Mirjam Sick and Alexander Schwab discuss how pumped storage can be used as an integrator of wind energy in the electrical grid
During the last 10 years the liberalisation of the energy market in Europe has changed the rules for major investment decisions in the electricity market. At the same time there are considerable political efforts to increase the percentage of renewable energy sources in power production. Wind energy has become an important source of renewable energy, and has grown to become the second largest renewable energy source worldwide after hydro power. In some countries wind power has even become as important as hydro. Recent developments in tidal power and wave power will bring a further increase of renewable energy sources in certain regions of Europe. The increase of wind energy in the national electricity grids leads to new tasks with respect to grid stability and stability of power supply. This is due to the stochastic nature of wind energy. There are several means to deal with the intermittency of wind energy: forecasting of wind, overall balancing within a large grid with high transmission capacities, balancing of the grid with thermal power and hydro power stations and storage of energy in pumped storage plants.
Currently the electrical grid is balanced by regulating thermal power stations and hydro power stations to provide additional power within a few minutes such that the need for electricity is met and both voltage and frequency of the electrical grid are kept constant. With the increased percentage of wind power there is more need for balancing power than before. Although it is hard to give a precise figure of how much balancing power is needed with a higher grid penetration by wind power, German utilities clearly state that the need for balancing power has been increased with increased penetration with wind energy. In fact the price for balancing power is high and pays for running thermal power stations at part load which is the main method in the UK with its over capacity of energy production.
Energy storage is an obvious means to balance supply and load of energy. Until now pumped storage is the only mature technology able to store amounts of energy sufficiently large to contribute to grid regulation. Pumped storage plants can be regulated easily and quickly and have a relatively high efficiency (between 65% and 80%).
A newer technology which directly competes with pumped storage is compressed air technology. When there is excess electrical energy a compressor takes air from the environment and feeds it into a reservoir at high pressure. When energy is needed the air is released from the reservoir thereby driving a gas turbine. Until now there are only a few pilot projects installed. The technology is being improved with support of EU funding to reach an efficiency of about 75%.
Major problems of energy storage are the high costs and the energy loss depending on the technology. Generally, transmission is the most efficient way to deal with local excess or lack of electricity. But transmission capacities are restricted, they require high investments for being extended and they cannot be extended without limit for environmental reasons.
The electricity market: major issues
During the last 10 years the energy market in Europe has been rapidly changing in several respects. Firstly, liberalisation has changed the rules for the four major activities in the electricity market: production, grid operation, trade and distribution of power. In most European countries public energy companies have been privatised and unbundled. At least with respect to controlling, these four activities have to be treated separately by each company. Legislation in Europe aims at disestablishment of monopoly in the electricity market even though in some European countries the electricity market is still in the hand of public monopolists. However, grid operation remains a monopoly because it is not feasible to install two competing electricity grids in one region. Trade of electricity has become a major issue within the electricity market.
Until today the production capacity for electricity exceeds the necessary safety limit of 15% overcapacity in most European countries.1 As a consequence, electricity is sold at marginal costs and a fierce competition between power producers takes place.
The most important renewable energy sources are hydro power and wind power. Although hydro is worldwide clearly the leading renewable, in the relatively flat low-lying areas of Denmark and in the North of Germany wind power already delivers more energy than hydro power. For the next 20 years a massive increase of installed capacity is planned in many European countries (Figure 1). Even if the increase will not take place to full extent it will still have a major impact on the European electricity grids. The main technical features that distinguish wind power from traditional generating capacity are its intermittency, its low reliability and problems involved in connecting it to the grid. In its World Energy Outlook 2004 the International Energy Association (IEA) gives some figures for the extra costs of integrating wind energy in the electrical grid:
• Back-up capacity and operational costs: Alternative generating capacity must be available to supply electricity when there is no wind. Because of the extreme difficulty of predicting wind patterns more than 36 hours in advance, the provision of a steady supply of power is complex and costly. Access to alternative flexible resources is necessary. Flexibility has a cost. Assessed cost range: $5 to $10 per MWh.
• Grid costs: Wind turbines are often connected to the grid at low-voltage levels, a practice that may save grid losses but adds to the complexity of system control and operation. Offshore wind farms extend the transmission system to new territories and this adds costs. Assessed cost range: $2.5 to $4 per MWh.
The integration of wind energy in a hydro power system and the required extra effort is described for the Canadian Manitoba region by Hurdowar et al.6
Some dispute is going on questioning whether or not the intermittent nature of wind energy can be completely balanced within a large grid. Theoretically, a very large grid with no limits to transmission capacity is perfectly able to balance supply with load. But no electricity grid is perfect. While supporters of wind energy tend to claim that there is no problem with wind energy in a large grid because its intermittency is levelled out, grid operators report of great difficulties with the intermittency of wind power, see for example Neldner et al.11 However, due to the high investment costs and environmental aspects there are regions in Europe for which new transmission lines are ruled out. For such regions energy storage may become an important option.
Regulations and remuneration schemes
Despite the overall tendency to de-regulate the energy market, the price for electricity produced from renewable energy sources is directly regulated or at least strongly influenced by the regulator.
Wind energy is not yet competitive as an energy source in the electricity market. There is some debate going on about whether or not wind energy is competitive when environmental costs are included into the cost calculation14 but this discussion is not relevant for profitability considerations as long as environmental costs are generally not charged. In order to foster wind energy and to fulfil the targets with respect to the share of renewables in the national grid several remuneration schemes have been installed which directly affect the profitability of wind energy. There are two main systems to directly subsidise renewables: The fixed price schemes with either fixed feed-in tariffs or other tariff regimes, and the quota-based schemes by which a minimum percentage of renewable sources is prescribed to the power producers and failure is penalised.
In a similar way small hydro in most European countries is either indirectly subsidised or the price per kWh is fixed and guaranteed on a relatively high level. For all considerations of profitability it is important to be thoroughly aware of the regulations in each individual country.
Benefits of pumped storage
Pumped storage plants provide balancing power and ancillary services as well as peak shaving capability.
• Fast response: In order to regulate the electrical grid the transmission system operators need to be able to feed in additional electricity within seconds, minutes and hours. For this purpose utilities keep power plants as spinning reserve (dispatchable within seconds). Depending on the time frame in which the energy is dispatchable utilities can charge higher prices. Hydro power stations are dispatchable within minutes and therefore provide highly valuable energy.
• Balancing power: Grid stability with respect to voltage and frequency is of high interest for many industrial consumers because their machines or processes fail if the power supply is not stable. This applies to producers of microprocessors or technical glass, to give two examples. Pumped hydro stations are used by the utilities in order to deliver balancing power. In Germany pumped storage stations are permanently active, either using excess power for pumping or releasing the water through the turbine in order to deliver the contracted balancing power.
• Blackstart capability: The capability to start a power station without the help of the electrical grid is of importance in case of complete failures of the grid. Most types of power plants need supply of electricity in order to start operation. Hydro power plants generally can be started without supply of the electrical grid.
• Commodity storage: Peaks in the energy system are levelled out by using excess electrical power at low price to pump the water in the upper reservoir and by delivering peak energy at times of high demand and at high price
Today, balancing power is supplied within the local or regional grid and there is no significant trade on the electricity market. Typically, pumped hydro stations delivering balancing power are operated with a high frequency of load changes (Figure 2). Technical solutions for pumped storage plants with the purpose of delivering balancing power are either pump turbines with variable speed or separate pumps and turbines which can be operated in closed loop.
Peak energy on the other hand is traded in the international grid. Pumped storage plants which run mainly for commodity storage in order to transform cheap to high price energy can make use of a relatively small price span but need to shift high volumes of water. As the fixed costs of pumped storage schemes are high and the variable costs are low, they are operated at least six hours per day in turbine operation plus six hours per night in pump operation in order to be profitable. Ideally they have very large storage capacities and are permanently loaded.
Energy storage systems
Pumped storage is one of many means to store energy. Energy storage for the supply of electricity is expensive and, in case of many technologies, relatively inefficient. Most energy storage systems are indirect storage systems transforming electricity into another form of energy. Today’s energy storage systems are based on:
• Magnetic energy (Super capacitors)
• Electrical energy (Super conducting Magnetic Energy Storage, SMES)
• Mechanical energy (Fly Wheels, pumped storage, compressed air)
• Chemical energy (Batteries, Reversible Batteries)
Energy storage systems have to fulfill different tasks of the energy management. There are only two energy storage technologies which, due to their large capacities with respect to power output and with respect to the amount of energy, are very well suited for energy management in an electrical grid: pumped storage and compressed air storage. Figure 3 shows that both technologies are able to deliver 100MW and more for several hours, a capacity which today no other technology provides.
At present the efficiency of pumped storage schemes is higher than the efficiency of compressed air storage systems: it ranges from 70% to 80%. But new developments of Compressed Adiabatic Air Energy Storage aim at an overall efficiency of 70% which is similar to the efficiency of a pumped hydro plant, see Bullough et al.3 The comparison of costs of capital per energy unit and power unit respectively shows that compressed air technology requires lower costs of capital, mainly because only one reservoir is needed which means lower costs for civil works. A different picture results from the comparison with respect to capital costs over lifetime: with the long lifetime of pumped hydro the capital costs are lowest compared to all other technologies. Still, the initial capital costs and the long pay back period represent a severe disadvantage for pumped hydro.
Options for pumped storage
Large, centralised pumped storage schemes
The common application for pumped hydro today is a central pumped storage scheme which is operated by a utility, delivering balancing power to the grid or transforming low price electricity into high price electricity. Its working principle is shown in Figure 4. A pumped storage scheme consists of two water reservoirs at different elevations. In times of excess electricity water is pumped from the lower to the upper reservoir. When electricity is needed water is directed from the upper reservoir through the turbine to the lower reservoir and the resulting electricity is fed into the grid.
Pumped storage schemes are used for two main purposes: balancing power and ancillary services; and transforming low price energy into high price energy. With the increase of wind power in the European electrical grids large pumped storage schemes have become very important because of the capability to balance load and stabilise frequency and voltage. It is generally recognised that the potential for a further increase of hydro power stations in Europe is strongly limited, but there is still potential for pumped storage schemes which will be further exploited in the near future.
Even in the Alpine region of Europe where hydro power is exploited to a large extent there is still potential for the extension of existing storage schemes into pumped storage plants. A recent example is the Kops II extension scheme: The Austrian power producer Vorarlberger Illwerke is extending an existing hydro storage scheme into a pumped storage scheme. The purpose of the Kops II scheme is to deliver balancing power to the German utility EnBW. The existing hydro power plant will be extended by three pumps and three Pelton turbines, the output of each machine being 150MW, with generator motors, piping and civil works. The two reservoirs already exist. Total costs of the scheme will be 330M Euro (US$395M) including civil engineering. Technically, the solution is interesting because the pumps and the turbines are separated such that they can be operated in a closed loop. This is necessary in order to deliver or use any amount of energy between zero and 100% of the maximum output at highest possible efficiency and thus serve the purpose of balancing the electrical grid.
Small, decentralised pumped storage schemes
In contrast to large central pumped storage units small pumped storage schemes (output < 20MW) generally cannot deliver balancing power and other ancillary services because they are too small. Consequently, the economics of small pumped storage relies entirely on the price span of electricity and the duration of high price periods and low price periods respectively. In a recent investigation Figueiredo et al.5 compare the economics of pumped storage for 14 different deregulated markets worldwide based on the respective hourly price data for electricity. It is shown that in 7 out of 14 markets only an investment into pumped storage of US$300 per kW gives a return on investment of 10% or more. As the price per kW investment into small pumped storage is relatively high, this investigation would lead to the conclusion that in many markets there is no point in investing in small pumped storage.
Small pumped storage schemes are attractive if the infrastructure already exists and if they fulfil multiple purposes. In certain regions in Europe the transmission capacities are not sufficient but cannot be extended mainly for environmental reasons. Decentralised pumped storage schemes relieve the transmission problem by delivering additional power during the peak times and storing energy during the night. The value thus is not only peak shaving but also preventing the need for more transmission capacity.
Further possible applications are: existing water reservoirs in ski resorts or river locks. Here, all the civil engineering is already done which means that the additional investment is far lower. In addition, such schemes always serve another primary purpose than energy storage and energy production.
Coupled wind – hydro schemes
As the irregular output of a wind farm is the root cause of the grid problems the direct coupling of a wind farm with a storage system seems to be an obvious answer. Consequently the third option presented and discussed in this study is the coupling of a pumped storage plant with a wind farm as illustrated in Figure 5. Instead of delivering the wind energy directly to the grid this scheme optimises the electricity supply beforehand.
Several studies have been carried out in order to determine the economics and profitability of wind-hydro schemes. Castronuovo4 finds an optimal layout for the pumped storage scheme based on wind data and the remuneration scheme in Portugal. He shows that without a storage scheme 57% is sold at high price while the storage scheme enables 65% of the energy to be sold at high price.
Korpås et al.10 report about the simulation of a 10MW wind farm which is operated together with a storage system, participating in the spot market. Income is generated by sales of electricity at spot price varying from $18/MWh to $42/MWh, which is found to be insufficient to profitably operate a wind-hydro scheme. Bose et al.1 have developed a tool to optimise wind-hydro schemes for island grids which delivers the respective generation costs for each variation, showing that generation costs of wind-hydro schemes tend to be higher than those of wind only schemes. They calculate that generation costs for wind-hydro schemes range between 60€ per MWh (US$72) and 110€per MWh (US$132).
There are some initiatives for wind-hydro schemes on Greek islands as well as on El Hierro (Canary islands). Given the enormous costs of electricity generation using diesel oil in diesel generators as reported by Kaldellis et al.,9 the profitability of a wind-hydro scheme in comparison to a diesel generator is not in doubt. Furthermore, a wind farm without storage scheme is not an option for a small island because of grid stability issues. Still, the following problems remain: high initial investment of wind-hydro plants; long pay back period
The El Hierro project11 is strongly supported by the EU and will serve as demonstration case for other Canary and European islands:
Today, most of the electricity supply of the Canary island El Hierro (about 10,000 inhabitants) is covered by a conventional diesel power plant. El Hierro, a declared Unesco Biosphere Reserve, is working on a project for 100% renewable supply of electricity. At the heart of the energy system is a combined wind – hydro scheme with a desalination plant which is used to fill the reservoirs and replace the evaporated water.The wind-generated energy surplus can be accumulated by desalinated water. In order to stabilise the electrical grid the hydro plant will consist of a set of Pelton turbines and a set of pumps which can be operated in closed loop. Installed wind capacity is planned to be 15MW. The system should be operative in 2008.
Pumped storage has an important role in the electrical grid both for grid stabilisation and for peak shaving. The increased share of wind energy in the European electrical grid can be managed most efficiently with the help of pumped storage schemes because of their fast response and high efficiency. Consequently, the use of large pumped storage schemes for balancing power and ancillary services has gained in significance leading to new extension schemes, especially in the Alpine region of Europe.
Small pumped storage plants can be used for energy storage and peak shaving but not for balancing power. The profitability of small pumped storage schemes depends very much on the individual market conditions and on multi-functionality. Studies show that small pumped storage schemes with no other purpose other than energy storage and peak shaving are profitable in few markets only. But small pumped storage plants which fulfil additional purposes may very well be profitable.
A small pumped storage scheme directly coupled to a wind farm can be used to regulate the power output of a windfarm. So far all studies show that this solution is less profitable than energy production in a wind farm alone. The main reasons for this are the relatively high investment costs for a small pumped storage scheme and the relatively low degree of utilisation. But in the case of remote island, wind farms need to be operated in connection with a storage scheme for grid balancing purposes. Compared to diesel power stations which are most commonly used for electricity production on small islands a wind-hydro scheme is the more economical solution. A demonstration project on the Canary Island El Hierro will provide valuable experience with this new technology in the near future.
The Authors are Mirjam Sick, VA Tech Hydro, Hardstrasse 319/P.O. Box, 8023 Zurich Switzerland, email@example.com; and Alexander Schwab, VA Tech Hydro, Penzingerstrasse 76, 1141 Vienna, Austria, firstname.lastname@example.org
This article is based on a paper presented at Hydro 2005, Villach, Austria (17-20 October), organised by Aqua-Media International