operators receive measured values but have no legally and technically clear control options, their response remains limited. Whether real-time information becomes practically usable is decided by the rules: data access, control rights, remuneration, liability, data protection and IT security requirements.

Costs also do not disappear through digitalisation. Sensors, communication infrastructure, control systems, data platforms, certification and operation cause expense. The expense is worthwhile if it allows grid expansion to be planned more precisely, equipment to be better utilised or flexibility to be deployed more efficiently. But it can also end in expensive data collection if it remains unclear which decision the data are supposed to improve.

Typical misunderstandings

Real-time data are supposedly always better than less timely data: for many tasks this is not true. Grid planning needs long-term load and feed-in data, scenarios, connection requests and operating experience. Annual statistics, standard load profiles and weather years continue to serve their purpose. Real-time data improve decisions in ongoing operation, but replace neither planning nor reliable forecasts.
More transparency supposedly means more controllability: a control centre can only act if responsibilities, technical interfaces and permissible interventions have been clarified. This applies in particular to plants behind the grid connection point, such as home charging points, heat pumps or battery storage systems. There, property rights, comfort demands, supply contracts, grid charges and safety requirements meet.
Real-time data are supposedly public data: many real-time data are operationally critical or personal. Detailed consumption data can reveal when someone is at home and how they live. Grid data can be security-relevant. Outside the cases regulated by law, the MsbG permits the collection of grid state data only if they are not personal data. A good data regime therefore distinguishes between necessary access, aggregation, anonymisation, role-based rights and cybersecurity.

Real-time data make the current state of the power system visible to the extent that faster and more targeted responses are possible. Their value does not lie in the data stream itself. What matters is the appropriate time resolution, a reliable technical interpretation and the rules by which information becomes a permissible action.

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. 3 (Begriffsbestimmungen), Art. 40–44 (Datenaustausch, Echtzeit-Datenaustausch zwischen ÜNB sowie zwischen ÜNB und VNB)
Gesetz über den Messstellenbetrieb und die Datenkommunikation in intelligenten Energienetzen (Messstellenbetriebsgesetz – MsbG) — Bundesministerium der Justiz / juris; unbekannt (konsolidierte Fassung). § 2 Nr. 16 (Netzzustandsdaten), § 34 (Leistungen des Messstellenbetreibers), § 56 (Erhebung von Netzzustandsdaten)
Intraday-Handel — Next Kraftwerke GmbH; unbekannt. Abschnitt Vorlaufzeiten und Ablauf des Intraday-Handels

Related glossary entries

Edited by eCMS on 11.11.2026 19:42