Connecting countries to climate technology solutions
English Arabic Chinese (Simplified) French Russian Spanish Yoruba

Pumped hydroelectric energy storage

Technology group:

Next to the other energy storage technologies, such as phase change materials, batteries and CAES, pumped hydro is another option for energy storage. Pumped hydro storage uses two water reservoirs which are separated vertically. In times of excess electricity, often off peak hours, water is pumped from the lower reservoir to the upper reservoir. When required, the water flow is reversed and guided through turbines to generate electricity.

Pumped hydro is the most developed energy storage technology, with facilities dating from the 1890s in Italy and Switzerland. Currently, there is over 90 GW of pumped storage in operation world wide, which is about 3 % of global generation capacity (CPUC, 2010). The main applications of pumped hydro are for energy management, frequency controal and provision of reserve (CPUC, 2010). Pumped storage plants are characterized by long construction times and high capital expenditure. However, with rising electricity prices and an increasing use of intermittent energy sources, it can be very economic to store electricity for later use.


Pumped hydro storage uses two water reservoirs which are separated vertically. In times of excess electricity, often off peak hours, water is pumped from the lower reservoir to the upper reservoir. When required, the water flow is reversed and guided through turbines to generate electricity. This operating principal is illustrated in a simple way in Figure 1.

Feasibility of technology and operational necessities

Conventional pumped hydro storage

Conventional pumped hydro uses two water reservoirs, separated vertically. During off peak hours water is pumped from the lower reservoir to the upper reservoir. When required, the water flow is reversed to generate electricity. Some high dam hydro plants have a storage capability and can be dispatched as a pumped hydro. Underground pumped storage, using flooded mine shafts or other cavities, are also technically possible. Open sea can also be used as the lower reservoir. A seawater pumped hydro plant was first built in Japan in 1999.

While pumped hydro is often performed in geologically appropriate areas, in other words areas which have high elevation differences, pumped hydro can be constructed in such a way that it is applicable anywhere where water is available. An example of such a construction is shown in Figure 2. However, the construction of such facilities is more expensive as absence of the natural vertical difference has to overcome due to the artificial construction of vertical difference.

Pumped hydro with variable speed turbines for bulk storage

A variation on the conventional pumped hydroelectric plant is based on the use of adjustable- or variable-speed turbines. No adjustable-speed pumped hydro plants exist in the U.S. This is because the last pumped hydro plant constructed in the U.S. was completed in 1995. At the time this plant was being designed, the variable-speed turbine and associated electrical and electronic equipment needed was still under development. The main effort to bring this technology to market has been in Japan. After several model plants were constructed and tested in Japan, a full-scale developmental installation was designed in the late 1980s and was completed in 1993. The 395-MW variable-speed turbine and associated motor generator were designed and built by Hitachi and installed in Kansai Electric’s Ohkawachi No. 2 unit. An adjustable-speed pumped hydro plant is based on a turbine and motor generator that will operate over a range of rotation speeds that is 10% above and below nominal. This speed variation is accomplished by using a solid rotor with a three-phase winding that can be driven at a variable frequency. A cycloconverter changes AC power to the appropriate frequency via an AC-DC-AC converter. The stator, as in a conventional pumped hydro plant, is directly connected to the three-phase bus and is energized at line frequency. The modest ±10% change in rotation frequency has a significant impact on performance. The 20% variation in rotational velocity translates into a 50% change in power output because mass flow is related to the third power of the rotational speed.

An adjustable-speed pumped hydro plant has several advantages:

  1. There is no need for a pony motor to start pumping.
  2. Synchronous power operation covers a wider range than for a conventional turbine, from below 70% to full power.
  3. The rate of changing the power output is driven by inertia of the water flow and is much more rapid than conventional pumped hydro.
  4. Full power output can be delivered from water-head variations of a factor of two.
  5. Rotational speed can be adjusted to avoid resonances within the equipment and cavitation modes in the water flow. This leads to longer life and less maintenance.
  6. Higher overall efficiency, as high as a 0.3 improvement on an annual basis is possible.
  7. Speed regulation provides frequency regulation to the grid.

However, improved performance comes with extra design and construction issues and costs:

  1. Motor/generator balance is more critical and takes multiple adjustments prior to initial operation.
  2. Higher speed rotation >350 rpm is required to obtain maximum performance.
  3. The motor/generator/turbine shaft is longer, which requires more excavation, increasing civil engineering costs.
  4. There is some additional cost for the cyclo-converter, but it is partially offset by the electronic conversion needed for starting conventional plants.

Several of these plants are now installed in Japan and others are in design or under construction. The general philosophy of recent pumped-hydro plants has been to install equal numbers of adjustable speed turbines and conventional turbines. This approach seems to provide adequate flexibility for operation over a wide range of pumping and generating scenarios (Schoenung and Hasselzahn, 2002).

Typically, upper and lower reservoirs have been constructed by dams. Within the last 10 years, advanced pumped storage (APS) technology has been developed to improve speed, reliability and efficiency. These plants are designed hydraulically and mechanically for ultrafast loading and ramping, allowing frequent and rapid (<15 sec) changes among the pumping, generating and stand-by spinning modes (Schoenung, 2001).

Operational necessities

Hydroelectric power requires a considerable volume of water to produce energy. The following equation describes the relationship between the volume of water (V, in cubic meters), the stored energy (E, in kWh), and the average head driving a turbine (h, in meters), and assumes 0.90 efficiency in energy conversion for electricity production:

V(m3)≈ 400 * (E(kWh)/h(m)) (Schoenung and Hassenzahl, 2002)

Using this relationship, a reservoir one kilometer in diameter, having an average head of 200 meters, and holding enough water for 10,000 MWh, would be filled to a depth of 25 meters. These are large installations and require considerable planning as well as environmental and other permits.

Status of the technology and its future market potential

Development status of pumped hydro compared to other energy storage technologies

The 2007 APS study identified the development status of several key energy storage technologies. Of the possible energy storage technologies, pumped hydro is identified as the most mature technology. Currently, over 90 GW is in operation world wide, which is approximately 3 % of global generation capacity (CPUC, 2010). This places pumped hydro technology in the commercial stage of development.

Table 1. Development status of several key energy storage technologies. Source: APS, 2007
Commercial Pre-commercial prototype Demonstration stage Developmental stage
Pumped hydro

Flywheels for power quality applications at the consumer site


Lead-acid battery

NI-Cad battery

Flywheel (as load device)

Micro - SMES (as load device)

Zinc-bromine battery

Flywheel (as grid device)

Vanadium redox battery

Electrochemical capacitor

Lithium-ion battery for grid applications

SMES (as grid device)

Electrochemical capacitors

Other advanced batteries

Potential of pumped hydro

While no new pumped storage capacity has been added between 1990 and 2000 (IEA, 2009), approximately 9 GW of new storage capacity is planned in the European Alps (Schwab, 2007). Globally, there is potential for about 1000 GW of pumped storage capacity, equal to about half of all realistic hydropower potential (Taylor, 2007). This is about ten times the current pumped storage capacity (IEA, .

New novel approaches to pumped hydro storage increase the potential of the technology. A novel example of the application of pumped storage is the recent commissioning of the NorNed cable. The cable interconnects the electricity grids of the Netherlands and Norway and is a submarine high voltage cable. The cable enables transportation of electricity from the Netherlands to Norway and vice versa. With a total length of 580 kilometres, the NorNed cable has a capacity of 700 megawatts (MW) – enough to supply power to half of Amsterdam or Oslo.

The cable is used to transmit electricity to Norway from the Netherlands. In turn, this electricity is used in pumped hydro storage facilities to pump water to elevated levels. Norway and the Netherlands complement each other when it comes to the production and consumption of electricity. In the Netherlands, less electricity is consumed at night than during the day, while the energy consumption in Norway is relatively high during the night (Tennet, no date). Figure 3 illustrates the energy mixes of the two countries and how the energy demand patterns of the two countries are throughout the year.

Interconnecting countries through high voltage cables with pumped storage capacity in other countries results in a higher efficiency use of the pumped storage capacity. As such, the expected increasing interconnectedness of electricity grids allows existing pumped storage facilities to be used more effectively.

Potential of pumped hydro to address intemittency of renewable energy technologies

Pumped hydro has large potential to be used as an electricity storage medium for intermittent renewable energy technologies. Originally, pumped hydro storage has been used for off-peak energy storage stemming from coal and nuclear power plants to sell in high-peak demand, and thus generate revenue.

However, the increasing use of renewable energy technologies, such as wind and solar energy, opens new opportunities for the pumped storage technology. Due to the intermittent nature of wind and solar energy, electricity is often delivered when it is not needed. Pumped hydro storage technology would be able to store the excess electricity for later use. Such an application would support the penetration of renewable energy technologies.

Novel new ideas for pumped storage might be able to use this potential. The new idea presented in Figure 2, for instance allows pumped hydro storage to be applied irrespective of geologically available vertical differences. Applying this method near intermittent sources enhances both the market potential of the pumped hydro energy storage technology as well as the market potential of renewable energy sources. Another novel new idea for pumped hydro storage is illustrated in Figure 4. The energy island illustrated in Figure 4 combines the intermittent energy sources wind and solar energy with pumped hydro storage technology to more effectively deploy wind and solar energy. Constructing these energy technologies at the same location supports the effectiveness of both technologies. Moreover, through the grid connection with the mainland, the energy island can also draw energy from the grid to fill the lake to allow peak shaving of the electricity grid.

How the technology could contribute to socio-economic development and environmental protection

Energy storage can reduce costs for consumers of electricity. In general, off-peak electricity is cheaper compared to high-peak electricity. This is due to the base-load characteristics of off-peak electricity. Energy storage in the form of pumped hydro provides customers with off-peak electricity in high-peak situations. Next to being an economic benefit to the seller of the electricity, it might also be an economic benefit to the customer when the electricity is sold at lower prices compared to the high-peak generated electricity.

A California Public Utilities Commission (CPUC) study (2010) identifies the following factors that contribute to economic development through the use of pumped storage technology: a) reduced need for peak generation capacity; b) employment and other economic growth; c) reliable and cleaner back-up power; d) transmission support and congestion relief; e) more efficient use of renewable and off-peak generation capacity; f) Reduced need for transmission and distribution capacity upgrades; and g) Increased and improved availability of ancillary services.

a) By allowing customers, utilities or power generators to store energy off-peak and discharge on-peak, storage provides an alternative to the construction and operation of new generation and reserve capacity. As an example, the CPUC illustrates that peak demand growth is a major concern in California which is exacerbated by the fact that populations in the hotter central and southern parts of the State are growing fastest. The value of the avoided cost of peak generation capacity will continue to increase as peak demand grows and as carbon emissions become more expensive.

b) As more storage is deployed, new jobs could be created in manufacturing and installation, boosting the economy and providing a new source of tax revenue.

c) Pumped hydro can provide customers with electricity for a period of hours when utility power is not available. While economics prevent energy storage to be used as a long-term backup, i.e. for multiple days at once, pumped hydro can provide a source of back-up power for shorter outages.

d) Transmission support and congestion relief: pumped hydro can be used to improve transmission and distribution performance by alleviating problems like voltage sag and unstable voltage. In addition, pumped hydro can help to avoid transmission congestion by discharging in congested areas at times of peak demand.

e) More efficient use of renewable and other off-peak generation: pumped storage can provide energy market support to renewable energy technologies. Pumped hydro can be used to prevent a mismatch between load and generation in the off-peak hours. Rather than forcing renewable generators to curtail off-peak production, energy storage can allow excess wind and other off-peak energy to be stored and used during high demand times.

f) Reduced need for transmission and distribution capacity upgrades: Similar to the reduced need for additional power plants, energy storage can be used to maximixe existing transmission and distribution resources. Energy storage technologies in general can delay the need for new transmission and distribution capacity upgrades by shifting demand off-peak.

g) Increased and improved availability of ancillary services: Ancillary services are services necessary to support the transmission of energy from generation resources to consumers, while maintaining the reliable operation of the transmission system. For example, there are two primary types of ancillary services sold in California, both of which could be provided by energy storage technologies: frequency regulation, which ensures the grid operates within an allowable range of interconnection frequencies, and operating reserves, which ensure that more energy can be added to the system within a short period of time to meet unexpected increases in demand or reductions in supply. This can also be an economic benefit to the energy storage system as it sells the electricity into the ancillary market.


Lower GHG and other emissions: EES can reduce emissions by shifting on-peak energy use to off-peak periods. In California, relatively little baseload power comes from coal and much comes from hydroelectric and nuclear power, such that off-peak generation generally has a cleaner emissions profile than largely gasfired peak power. However, as renewables like wind increase as a percentage of the off-peak power mix, the emissions benefits of EES will continue to grow. (CPUC, 2010)

EES is also a lower-emissions alternative for providing ancillary services. A study by KEMA found that regulation provided by a 20 MW flywheel EES system created less than half the GHG emissions of equivalent regulation from a combined cycle gas turbine and less than three quarters of the emissions of a pumped hydro plant providing equivalent regulation. (CPUC, 2010).

Financial requirements and costs

The California Public Utilities Commission (CPUC) identifies several general factors that influence the economics of energy storage technologies. First, storage tends to be an application specific resource and therefore the costs and benefits can vary greatly between technolgies. One of the complications in developing detailed cost estimates of energy storage technologies is that the costs of a technology are greatly influences by the particular application in which that technology is deployed. From this the CPUC concludes that any generalized cost estimates are of questionable value. Second, an energy storage system varies on two dimensions: power (how much electricity can be discharged at one time) and energy (how many hours can be discharged continuously). Energy storage system costs are impacted by system efficiency. In addition, the frequency of how often and deeply the system is charged and discharged also impacts costs. From this, the CPUC concludes that an energy storage technology's cost cannot be meaningfully estimated independently of the way in which it is used. third, the lifecycle cost of an energy storage system is made up of two basic components: capital costs and operating and maintenance costs (O&M). Energy storage O&M costs include the cost of buying the energy used to charge the system, fixed costs that do not depend on how much or often the system is used, and variable costs, the bulk of which are usually replacement costs.

While general cost estimates are of questionable value as to the accuracy of the cost of the system, such an analysis does indicate the relative cost of the technology compared to the other options of energy storage. Considering pumped hydro, this technology's efficiency and costs depend on a variety of factors. Schoenung and Hassenzahn (2002) identify the head of water, the civil costs of excavation, tunneling and dam building as several of these factors. An average value for the power-related part of installations under construction today is $1000/kW, while the cost of the storage component is relatively inexpensive, at about $10/kWh. The typical round-trip efficiency of large plants is about 0.75. This technology has been the primary type of energy storage for utilities to date. Today, however, only a few locations exist where adequate water and sites for upper and lower reservoirs are available. (Schoenung and Hasselzahn, 2002)

Schoenung and Hasselzahn (2002) conducted a life cycle analysis (LCA) on various energy storage technologies, including pumped hydro storage. Image X, XX and XXX show the results of their analysis.

Revenue requirements for bulk energy storage systems are shown in Figure 6. These represent the cost of the energy served, or the minimum price for energy that would cover all expenses to generate that energy. The costs all decrease for larger systems.

CAES is shown to be the least expensive technology, followed by pumped hydro. This is primarily because of the low cost of storage. Pumped hydro with variable speed drive is slightly more expensive than conventional pumped hydro, even though it is more efficient. This is offset by large benefits due to greater operational flexibility.

On a component basis, it can be seen that CAES and pumped hydro have relatively low costs. This is partially due to the lack of replacement costs for the installation. For instance, batteries have components which have a maximum number of charges after which capacity losses occur. In the study by Schoenung and Hassenzahn (2002) it is assumed that pumped hydro and CAES have no replacement costs throughout the lifetime of the installation.


  • EPRI, 2002. Handbook for Energy Storage for Transmission or Distribution Applications. Report No. 1007189. Technical Update December 2002. Document can be found at:
  • IEA, 2008. Energy Technology Perspectves: Scenarios and Strategies to 2050.
  • NETL, 2008. Market Analysis of Emerging Electric Energy Storage Systems. National energy technology laboratory and Department of Energy report with code DOE/NETL-2008/1330 of July 31, 2008.
  • APS, 2007. Challenges of Electric Energy Storage Technologies: A Report from the APS Panel on Public Affairs Committee on Energy and Environment.
  • CPUC, 2010. Electric Energy Storage: An Assessment of Potential Barriers and Opportunities. A Policy and Planning Division White Paper of the California Public Utilities Commission.
  • Schoenung, S., M., & Hassenzahn, W., V., 2002. Long- vs Short-Term Energy Storage Technology Analysis: A life cycle cost study. A study for the Department of Energy (DOE) Energy Storage Systems Program. Document can be found online at: [[1]]
  • Schwab, A. (2007), “Pumped Storage: Innovation and Demand”, paper presented at the International Hydropower Association (IHA) World Congress on Advancing Sustainable Hydropower, May 29-31 2007, Antalya, Turkey.
  • Taylor, R. (2007), Hydropower Potentials, International Hydropower Association.
  • Schoenung, S., M., 2001. Characteristics and Technologies for Long- vs Short-Term Energy Storage. A study by the Department of Energy (DOE) Energy Storage Systems Program. Document can be found online at: [[2]]
  • Tennet, no date. NorNed: Europe's link for the future. Document can be found online at: [[3]]
  • Lievense, no date. Image retrieved from: [[4]]