EENS, Expected Energy Not Served · Glossary

EENS stands for Expected Energy Not Served. It refers to the amount of energy that, in a power system and within a given period, is expected not to be supplied because the available resources do not cover demand. The official German version of EU law calls the metric "erwartete Energieunterdeckung". The indicator belongs to resource adequacy, that is, to the question of whether generation, storage, imports and flexible demand together are sufficient. EENS does not describe a power outage that has actually occurred. It is an expected value derived from many simulated combinations of load, weather, power plant outages, grid availability and market outcomes.

Definition in the European regulatory framework

The ERAA methodology of the EU agency ACER builds up the metric in two stages. Energy Not Served (ENS) is the energy that is not supplied in a market time unit and in a modelled zone because resources are insufficient. EENS is the expected value of this ENS for a zone and a period. ERAA stands for European Resource Adequacy Assessment, the European assessment of the adequacy of resources. Under the EU Electricity Market Regulation (EU) 2019/943, this assessment must contain at least EENS and LOLE and must be calculated probabilistically. Member States that use a capacity mechanism also need a reliability standard. It is calculated at least from the value of lost load (VoLL) and the cost of new entry (CONE) of the cheapest market entry, and is expressed as EENS and LOLE (Article 25).

Unit and worked example

EENS is a quantity of energy. The ERAA reports give it in GWh per year, while smaller studies often use MWh per year. EENS therefore differs from power, which is an instantaneous quantity in megawatts or gigawatts, and also from load and peak load. A simple example: if 500 megawatts are missing for one hour, that is 500 megawatt-hours of unserved energy. If 100 megawatts are missing in each of three hours, the result is 300 megawatt-hours. If the value of 300 MWh occurs in only one of ten simulated years and all other years show no deficit, the EENS is 30 MWh per year. The indicator thus combines the magnitude, duration and probability of a shortage.

Distinction from LOLE and from outage indicators

EENS is often confused with LOLE (Loss of Load Expectation). LOLE gives the expected number of hours per year in which demand cannot be fully met. EENS measures how much energy is missing in those hours. Two systems can have the same LOLE value and still very different EENS values. In one case only a few hundred megawatts are missing in several hours, in the other several gigawatts are missing in the same hours. LOLE therefore shows how often the system reaches its limit. EENS shows how deep the gap is in expectation.

EENS must also be distinguished from actual interruption indicators such as SAIDI. SAIDI describes the average interruption duration for final customers, which is usually caused by local grid faults. The EENS of an adequacy analysis, by contrast, relates to a bidding zone or modelled zone as a whole. A cable fault in the distribution grid can affect many customers even though the system as a whole has enough energy.

Probabilistic calculation

EENS is generally calculated using Monte Carlo simulations. The model runs through many combinations of climate or weather years and random power plant and line outages. For each run, a market model with economic dispatch of the plants is calculated, in Europe including the exchange capacities between zones. In every hour in which demand remains unmet after all modelled resources have been dispatched, ENS is recorded. EENS then results as the arithmetic mean of ENS across all model runs. With several nodes, it is summed for the area under consideration. In addition to the mean, percentiles are also reported, for example the 95th percentile of ENS. This is important because a mean can be strongly shaped by a few extreme years. The mean is therefore not a forecast of the quantity of electricity that "will be missing". It is the statistical result under certain assumptions.

These assumptions determine what the indicator tells us. They include the development of demand, for example through heat pumps, electromobility and industrial electrification, as well as decommissioning and new build of power plants, the expansion of storage, the selection of weather years and the availability of imports in scarcity situations. Demand flexibility also counts: if it is assumed too optimistically in the calculation, EENS falls only on paper. The ERAA methodology also requires results before and after the activation of non-market reserves to be reported separately. An EENS value is therefore comparable only if the model framework is stated.

Use in assessment and regulation

Unlike pure hour-based measures, EENS shows how severe a shortage is. The indicator is therefore suitable for quantifying the benefit of additional resources. In the ACER methodology for VoLL, CONE and the reliability standard, it is examined how strongly additional capacity reduces EENS. This reduction is then valued using the Value of Lost Load and set against the costs of new capacity. A gas turbine, a battery storage system, load shifting in industry or a better grid connection, however, reduce EENS only if they are actually available in the critical hours. For this purpose, the methodology takes into account the expected availability as well as energy and activation limits in hours with ENS. A VoLL is always a flat-rate estimate: a announced load reduction in a controllable process causes different damage than the failure of critical infrastructure.

In Germany, a reliability standard of 2.77 hours per year applies to the German-Luxembourg bidding zone. It was set by the Federal Ministry for Economic Affairs together with Luxembourg and is communicated as a LOLE value. It is necessary because Germany has operated a capacity reserve as a capacity mechanism since 2019. German security-of-supply monitoring reports EENS in addition to LOLE, but LOLE is decisive for the assessment.

Common misinterpretations

EENS is not a blackout forecast. In German reports, hours with a market-side shortfall expressly do not count as a blackout. In such hours, instruments outside the market, such as reserves, are deployed first. Whether controlled load shedding actually occurs depends on which resources are included in the model and on the operating processes of the transmission system operators.
A low EENS does not mean comprehensive security of supply. The indicator concerns the adequacy of resources. It does not say whether the grid can transport certain power flows, whether frequency and voltage remain stable, or whether IT security is ensured.
An EENS of zero in all runs means only that no shortage occurred within the model framework. Events outside the weather years and outage patterns considered are therefore not ruled out.

EENS in a weather-dependent power system

When wind and photovoltaics supply a large share of generation, installed capacity no longer determines EENS. What matters is what is available in hours with high residual load, that is, with high demand and low feed-in from wind and sun. Photovoltaics deliver a lot of energy at midday in summer, but hardly help against winter evening scarcity. A battery storage system with high power and short storage duration can close short deficits well, but can close multi-day dark doldrums only partially. Imports reduce EENS only if neighbouring countries are not scarce at the same time and line capacity is free. Demand flexibility works only if it can be called upon contractually, technically and operationally. EENS therefore forces us to look at the temporal structure of availability.

Which EENS is acceptable cannot be answered purely technically. More firm capacity and reserves reduce the risk but cost money, which is borne through capacity payments, levies, grid charges or market prices. The European approach to the reliability standard is therefore based on an economic trade-off between the value of damage and capacity costs. An EENS value is robust only once the model framework is disclosed: period, weather data, load assumptions, availabilities, imports, grid limits and whether reserves are taken into account.

Sources

Verordnung (EU) 2019/943 über den Elektrizitätsbinnenmarkt — Europäische Union, EUR-Lex; 2019-06-05. Art. 23 Abs. 5 Buchst. h und j; Art. 25 Abs. 1–3
ACER Decision on the ERAA methodology: Annex I — ACER / ENTSO-E; 2020 (ACER-Entscheidung 24/2020). Art. 2 (Definitionen ENS, EENS); Anforderungen an Adequacy-Metriken
European Resource Adequacy Assessment 2023 Edition, Annex 2: Methodology — ENTSO-E; 2023. Abschnitt 11.2 Adequacy Indicators
Der Zuverlässigkeitsstandard für die Stromversorgung in Deutschland — Bundesministerium für Wirtschaft und Klimaschutz; unbekannt (laut Monitoring-Gutachten 2022 veröffentlicht). gesamtes Informationsblatt
Bericht Versorgungssicherheit Strom – Stand und Entwicklung der Versorgungssicherheit im Bereich der Versorgung mit Elektrizität — Bundesministerium für Wirtschaft und Klimaschutz / Bundesnetzagentur; 2022. Abschnitt zur marktseitigen Versorgungssicherheit, Fußnoten zu LOLE/EENS

Related glossary entries

Edited by eCMS on 11.11.2026 19:42