Compressed Air Energy Storage · Glossary

Compressed air energy storage (CAES) is a mechanical electricity storage technology. During charging, electricity drives a compressor that compresses ambient air to high pressure and forces it into a storage space. During discharging, the compressed air flows through an expansion machine or turbine that drives a generator. What is stored is therefore not the electricity itself but a physical state: compressed air in a cavity. It can be pictured as a very large underground balloon that is inflated when electricity is cheap and releases its air through a turbine when electricity is needed.

Two parameters describe the technical design, and they serve different purposes in the power system. The power rating, usually stated in megawatts (MW), indicates how quickly the storage system can charge or discharge. The storage capacity in megawatt-hours (MWh) indicates how much energy it can deliver in total. A compressed air storage system with 100 MW of power and 800 MWh of usable capacity could, in theory, feed in at full power for eight hours. In compressed air storage systems, charging and discharging power can also differ considerably. At the Huntorf power plant, the compressor draws significantly less power than the turbine delivers. Charging there accordingly takes longer than discharging. These key figures are more informative for classification than the mere label "storage," because the various storage technologies differ greatly in power, capacity, cost and siting requirements.

For large plants, underground cavities are mostly used, above all salt caverns created by solution mining. Rock salt is virtually gas-tight, and the technology is familiar from natural gas storage. At Huntorf, two caverns with a combined volume of around 300,000 cubic metres lie at a depth of roughly 650 to 800 metres. Their operating pressure is about 50 to 70 bar. Pressure vessels, tube storage, mined cavities or porous rock layers are also possible. For large amounts of energy, however, they are often more expensive or less proven. The question of location is therefore central: a compressed air storage system depends far more on geology, permitting, grid connection and land area than a modularly built battery storage system.

Compression, heat and efficiency

The central technical question is: what happens to the heat generated during compression? Anyone who pumps a bicycle pump vigorously notices that it gets warm. Conversely, air cools down sharply when it expands. If the heat of compression is lost and new heat has to be supplied during discharging, the efficiency of the overall system drops significantly.

In diabatic compressed air storage systems, the first generation, the heat of compression is largely released to the environment. Before the turbine, the compressed air is reheated, in the existing legacy plants by burning natural gas. One example is Huntorf in Lower Saxony: the plant went into operation in 1978 as the world's first commercial CAES power plant. Uniper operates it today with 321 MW, and the gas turbine reaches its full output within around twelve minutes. For each kilowatt-hour of electricity delivered, the plant uses about 0.83 kWh of electricity for the compressor and around 1.56 kWh of natural gas. Relative to the total energy input, this yields an efficiency of about 42 percent. The second well-known legacy plant, McIntosh in Alabama (110 MW, since 1991), achieves about 54 percent.

This calculation example shows why diabatic plants are not pure electricity storage systems in the narrow sense. If one only compared electricity in with electricity out, more electricity would come out than was put in. The reason is the additional fuel. Such plants are hybrids of storage and gas-fired power plant. For climate balance, fuel costs, fuel dependency and market role, electricity and gas inputs must therefore be reported separately. Uniper has announced that it intends to replace the natural gas at Huntorf with hydrogen in the medium term. This is a statement of intent and not an implemented operating status.

Adiabatic compressed air storage systems (A-CAES or AA-CAES) capture the heat of compression in a heat storage unit, for example in pressurised water, molten salt or solid packed beds. During discharging, they use it to preheat the air again. The plant thus operates without fuel and can achieve a higher electricity-to-electricity efficiency. This does, however, require heat exchangers, high-temperature heat storage and precise operational management. Larger plants of this type are now in operation. In Yingcheng (Hubei Province) in China, the Nengchu-1 plant went online commercially in January 2025. It operates without supplementary firing, uses two salt caverns and has 300 MW of power with about 1,500 MWh of storage capacity. The operator states a conversion efficiency of about 70 percent. This figure comes from the operator and has not been independently verified. Isothermal concepts pursue a different approach: compression and expansion are meant to take place at as near-constant a temperature as possible through intensive heat transfer. So far they are mostly at the development and demonstration stage. The term compressed air storage therefore covers several variants that are not equivalent in energy or economic terms.

Distinction from batteries and pumped storage

Compressed air storage systems are often grouped in the same category as battery and pumped storage systems because all three shift electricity in time. Technically and economically, however, they differ considerably. Battery storage systems respond within fractions of a second, can be built in small units and are well suited to short storage durations, frequency services, local grid support and daily cycles. For long storage durations, however, they become increasingly expensive, because every additional kilowatt-hour requires additional cells.

Pumped storage plants store energy as the potential energy of water. They have high efficiencies and decades of operating experience, but require differences in elevation, water, permits and interventions in the landscape. Compressed air storage systems lie between these categories. Where suitable caverns are available, they can provide large amounts of energy over several hours. Their start-up times in the range of minutes are sufficient for many tasks in the power system, but are considerably longer than those of batteries.

The distinction is also important economically. In compressed air storage systems, the costs of the power components (compressor, turbine, generator, grid connection) scale differently from the costs of the storage volume. If a large cavity is available at low cost, additional storage duration can be cheaper than with technologies that need new, expensive modules for every additional kilowatt-hour. This makes compressed air storage interesting above all for medium storage periods, but not automatically for every storage need.

Significance in the power system

Compressed air storage systems can absorb electricity when wind and solar power are abundant and demand is low. When the residual load rises later, that is, demand minus weather-dependent renewable feed-in, they feed it back in. As shares of fluctuating generation grow, this time shifting becomes more valuable. Storage can reduce curtailment, dampen price peaks and avoid start-ups of conventional power plants, provided market rules and the grid situation allow it. The plant manager at Huntorf once put it bluntly: even 42 percent storage efficiency is better than shutting down wind farms because of a lack of transport capacity.

Nevertheless, a compressed air storage system is not firm generation. It can only deliver what it has stored beforehand, minus losses and, in diabatic plants, plus the fuel energy. Its availability depends on state of charge, technical readiness, grid connection and operating strategy. For security of supply, installed power alone therefore does not count. What is decisive is whether the storage system is sufficiently charged in critical hours and whether its discharge duration matches the duration of the scarcity. Huntorf, for example, can deliver its full power for only a few hours.

A compressed air storage system can also play a role in grid operation. Large plants can absorb or feed in power at certain grid nodes and thereby influence congestion. Uniper states that Huntorf can provide grid services. The grid effect, however, depends on location. An unfavourably located storage system can smooth market prices without relieving a specific grid bottleneck. A storage system placed in a grid-serving location, by contrast, can help make better use of renewable feed-in.

Typical misconceptions

Efficiency equals system value: A lower efficiency means that more charging electricity is needed. Nevertheless, a storage system can make sense if it stores large amounts of energy cheaply, covers rare scarcities or avoids curtailment whose electricity would otherwise go unused. The assessment depends on electricity prices, the CO₂ intensity of the charging electricity, storage duration, number of cycles and alternatives. In addition, efficiency figures must be comparable: for diabatic plants, it must be clear whether the fuel input was counted.
Compressed air equals emission-free: This only applies to plants without fossil supplementary firing and with low-emission charging electricity. Diabatic plants using natural gas are hybrids of storage and thermal generation. Their emissions and fuel costs must be accounted for openly.
Storage as a universal solution: Flexibility can arise from storage, but also from flexible demand, dispatchable power plants, grid expansion, European electricity exchange or sector coupling. Compressed air storage systems mainly solve shifting tasks over several hours. They replace neither fast regulation in every distribution grid nor seasonal energy provision.

Market, regulation and investment conditions

Whether compressed air storage systems get built is not decided by technical feasibility alone. Investors need revenues from price differences between charging and discharging times, from system services, from possible capacity mechanisms or from grid-serving services. If the markets mainly deliver short-term price signals and hardly remunerate long-term availability, the revenue prospects for capital-intensive storage are uncertain. This affects compressed air storage in particular, because permitting, cavern development and construction take considerably longer than for many battery projects. Since Huntorf, no further commercial compressed air storage system has gone into operation in Germany.

From a regulatory perspective, it matters whether storage is treated as a final consumer, as a generator or as a category of its own. Grid charges, levies, connection conditions and redispatch obligations depend on this. In Germany, § 118 para. 6 EnWG exempts facilities for storing electrical energy that were built after 31 December 2011 and are put into operation by 4 August 2029 at the latest from grid charges for the stored electricity drawn, for 20 years. The rule is technology-neutral and thus in principle also applies to new compressed air storage systems, not only to batteries. The exemption does not cover levies and charges other than the grid charge. For facilities that go into operation later, the Bundesnetzagentur intends to redefine grid charges as part of its reform. New storage systems will then presumably be subject to grid charges. As of November 2026, the details had not yet been conclusively determined. For compressed air storage systems with long lead times, the commissioning window until 2029 is hardly achievable. The future legal framework is therefore decisive for new projects.

Compressed air storage systems illustrate a basic characteristic of the future power system: the need for storage arises not only from a lack of energy, but from the fact that generation, consumption and grid capacity diverge in time. The technology can partly offset this mismatch if location, storage duration, efficiency, market rules and grid requirements fit together. A compressed air storage system is therefore neither a general substitute for power plants nor a technical curiosity. It is a site-specific option for shifting large amounts of electricity over hours. Its value only becomes apparent from its concrete task in the power system.

Sources

Energy Transformation Hub Northwest – Compressed air energy storage plant Huntorf — Uniper SE; unbekannt. Abschnitt Compressed air energy storage plant Huntorf
Diplomarbeit zu Druckluftspeicherkraftwerken (Kapitel 4: Grundlagen) — Hochschule für Angewandte Wissenschaften Hamburg; unbekannt. Kapitel 4, Vergleichstabelle Huntorf/McIntosh und Wirkungsgradrechnung
Druckluftspeicher Huntorf: Der Energiewende-Exot — Ørsted (Energiewinde); unbekannt. Abschnitte zu Funktionsweise und Wirkungsgrad
World's largest compressed air energy storage facility commences full operation in China — ESS News; 2025-01-10. Gesamter Artikel
Batteriespeicher: Neues zu Netzanschlüssen und Netzentgelten — Becker Büttner Held (BBH-Blog); unbekannt. Abschnitt Diskussion um Netzentgeltbefreiung (§ 118 Abs. 6 EnWG)

Related glossary entries

Edited by eCMS on 11.11.2026 19:32