6.

15 Pumped Storage Hydropower Developments

T Hino, CTI Engineering International Co., Ltd., Chu-o-Ku, Japan
A Lejeune, University of Liège, Liège, Belgium
© 2012 Elsevier Ltd. All rights reserved.

6.15.1 Inroduction 406

6.15.2 Electrical Energy Storage 406

6.15.2.1 General Issues 406

6.15.2.1.1 Benefits of storage 406

6.15.2.1.2 Barriers to the deployment of electrical energy storage 406

6.15.2.1.3 Location of storage systems 407

6.15.2.2 Applications 407

6.15.2.2.1 Load management 408

6.15.2.2.2 Spinning reserve 408

6.15.2.2.3 Transmission and distribution stabilization and voltage regulation 408

6.15.2.2.4 Transmission upgrades deferral 408

6.15.2.2.5 Distributed generation 408

6.15.2.2.6 Renewable energy applications 409

6.15.2.2.7 End use applications 410

6.15.2.2.8 Miscellany 410

6.15.2.3 Storage Technologies 410

6.15.3 Pumped Storage Hydropower Plant 412

6.15.3.1 Characteristics 412

6.15.3.2 History 412

6.15.3.3 Characteristics of Pump–Turbines 413

6.15.3.3.1 Elements of pump–turbine hydraulic design 413

6.15.3.3.2 Elements of pump–turbine mechanical design 416

6.15.4 Examples of Remarkable Pumped Storage Power Plants 418

6.15.4.1 Okinawa Seawater Pumped Storage Power Plant 418

6.15.4.1.1 Outline of the project 419

6.15.4.1.2 Features of the project area 420

6.15.4.1.3 Major impacts 421

6.15.4.1.4 Mitigation measures 422

6.15.4.1.5 Measures during construction 422

6.15.4.1.6 Permanent measures 423

6.15.4.1.7 Results of the mitigation measures 424

6.15.4.1.8 Reasons for success 426

6.15.4.2 Goldisthal Pumped Storage Power Plant 426

6.15.4.2.1 Introduction 426

6.15.4.2.2 Developing the Goldisthal project 426

6.15.4.2.3 Main features of the project 427

6.15.4.2.4 Choosing variable-speed machines 427

6.15.4.2.5 Operation to date 428

6.15.4.3 Tianhuangping Pumped Storage Power Plant 429

6.15.4.3.1 Introduction 429

6.15.4.3.2 Two pump storage reservoirs 429

6.15.4.3.3 Benefits of the project 430

6.15.4.4 Coo-Trois Ponts Pumped Storage Power Plant 431

6.15.4.4.1 Introduction 431

6.15.4.4.2 Main features of the project 431

6.15.4.4.3 Generating equipment 432

6.15.4.4.4 Special features of interest 433

References 434

Comprehensive Renewable Energy, Volume 6 doi:10.1016/B978-0-08-087872-0.00616-8 405

406 Design Concepts

6.15.1 Inroduction

Pumped storage hydroelectric projects differ from conventional hydroelectric projects. They store electrical energy normally by
pumping water from a lower reservoir to an upper reservoir when demand for electricity is low. Water is stored in the upper reservoir
for release to generate power during periods of peak demand. These projects are uniquely suited for generating power when demand
for electricity is high and for supplying reserve capacity to complement the output of large fossil-fueled and nuclear steam electric
plants. Start-up of this type of plant is almost immediate, thus serving peak demand for power better than fossil-fueled plants do,
which require significantly more start-up time. Like conventional projects, they use falling water to generate power, but they use
reversible turbines to pump the water back to the upper reservoir. This type of project is particularly effective at sites having high
heads (large difference in elevation between the upper and the lower reservoir) [1].

6.15.2 Electrical Energy Storage

6.15.2.1 General Issues
The concept of electrical energy storage has become a controversial issue in recent years. Many questions are raised in the electricity
sector: Why is energy storage needed? What are the alternatives? How much do storage systems cost, and how much added value
does a storage system provide? Will storage contribute to the increased utilization of renewable sources?
The storage issue must be viewed in the frame of a changing electricity sector characterized by

• restructuring of the electricity market;

• growth in new/renewable energy sources;
• increasing reliance on electricity and demand for higher-quality power;
• move toward distributed generation; and
• more stringent environmental requirements.

As part of these changes, there are growing pressures to operate the electrical network more efficiently while still maintaining high
standards of reliability and power quality. Accommodation of renewable generation and ever more stringent environmental
requirements are combining strongly to further influence electricity companies’ decisions on how they should be developing
their future network designs. With these driving forces as a backdrop, the rapidly accelerating rate of technological development in
many of the emerging electrical energy storage technologies, with anticipated system cost reductions, now makes their practical
application look attractive.
Energy storage is not a new concept in the electricity sector. Utilities across the world have built a number of pumped hydro
facilities in the last few decades, resulting in a storage component of roughly 5% of the power generation capacity of all the
European countries, which is 3% in the United States, and 10% in Japan. These pumped hydro plants, and to a lesser extent
compressed air storage systems, have been used for load leveling, frequency response, and voltage/reactive control. Likewise, storage
facilities based on other technologies such as lead–acid batteries have been installed by a number of utilities to fulfill a variety of
functions. At a different scale, energy storage is also commonly used at the user level to ensure reliability and power quality for
customers with sensitive equipment. Another traditional application is the electrification of off-grid networks and remote tele­
communications stations, mostly in connection with renewable sources. The market penetration achieved by electrical energy
storage to date has been heavily constrained by its cost and the limited operational experience, resulting in high technical and
commercial risk. However, the presence of storage systems is growing fast owing to the circumstances mentioned above.

6.15.2.1.1 Benefits of storage

Storage contributes to optimizing the use of existing generation and transmission infrastructure, reducing or deferring capital
investment costs. It contributes to integrating renewable sources (and in general distributed sources) into the system, enhancing
their availability and market value. The environmental benefits must be highlighted, in terms of both reduction of emissions from
conventional power plants and increase of renewable sources’ penetration. Energy storage facilities can also help maintain
transmission grid stability by providing ancillary services, including black start capability, spinning reserve, and reactive power.
At the consumer level, storage improves power quality and reliability, and can provide capability to control or reduce costs.
Energy storage is of growing importance as it enables the smoothing of transient and/or intermittent loads, and downsizing of base
load capacity with consequent substantial potential for energy and cost savings.
However, it is acknowledged that energy storage systems will have to compete within the context of present overcapacity of
power stations and power generators with short start-up times, such as open-cycle gas turbines and gas or diesel motors with
appropriate emission controls.

6.15.2.1.2 Barriers to the deployment of electrical energy storage

Electrical energy storage involves significant investment and energy losses, which must be weighed against the benefits and
compared to other nonstorage solutions. There are a number of key barriers to a more widespread use of storage systems:
 Pumped Storage Hydropower Developments 407

• Immaturity of some technologies and lack of operating experience. More demonstration projects are needed to gain customers’
confidence. Further research and development is necessary in some aspects such as the implementation of power conditioning
and control process for a multi-application energy storage system.
• High initial capital costs. Technological advances and large manufacturing volumes will bring these costs down.
• Uncertainty over the quantified benefits. This is true especially when, as usually happens, there are multiple different benefits
associated with a storage system.
• Uncertainty over the regulatory environment. The future shape of the electricity market, in relation to not only energy trading but also
ancillary services trading, will affect decisively the viability of electrical energy storage. The use of storage systems for the provision
of ancillary services currently provided by the system operator will depend on the deregulatory process.

6.15.2.1.3 Location of storage systems

Utility-scale energy storage systems are envisaged as forming an integral part of the future energy system. Depending on the
application, they can be implemented in any of the different segments of the electricity supply system (Figure 1). In a liberalized
market, the different segments of the electricity sector are increasingly being separated. Each segment offers different potential
opportunities to energy storage applications. Correct location of the storage systems is important to maximizing the benefits.
Large-scale, that is, multimegawatt, centralized storage could improve generation and transmission load factors and system stability.
Smaller-scale, localized, or distributed storage could deliver energy management and peak-shaving services, as well as improving
power quality and reliability. Distributed storage would be an ideal complement to distributed generation, especially on account of
the increasing levels of renewable energy generation.
One of the axioms of energy storage is that storage units should be located as close as possible to the end consumer of electricity.
This is because the storage device can improve the utilization of all components in the network. In order to place a storage device
close to the end consumer, the device would need to be matched for both power and energy storage capacity to the requirements of
the consumer. Since the specific capital cost increases as the system becomes smaller, the optimum position for a storage device in
the network tends to move closer to the generation source. For this reason, Price [2] maintains that many storage systems can, and
should be, located near to substations or grid distribution points. When storage systems are utilized to facilitate renewable energy
source integration, the picture changes, however, since the fluctuations in the generated power are usually greater than those in the
load. As a result, the optimum location is likely to be close to the generation points, thus maximizing the capacity of the
transmission and distribution lines.

6.15.2.2 Applications
Applications of electrical energy storage are numerous and varied, covering a wide spectrum, from larger-scale generation- and
transmission-related systems to smaller-scale applications at the distribution network and the customer/end use site. This chapter
deals specifically with the application of storage for renewable energy source integration; however, this is closely connected to other
applications. Storage systems usually provide multiple benefits, and thus it is necessary to review all their possible functions.
Interesting reviews of the applications can be found in Schoenung [3], Herr [4], and Butler [5]. Ultimately, the purposes of all these
applications come down to

• improved load management

• provision of spinning reserve
• transmission and distribution stabilization and voltage regulation
• transmissions system upgrade deferral
• facilitating distributed generation
• facilitating renewable energy deployment
• end use applications
• miscellaneous (including ancillary services)

Bulk Distributed
Transmission Distribution
generation generation

Storage End use

Figure 1 Storage locations in the electricity supply system.

408 Design Concepts

6.15.2.2.1 Load management

Load management includes the traditional ‘load leveling’, a widespread application for large energy storages, in which cheap
electricity is used during off-peak hours for charging, while discharging takes place during peak hours, providing cost savings to the
operator. In addition, load leveling can lead to more uniform load factors for the generation, transmission, and distribution
systems. Although load leveling was the first application of energy storage that utilities recognized, the difference in the marginal
cost of generation during peak and off-peak periods for many utilities is moderate. Therefore, Butler [5] concludes that load leveling
is likely to be a secondary benefit derived from an energy storage system installed for other applications that offer greater economic
benefits. Load leveling requires energy storage systems on the order of at least 1 MW and up to hundreds of megawatts, and with
several hours of storage capacity (2–8 h). For utilities without a strong seasonal demand variation, a system used for load leveling
would operate on weekdays (250 days per year). Other types of load management are ‘ramping and load following’, in which energy
storage is used to assist generation to follow the load changes. Instantaneous match between generation and load is necessary to
maintain the generators’ rotating speed and in turn the frequency of the system. Storage systems serving this application should be
able to deliver on the order of 10–100 MW to absorb and deliver power as demand fluctuates. The system would have to be able to
dispatch energy continuously, especially during peak-load times, in frequent, shallow charge–discharge cycles that would occur.
This service is usually provided by conventional generation.

6.15.2.2.2 Spinning reserve

The category ‘fast-response spinning reserve’ corresponds to the fast-responding generation capacity that is in a state of ‘hot standby’.
Utilities hold it back to be put in use in case of a failure of generation units. Thus, the required power output for this application is
typically determined by the power output of the largest unit operating on-grid. The conventional spinning reserve requires a less
quick response. Storage systems can provide this application in competition with standard generation facilities. Since the power
plants that they would temporarily replace may have power ratings on the order of 10–400 MW, storage systems for reserve must be
in this same range. Generation outages requiring rapid reserve typically may occur about 20–50 times a year. Therefore, storage
facilities for rapid reserve must be able to address up to 50 significant discharges that occur randomly through the year.

6.15.2.2.3 Transmission and distribution stabilization and voltage regulation

Transmission and distribution stabilization are applications that require very high power ratings for short durations in order to keep
all components of a transmission or distribution line in synchronous operation. This includes phase angle control, and voltage and
frequency regulation. In the event of a fault, generators may lose synchronism (due to difference in phase angle) if the system is not
stabilized, making the systems collapse. Energy storage devices can stabilize the system after a fault by absorbing/delivering power
from/to the generators as needed to keep them turning at the same speed. Fast action is essential for quick stabilization. Response
time limitations demand an appropriate power-conditioning interface designed to ensure a reliable mitigation of short-duration
electrical disturbances, which can range from a couple of cycles to 2 min. The portability of the storage systems might be an
important factor in many cases. Some applications are temporary in nature, and Boyce [6] points out that for a storage system the
ability to be transferred from site to site can significantly increase its overall value. With the liberalization of the electricity market
there will be an increasing need to maintain and to improve the stability of the electrical grid. The risk of voltage instability, being
the source of failures in automatic production centers and the base of cascading outages, will become more and more serious. Many
of the utility grids cannot properly react to transient events with the limited transmission capacity they have. In cases of
fast-changing load flow patterns or changes in the distribution of the loads or power plants on the grid, the risk of voltage instability
increases. To offset the effect of impedance in transmission lines, utilities inject reactive power and maintain the same voltage at all
locations on the line. Traditionally, fixed and switched capacitors have provided the reactive power necessary for ‘voltage regula­
tion’. Storage systems deployed by transmission or distribution network operators for any other primary application can provide
reactive power to the system to augment the existing capacitors and replace capacitors planned for future installation. An energy
storage system for voltage regulation should provide reactive power on the order of 1–10 MVAR for several minutes, mainly during
daily load peaks.

6.15.2.2.4 Transmission upgrades deferral

When growing demand for electricity approaches the capacity of the transmission system, utilities add new lines and transformers.
Because load grows gradually, new facilities are designed to be larger than necessary at the time of their installation, and utilities
under-use them during their first several years of operation. To defer a new line installation or transformer purchase, a utility can
employ an energy storage system until the load demand will make better use of a new line or transformer. The power requirement
for this application would be on the order of hundreds of kilowatts to several hundred megawatts. Butler [5] states that the energy
storage system should allow for 1–3 h of storage to provide support to the constrained transmission facility.

6.15.2.2.5 Distributed generation

The growing presence of distributed sources opens a new market for storage systems, which can assist during transient conditions of
generation units such as microturbines and diesel engines, with a slower dynamic response and thus limited capability to adjust to
load changes. In this way, storage can increase the distributed generation capacity that can be embedded on a distribution network
and avoid cost-intensive reinforcements. A less-demanding application of storage technologies in distributed generation is ‘peaking
 Pumped Storage Hydropower Developments 409

generation’, which can also avoid reinforcement of distribution lines. Areas with temporary high demands, for example, at daytime,
could be equipped with storage that supply power at peak times and are recharged through off-peak hours. These applications are
often referred to as ‘distribution capacity deferral’. An energy storage system deployed to defer installation of new distribution
capacity requires power on the order of tens of kilowatts to a few megawatts, and must provide 1–3 h of storage.

6.15.2.2.6 Renewable energy applications

Electrical energy storage is very promising as a means of tackling the problems associated with the intermittency of renewable
sources such as wind and solar energy. This application will cover a wide range of power capacity and discharge duration. With the
increasing market penetration of renewable sources, these applications are more and more likely to gather momentum within future
energy systems, as conventional generation utilities’ ability to even out the intermittent renewable energy production is limited.
There is a variety of denominations in the technical literature for the use of storage in connection with renewable energy-related
applications. Butler [5] states that some authors call it ‘renewable integration’ or ‘renewable energy management’. Schoenung [3]
identifies only one utility-scale application under the term ‘renewable matching’, referring to the use of storage to match renewable
generation to any load profile, making it more reliable and predictable and hence more valuable. This does not seem to be
applicable to the storage of renewable energy at off-peak times to be delivered at peak times. Herr [4], however, broadens the scope
of renewable matching, by referring to applications making renewable electricity production more predictable throughout the day
and bringing renewables closer to demand profiles, especially providing high power outputs at peak hours. Baxter and Makansi [7]
identify four categories within renewable energy-related storage: distributed generation support, dispatchable wind, base load wind,
and off-grid applications. Storage systems with a longer discharge duration can cover longer mismatches (up to several hours). In
the longer term, a utility with a significant percentage of renewable power may require storage capacity of days to ride through
periods with windless days. In Table 1, a number of short- and long-discharge renewable matching applications are included. Both
will be referred to later as ‘renewable integration’. Indeed, a broader scope can be given to renewable integration, including
short-time applications that also contribute to tackling the problems associated with intermittent sources. The storage system
required for either application would need to provide from 10 kW to 100 MW. According to Butler [5], the storage system would
need response time on the order of fractions of a second if transient fluctuations are to be addressed. The cycling of the storage

Table 1 Applications of storage systems with different discharge times

Discharge
Application Power rating duration Storage capacity Response time System location

Fast discharge Transit and end <1 MW Seconds ∼ 2 kWh <1/4 cycle End use
use ride-through
Transmission and up to 100 s MVA Seconds 20–50 kV Ah <1/4 cycle Transmission
distribution and
stabilization distribution
Short to long Voltage regulation up to 10 MVAR Minutes 250–2 500 kV Arh <1/4 cycle Transmission
discharge Fast response 10–100 MW <30 m 5 000–500 000 kWh <3 s Generation
spinning reserve
Conventional 10–100 MW <30 m 5 000–500 000 kWh <10 min Generation
spinning reserve
Uninterruptible <2 MW ∼2h 100–4 000 kWh Seconds End use
power supply
End use and <5 MW 1–3 h 1 000–150 000 kWh Seconds End use and
transmission distribution
peak shaving
Transmission up to 100 s MW 1–3 h 1 000–500 000 kWh Seconds Transmission
upgrade deferral
Renewable <100 MW Min–1 h 10–100 000 kWh <1 cycle Generation
matching (short
discharge)
Long discharge Renewable <100 MW 1–10 h 1 000–100 000 kWh Seconds Generation
matching (long
discharge)
Load levelling 100 s MW 6–10 h 100–10 000 MWh Minutes Generation
Load following 10–100 s MW Several 10–1 000 MWh < cycle Generation and
hours distribution
Emergency back­ <1 MW 24 h 24 MWh Seconds End use
up
Renewables back­ 100 kW–1 MW Days 20–200 MWh Seconds–Minutes Generation and
up end-use
410 Design Concepts

systems coupled with wind energy will be rather unpredictable and could range from 100 to 1000 cycles per year or more. In remote
locations not connected to the grid, it may be useful to include energy storage to minimize the generation capacity. This is especially
attractive in renewable-based supplies. Renewable backup applications should be capable of substituting renewable production
when this is not available for time lengths that could go up to a week. The power rating would depend on the corresponding power
output of the renewable system.

6.15.2.2.7 End use applications

The primary end use application of energy storage is in maintaining/improving power quality. Outages and poor power quality
phenomena are important concerns for many business sectors – a survey estimated losses between $119 billion and $189 billion in
the US economy alone. Energy storage systems are being successfully deployed to provide reliable and high-quality power to
sensitive loads. ‘Transit and end use ride-through’ are applications requiring very short durations combined with very quick
response times. They cover electric transit systems with remarkable load fluctuations and customer power services like voltage
stabilization and frequency regulation to prevent the events that can affect sensitive processing equipment and can cause data and
production losses. The demand for quality power is growing within industry and is becoming a matter of concern also for electricity
suppliers, who may also install systems at the distribution level to improve the power quality. Uninterruptible Power Supply (UPS)
devices provide protection against electricity supply downtimes. Primarily they prevent production losses; however, if the serving
systems have very short response times they can also be used for power quality assignments (protection against voltage sags and
power surges, frequency regulation, etc.). UPS systems often consist of a storage device that usually acts for a short duration of time
until a generation set takes over. Although the provision of UPS is usually taken up at the user level, generation facilities can also use
storage systems to remove particularly the short-term fluctuations from their supply. The attractiveness of the investment will
depend on any penalties imposed on generating units that fail to provide a quality supply. There are other customer uses such as
‘end use peak shaving’, which can avoid additional demand charges by reducing demand peaks. ‘Emergency backup’ at customer site
requires power ratings of approximately 1 MW for durations up to 1 day. Presently, most of these applications are served by
reciprocating engines.

6.15.2.2.8 Miscellany
The provision of ancillary services by storage systems can also include black start capability, which consists of the supply of
electricity for the start-up of generators after a network failure. This is usually performed by relatively expensive diesel engines. Some
storage options also need an auxiliary electricity supply, but several can start without an electricity source. Once again, this service
can be provided as an addition to the other applications listed. A number of different lists and different descriptions of the storage
applications can also be found in the literature. Some authors include deferment of new capital equipment as a separate category.
This is in fact simply an aggregation of some of the applications already mentioned that can be performed by conventional
equipment such as peaking plants, new lines, substations, and so on. Installation of storage systems on the transmission and
distribution grid in order to expand the grid capacity, decouple generation and load, and thus reduce congestion has already been
included as a separate application (transmission upgrade deferral). The use of storage systems to improve transmission stability also
reduces or defers the need for transmission upgrades. Likewise, storage units installed for load leveling, spinning reserve, or peak
shaving delay the need for new generation capacity. Other authors also refer to the improvement of power plant efficiency as a
category, but this is rather a driving force for applications such as load management and spinning reserve. Environmental benefits
are also sometimes quoted as an application, but it is rather a consequence derived from the application of storage systems as
spinning reserve, peak shaving, and others that results in a cut in emissions that conventional technologies cause. Energy storage can
enhance the environmental performance of a network in a number of ways:

• Conventional generating units used to provide spinning reserve and other ancillary services could be replaced by energy storage.
• Generators, which operate best at constant load, can be combined to provide ramping and peaking duties.
• Grid upgrades can be avoided.
• System control issues arising from intermittent RE sources can be mitigated, thus increasing the proportion of renewable energy
generation that the system can absorb.

6.15.2.3 Storage Technologies

Storage systems generally comprise three key elements, namely, storage subsystems, power conversion systems (PCS), and balance
of plant systems (BOP), as illustrated in Figure 2. Depending on the storage system, certain elements within the scheme may be
unnecessary; for example, pumped hydro and compressed air energy storage do not need a rectifier and inverter, as pumps and
compressors operate using alternating current (AC).
There is a wide range of energy storage technologies at utility scale that are at various stages of development. Each technology has
different features that make it more or less desirable for the various applications. Table 2 provides an overview of possible selection
criteria. The relevance of the different features varies largely depending on the application to be served. Fundamental criteria for any
technology will be power capacity (including the reactive power capacity for some applications), energy capacity/discharge time,
 Pumped Storage Hydropower Developments 411

Grid/load interface

Power conversion system (PCS)

Transformer Transformer
and Control unit and
rectifier inverter

Storage subsystem
Converting devices Storage medium

Services Building

Balance of plant (BPO) Heating/ventilation/air conditioning

Figure 2 Scheme of a storage system.

Table 2 Criteria for the selection of a storage system

Design Operating Financial Others

Power rating Overall cycle efficiency Capital cost per energy stored Health and safety aspects
Storage capacity/discharge duration Lifetime/maximum Capital cost per power rating Environmental impacts
Response time number of Fixed O&M cost Synergies with other sectors
Energy density per unit area (foodprint) charge–discharge cycles Variable O&M cost
Energy density per unit volume and weight Parasitic losses Replacement cost
Maturity of technology Disposal cost
Reliability Commercial risk
Modularity
Siting reguirements
Portability
Synergies with other energy applications

and reaction time. Some applications, like grid support, require discharge to commence less than a second after starting; others, like
power sales, can be scheduled allowing for a reaction time of a few minutes.
Ideally, energy storage technologies should

• entail low capital, operating, and maintenance cost

• have a long lifetime
• be flexible in operation
• have high efficiency
• have a fast response
• be environmentally sustainable

There is a notable absence of detailed technical information on storage technologies. This is surprising given the growing need and
opportunities for storage technologies. Somewhat superficial reviews can be found in Gandy [8], Schoenung [9], Swaminathan [10],
Ter-Gazarian [11], and Herr [4]. Electricity storage systems can be technically categorized by their inherent physical principles into
mechanical, electromagnetic, and electrochemical storage devices (Figure 3).
412 Design Concepts

Mechanical Electromagnetic Electrochemical

Pumped hydro Supercapacitors Batteries

Compressed air Superconducting Flow batteries

magnets

Flywheel Hydrogen

Figure 3 Storage technology categories.

6.15.3 Pumped Storage Hydropower Plant

6.15.3.1 Characteristics
Pumped storage hydroelectricity works on a very simple principle. Two reservoirs at different altitudes are required. When the water
is released from the upper reservoir, energy is generated by the down flow, which is directed through high-pressure shafts, linked to
turbines. In turn, the turbines power the generators to produce electricity. Water is pumped back to the upper reservoir by linking a
pump shaft to the turbine shaft, using a motor to drive the pump.
This kind of plant generates energy for peak load, and at off-peak periods water is pumped back for future use. During off-peak
periods, excess power available from some other plants in the system (often a run-of-river, thermal, or tidal plant) is used for
pumping the water from the lower reservoir. A typical layout of a pumped storage plant is shown in Figure 4.
A pumped storage plant is an economical addition to a system, which increases the load factor of other systems and also provides
additional capacity to meet the peak loads. These have been used widely in Europe, and about 50 plants are operational in the
United States since the year 1990, reaching a total installed capacity of about 6700 MW.

6.15.3.2 History
Shoenung [12] acknowledges the role of pumped hydro energy storage (PHES) as the most widespread energy storage system
currently in use on power networks, operating at power ratings up to 4000 MW and capacities up to 15 GWh. PHES uses the
potential energy of water, swapped by pumps (charging mode) and turbines (discharge mode) between two reservoirs located at
different altitudes. Currently the overall efficiency is in the 70–85% range, although variable-speed machines are now being used to
improve this. Overall efficiency is limited by the efficiency of the deployed pumps and turbines (neglecting frictional losses in pipes
and water losses due to evaporation). Plants are characterized by long construction times and high capital costs. One of the major
problems related to building new plants is of an ecological/environmental nature. At least two water reservoirs are needed. Some
high-dam hydro plants have a storage capability and can be operated as a pumped hydro. A relatively new concept of pumped hydro
employs a lower reservoir buried deep in the ground. A good example of underground pumped storage is Dinorwig plant in the
United Kingdom, commissioned in 1982, which includes Europe’s largest man-made cavern under the hills of North Wales. Open
sea can also be used as the lower reservoir – a seawater pumped hydro plant was first built in Japan in 1999. Pumped hydro facilities
are available at almost any scale with discharge times ranging from several hours to a few days. PHES can be designed for fast
loading and ramping, allowing frequent and rapid (< 15 s) changes between the pumping, generating, and standby spinning modes
[13]. Dinorwig plant can go from 0 to 1890 MW (full capacity) in only 16 s. PHES is best suited to load leveling, storing energy

Upper reservoir

Water flow up
Penstock when pumping
(off peak)

Plant equipped with

Water flow down reversible pump–turbine
when generating and motor–generator
(on peak)
Lower reservoir

Power house
Figure 4 Typical layout of a storage hydroelectricity power plant.
 Pumped Storage Hydropower Developments 413

during off-peak hours for use during peak hours, and spinning reserve. They can provide energy to meet peak demands, and, in the
pumping mode, they serve as the source of load for base load during off-peak periods, helping to avoid cycling of the regular
generating units, thus improving their operating efficiency. PHES systems can provide other benefits, including black start capability
(they can begin generation without an external power source) and frequency regulation. There is over 90 GW of pumped storage in
operation worldwide in nearly 300 plants, which is about 3% of global generation capacity. In 1998, 10% of Japan’s total
instantaneous energy requirement was met by pumped hydropower generation [14]. Table 3 contains some of the most repre­
sentative pumped hydro plants in the world larger than 1000 MW in current net capacity, which are currently operational or under
construction. The 292 MW Turlough Hill Pumped Storage station (representing approx. 5% of installed capacity) is the only bulk
electrical energy storage facility in Ireland. Its construction was completed in 1974 and involved the building of a huge cavern in the
heart of the mountain, in which the generation plant and controls are housed. A pumped storage system allows for the use of excess
electricity capacity during non-peak hours to pump water from the lower to the upper lake at Turlough Hill, and then the release of
water in the reverse direction to create electricity in times of maximum demand. Unit prices for pump/turbines have leveled out as
the technology has matured. Thus, costs are typically around $600 kW−1. Reservoir costs can vary from almost nothing to more than
$20 kWh−1, according to Gordon [13]. Schoenung places that cost at $12 kWh−1.

6.15.3.3 Characteristics of Pump–Turbines

Reversible pump–turbines or separate generating and pumping equipment may be installed in pumped storage plants. Some are
equipped with a fixed or adjustable distributor. In the case of separate machines, a clutch operable at standstill, a starting turbine, or
a synchronizing torque converter facilitating extremely short changeover times can be provided. Whether a reversible pump–turbine
or a turbine and pump combination, these machines are extremely durable. Figure 5 gives the application range of pump–turbines,
and Figure 6 shows for selected pump–turbines the operation range in the pump mode [15]

6.15.3.3.1 Elements of pump–turbine hydraulic design

Designing a reversible pump–turbine is still a complex and challenging task for the reasons and the characteristics previously stated.
This is why the designing turns out to be an iterative process where all steps are interdependent [16]. For all components (spiral case,
stay vanes, guide vanes, runner, and draft tube), geometry definition is the first step of the hydraulic design process.
Capabilities of the latest computer systems and computational fluid dynamic (CFD) codes allow the intricate hydraulic
computations of all the parts of the pump–turbine with increasing accuracy. The objective is to check the behavior of flows in
the entire operating range, in both pump and turbine modes. It also allows verification of whether the hydraulic criteria are satisfied
or not. For each component of the machine, even if one of the hydraulic criteria is not satisfied, the geometry of the failing
component is modified to attain the required criterion. When the design of the runner and other components is hydraulically
validated by CFD, a mechanical check is performed, based on a structural analysis by means of a finite element model (FEM). The
objective is to check the mechanical behavior of the components. It has been pointed out that the runner blade thickness, and the
stay vanes and the guide vanes thickness have a strong influence on the static stress level.
These points are considered during the design development [17] through hydraulic and mechanical optimization.
Net heads of up to 817 m in the turbine mode and 832 m in the pump mode are considered as a limit for conventional
single-stage pump–turbines.
Even though all hydraulic components are important and can affect the pump–turbine performance as a whole, the runner
remains, of course, the key element to match the performances. This is the reason why special attention is paid to runner design.

6.15.3.3.1(i) Runner hydraulic design

Runner design is optimized using Navier–Stokes computations to deliver the best characteristics in terms of efficiency level and
cavitation behavior in both pump and turbine modes. The hydraulic criteria to be followed are mainly based on the energy results
(head vs. discharge and head losses) as well as the flow field at the runner outlet and cavitation behavior with specific reference to its
incidence at the blades’ leading edge. For cavitation behavior, specifically in the pump mode, the pressure level at the blades’ leading
edge is calculated in order to guarantee a free cavitation pitting area over the operating range. A large net head variation can be added
to a large frequency variation, thus increasing the operating range.
For various discharge values covering the whole operating range, calculation results can be compared with experimental
observations on the model for which the critical cavitation limit is determined by decreasing the net positive suction head
(NPSH) until efficiency drops (Figure 7).

6.15.3.3.1(ii) Pressure fluctuation level

Among all the parameters to be mastered in the pump–turbine design, vibratory behavior has a particular importance. Such
vibrations result from the mechanical response of the structure to hydraulic solicitations due, both to the operating conditions and
to hydraulic interaction between the rotor and the stator.
Since 2000, and especially for the latest PSP designs, a direct application of this R&D program has been performed by decreasing
the pressure fluctuation level especially between the runner and the guide vanes as well as in the casing. A comparison is made
between the pressure fluctuation levels measured on the model (Figure 8) for two hydraulic designs (Nq 32). The first one was
developed 6 years ago, whereas the second one has been developed recently.
414 Design Concepts

Table 3 Some of the largest pumped hydro facilities in the world

Capacity
Station Country (MW)

Bad Creek Hydroelectric Station United States 1065

Bailianhe Hydroelectric Station China 1224
Baoquan Pumped Hydroelectric Station China 1200
Bath County Pumped Storage Station United States 2772

Blenheim-Gilboa Hydroelectric Power Station United States 1057

Castaic Dam United States 1566
Chiotas Dam Italy 1184
Coo Hydroelectric Power Station Belgium 1164
Dinorwig Power Station United Kingdom 1728

Dniester Hydroelectric Power Station Ukraine 2268

Drakensberg Pumped Storage Scheme South Africa 1000
Goldisthal Hydroelectric Power Station Germany 1060
Grand Maison Dam France 1070
Grande Dixence Dam Switzerland 2069
Guangzhou Pumped Storage Power Station China 2400

Heimifeng Pumped Storage Power Station China 1200

Helms Dam United States 1200
Huhhot Dam China 1200
Huizhou Hydroelectric Power Station China 2400
Ingula Pumped Storage Scheme South Africa 1332
Imaichi Dam Japan 1050

Kannagawa Pumped Storage Power Station Japan 2700

Kazunogawa Dam Japan 1600
Kruonis Pumped Storage Plant Lithuania 1600
Lago Delio Hydroelectric Station Italy 1040
Liyang Hydroelectric Power Station China 1000
Ludington Pumped Storage Power Plant United States 1872

Markersbach Dam Germany 1050

Matanoagawa Pumped Storage Station Japan 1200
Minghu Dam Taiwan 1000
Mingtan Dam Taiwan 1602
Mount Elbert United States 1412
Mt. Hope Dam United States 2000

Muddy Run Pumped Storage Facility United States 1071

Northfield Mountain United States 1080
Okutataragi Pumped Storage Power Station Japan 1932
Okuyoshino Pumped Storage Power Station Japan 1206

(Continued)
 Pumped Storage Hydropower Developments 415

Table 3 (Continued)

Capacity
Station Country (MW)

Piastra Edolo Pumped Storage Station Italy 1020

Presenzano Pumped Storage Power Station Italy 1000

Pushihe Pumped Storage Power Station China 1200

Pyramid Lake United States 1495

Raccoon Mountain Pumped-Storage Plant United States 1530

Robert Moses Niagara Power Plant United States 2515

Shin Takasegawa Pumped Storage Station Japan 1101

Shintoyone Dam Japan 1125

Siah Bisheh Dam Iran 1140

Sir Adam Beck Hydroelectric Power Stations Canada 1600

Taian Pumped Storage Power Station China 1000

Tamahara Pumped Storage Power Station Japan 1200

Tashlyk Hydro-Accumulating Power Station Ukraine 1494

Tehri Pumped Storage Power Station India 1000

Tianhuangping Pumped Storage Power Station China 1800

Tongbai Pumped Storage Station China 1200

Tumut-3 Australia 1500

Vianden Pumped Storage Plant Luxembourg 1100

Xiangshuijian Pumped Storage Station China 1000

Xianyou Pumped-storage Power Station China 1200

Xilongchi Pumped Storage Power Station China 1200

Yangyang Pumped Storage Power Station South Korea 1000

Yixing Pumped Storage Power Station China 1000

Zagorsk Pumped Storage Station Russia 1200/1320

Zhanghewan Pumped Storage Station China 1000

Zhuhai Pumped Storage Station China 1800

1000
Setting getting deeper
Head (m)

Pump−turbine
100

10
1 10 100 1000
Output (MW)
Diameter increasing
Speed decreasing

Figure 5 Application range of pump–turbines.

416 Design Concepts

1000

29
4943
36 52
31
47 35 42 53
24 14 39
21 27
50 51
44 7 34 45 12
38 22 48

Pump head H (m)

37

100

25

K=
10
00
3

K=
80
0
K=
K = nq × √H

K=
K=

60
40
Q0.5

20

0
nq = n ×

0
0
10 H0.75
10 100 200
Specific speed (nq)

3 Lewiston 7 Coo I I12 Raccoon

14 Rodund II 21 Bath County 22 Coo II
24 Estangento 25 Gabriely Galan 27 Kühtai
29 Obrovac 31 Presenzano 34 Palmiet
35 Bad Creek 36 La Muela 37 Herdecke
38 Chiotas (4stage) 39 Mingtan 42 Shisanling
43 Guangzhou II 44 Edolo (5 stage) 45 Goldisthal (two variable speed)
47 Venda Nova 48 Tai An 49 Siah Bishe
50 Waldeck I 51 Limberg II 52 La Muela II
53 Ingula

Figure 6 Selected pump–turbines: Operation range in the pump mode.

Area without cavitation

Operating range
Experimental
NPSH

result
Calculation
result

Cavitation inception limit

Discharge
Figure 7 Experimental and calculation results comparison for cavitation behavior in the pump mode.

6.15.3.3.2 Elements of pump–turbine mechanical design

Mechanical design of these reversible pump–turbines is based on both experience and customer’s specifications.
The prominent characteristics of main projects are as follows (Figure 9):

• Shaft line is resting on three bearings: two for the generator–motor and one for the pump–turbine.
• Thrust bearing is supported by the lower bracket. The inlet valve is of the spherical type.
 Pumped Storage Hydropower Developments 417

Pressure fluctuation level: PUMP mode

More then 2.5% decrease

Peak-to-peak / u22/2g (%) between the runner and the
1%
guide vane
6 year ago

Today

6 years ago
0.5% decrease
in the stay vane Today
channel

0.85 0.95 Q/Qo pt 1.05 1.15

Gap runner/guide vane Stay vane channel

Figure 8 Pump model test result: Decreasing of the pressure fluctuation levels for 6 years.

Figure 9 Single-stage pump–turbine.

418 Design Concepts

Figure 10 Double-stage pump–turbine.

• The pump–turbine is to be dismantled through the generator–motor stator or at the pump–turbine pit level.
• The spiral case is pressure tested before concreting and is concreted with a partial water pressure inside.
• The bottom ring is embedded into the concrete.
• Guide vanes and distributor mechanism are guided by self-lubricated bushes.
• Guide vanes are operated either by two servomotors connected to one operating ring (mechanical synchronization) or by
individual servomotors (electronic synchronization).
• Runner is fabricated and the blades are machined by a five-axis numerically controlled milling machine.
• Pump–turbine guide bearing is pad or shell type.
• Shaft seal is of the hydrostatic axial type.
• Draft tube cone is embedded.
• Starting in pumping mode with dewatered runner can be performed either by a static frequency converter or back to back.

For a double-stage pump–turbine, some characteristics are specific (Figure 10).

• Thrust bearing can be located below the pump–turbine

• Draft tube cone is not embedded
• Dismantling of the lower runner by bottom

6.15.4 Examples of Remarkable Pumped Storage Power Plants

6.15.4.1 Okinawa Seawater Pumped Storage Power Plant
This project is located in a unique area in Japan where many original species of flora and fauna indigenous to the Okinawa Islands
have been preserved until then. Right from the start of construction activity, several measures were undertaken to mitigate its
environmental impacts. These included protection of habitat area, prevention of muddy water outflow, reduction of noise and
vibration of heavy construction equipment, installation of slope-type side ditch, restoration and revegetation of the disposal area,
and creation of a biotope. These mitigation measures were decided upon based on a study conducted by a special committee formed
by well-known local environmental specialists and the local community (Figure 11).
 Pumped Storage Hydropower Developments 419

Figure 11 Okinawa seawater pumped storage power plant.

6.15.4.1.1 Outline of the project

The project is a demonstration plant for seawater pumped storage power generation located at the northern part of Okinawa Island.
In making the concept of seawater pumped storage power generation practical, there was the necessity to find concrete solutions to
the technical problems arising from the use of seawater and to the problems of environmental impacts. There was no case of
seawater pumped storage power generation actually implemented anywhere in the world till that time, and this pilot plant
constituted a first example.
The Electric Power Development Co. Ltd. (J-POWER) [18] undertook implementation of the project on consignment from the
Ministry of International Trade and Industry (MITI) [19]. Surveys and research regarding the technical and environmental aspects of
using seawater were carried out for 6 years from 1981.
The feasibility of a seawater pumped storage power plant was studied, and, having gained a favorable outlook, construction was
performed from 1991 until 1999. Since 1999, a 5-year program of tests has been going on, its aim being to demonstrate the
practicality of seawater pumped storage power generation technology.
In Okinawa, where the project site is located, daily load fluctuations are being met with regulated operation of thermal and gas
turbine, and the need for pumped storage power generation is well recognized. Water resources are precious on Okinawa Island and
construction of a conventional pumped storage power plant using river water would not be appropriate. As for the northern part of
the island, it is mostly mountainous, and there are many locations that would be suitable for setting up a seawater pumped storage
power plant.
With these in mind, sites meeting the requirements were picked out from all over Okinawa Island, and comparison studies of the
impacts on the natural and social environments were made, the result being the selection of the present site.
Table 4 gives the specifications, and Figures 12 and 13 the general plan and profile, respectively, of the plant.

Table 4 Okinawa Seawater pumped storage power plant specifications

Item Specification

River system
Catchment area
Power plant Name Okinawa Yanbaru Power Plant
Max. output 30 MW
Max. discharge 26 m3 s− 1
Effective head 136 m
Upper regulating pond Type Excavated type, Rubber sheet-lined
Max. embankment height 25 m
Crest circumference 848 m
Max. width 251.5 m
Total storage capacity 0.59 � 106 m3
Max. depth 22.8 m
Waterway Penstock Inside dia. 24 m Length 314 m
Tailrace Inside dia. 27 m Length 205 m
420 Design Concepts

Upper pond N
Access road

Sheet lining
Intake

Penstock
Underground powerhouse Switchyard

Access shaft Tailrace tunnel

Pacific Ocean
Outlet
0 500 m

Figure 12 Okinawa project: General plan.

EL.
200 m Sheet lining

H.W.L152.00 154.00

150 H.W.L132.00

Intake
100
Upper pond Access tunnel to outlet
Penstock
50 Access shaft
Powerhouse Outlet
0
25.00
Tailrace tunnel

–50

Figure 13 Okinawa project: Profile of waterways.

6.15.4.1.2 Features of the project area

It is said that the Ryukyu Island chain of which Okinawa is a part was connected by land to China on the Asian continent
approximately 1.5 million years ago, but due to land upheavals and rising of sea level, a considerable area became submerged, the
Ryukyu Islands remaining as a result.
Consequently, fauna and flora that had existed in the area occupied by the islands have lived in an isolated state since then so
that original species have been preserved, and there are many species and subspecies indigenous to the Ryukyus and not found
anywhere else in the world.
The climate is the closest to tropical in Japan, and it is warm throughout the year with annual mean air temperature being 23 C.
Precipitation is approximately 2400 mm, 40% rainier than the national average, short bursts of heavy squalls occurring frequently.
Many typhoons strike the islands from July to September.
The regulating reservoir is located at the Pacific Ocean side of Kunigami Village, approximately 600 m from the shore and on a
tableland roughly 150 m asl. This area comprises a gently sloped plateau of elevation 150–170 m, gradually declining toward the sea
(east), and on nearing the sea, drops into it by a cliff of specific height from 130 to 140 m. Gullies 20–30 m deep and with gentle
gradients have developed in dendritic form on this plateau. These gullies gradually join together as the sea is approached, and on
reaching the cliff, feed into the sea as a continuous chain of waterfalls. The seashore immediately below the cliff consists of both
large boulders and sandy beaches (Figure 14).
A red soil classified as Kunigami maaji, a special type of soil in geological terms, covers the greater part of Okinawa Island.
Kunigami maaji is a soil from weathered sedimentary rocks, the parent rocks being of various kinds such as phyllite, andesite, and
sandstone. Molecular binding is weak and coefficient of permeability is low, from 10−5 to 10−6 cm s−1, so that surface flows readily
occur when it rains, and particles of this soil are washed out. Further features of Kunigami maaji include the following:

• It is highly weathered.
• It contains Kaolinite, a clay mineral.
 Pumped Storage Hydropower Developments 421

Okinawa
seawater
PSPP

Figure 14 Okinawa project: Site location.

• Its specific gravity is high (2.6–2.8).

• When it is washed out to the sea, it tends to not go out to the open sea, and causes serious damage to the coral reefs along the
coast, constituting a social problem for the local community.

The vicinity of the construction site is surrounded by state-owned forestland, and the area had been designated a US Marine Corps
training ground under the Japan-US Status-of-Forces Agreement, but a portion was released as demonstration tests were to be
conducted, while a separate portion is now leased and is being jointly used. The project site is not in an area designated as a natural
park, and there are no cultural assets or recreational facilities.

6.15.4.1.3 Major impacts

J-POWER under consignment from MITI to carry out environmental impact surveys and studies of seawater pumped storage power
generation, began field surveys in 1982, and compiled the environmental impact assessment (EIA) report in 1989. In Table 5 the
EIA items are shown.
The environmental problem of greatest concern was the potential effects of salt spray on the surrounding environment due to
pump-up of seawater. Wind tunnel water tank tests and simulations using numerical models were carried out in this regard, and it
was ascertained that there would be little difference from salt spray flying directly from the sea.
With regard to animal life, it was confirmed that there were 16 species of rare animals inhabiting the area of which five were
listed as endangered and seven as threatened in the Red Data Book. As for plant life, there were many indigenous species (Okinawa
Island-indigenous: three species, Okinawa Island northern limits-indigenous: five species, Ryukyu Islands-indigenous: nine spe­
cies), while in the sea area reef-building coral was found.
Under these circumstances, a study committee for the protection of invaluable assets was organized in 1989, comprising mainly
of local environmental specialists. Using the findings of the study committee as reference, fundamental principles of environmental
conservation were set up as follows:

1. Organisms inhabiting the project area are native to the area and, as such, are to be given consideration with a modest attitude.
2. The development area is to be held to a minimum in order that the ecosystem will be disrupted as little as possible.

Table 5 Okinawa project: Environmental impact assessment items

Environmental impact assessment items

Meteorology, weather, air quality, water quality, noise, vibration, offensive odor, soil contamination
Ground settlement, topography, geology, sea current, marine phenomenon
Salt spray, seawater seepage
Plant Vegetation, rare plant, soil profile, etc.
Animal Terrestrial animal Mammals, birds, reptiles, amphibians, insects, soil fauna
Aquatic organism Gully animals
Marine organism Coral, fishes, benthic organisms, plankton, eggs and fry, tideland organisms, seaweed, grasses
422 Design Concepts

3. The scope of implementation of the measures for natural environment protection is not to be limited to the construction area,
and is to be applied to the surrounding area as well.
4. Any damage caused to the environment is to be rectified without delay taking advantage of natural self-healing powers.

Based on these fundamental principles, the environmental impact factors listed below were extracted:

1. Outflow of muddy water (red water) produced from the construction area into the gullies and sea area near the river mouth;
2. Reduction of habitat area due to changes made in land;
3. Noise and vibration from heavy equipment; and
4. Damage to small animals from construction vehicles and accidents due to falling down into roadside gutters.

6.15.4.1.4 Mitigation measures

The environmental conservation measures in Table 6 were considered and implemented for each environmental impact factor.
Meetings were held with the local community to explain these measures and consent was obtained.
Items thought especially to be of unique nature (underlined portions in Table 6) are discussed in detail below.

6.15.4.1.5 Measures during construction

6.15.4.1.5(i) Capture and transfer of fauna and flora in the construction area
Since the construction area during the construction period would become an uninhabitable environment for rare animals due to
changes made in land and traffic of large heavy equipment, small animals of limited mobility such as frogs, turtles, and newts were
captured and moved to favorable environments outside the construction area. Prior to carrying out capture and removal, tests to see
whether the animals would survive resettling were conducted and it was confirmed that resettling would be possible. (Resettling
survival tests consisted of capturing several individuals of a rare animal species and marking them before release into an optimum
environment outside the construction area and confirming several weeks later that they were active and that no dead bodies were to
be found, by which it was concluded that resettling could be carried out.)

6.15.4.1.5(ii) Installation of intrusion prevention nets

In order to prevent small animals such as turtles from entering the construction area from outside and being harmed by construction
vehicles, polyethylene nets approximately 30 cm high were installed along 8 km of the outer perimeter of the construction area
(Figure 15).

Table 6 Okinawa project: Countermeasures for environmental impact factors

Environmental impact factor Countermeasure

Outflow of muddy water produced from the Construction Chemical treatment by turbid water plant
construction area into the gullies and sea area near water
the river mouth Turbid water Chemical treatment by turbid water plant
from red soil Reduction of turbid water by separation into red water and clear water
Reduction of red water by spraying asphalt emulsion or seeds on bare
ground
Install a gabion weir downstream of the gully
Reduction in habitat area due to changes in land Reduction of area Layout of powerhouse and waterways underground
changed Omit access road and work adit to outlet and powerhouse
Reduce construction area by balancing cuts and embankments as
much as possible
Protection, Landscape and green the construction site without delay
restoration of Protect the existing forest by planting low-height trees
Noise, vibration from heavy equipment vegetation Prohibit night time work in surface construction
Use low-noise machinery
Drive at low speed inside the construction area
Harm to small animals from construction vehicles and accidents due to falling Capture and remove animals and plants in the construction area
down into the roadside ditches Install facilities (intrusion prevention nets) to prevent the entry of rare
animals
PR activities using posters, lecture meetings, pamphlets, etc.
Prevent accidents of small animals falling down into roadside ditches
and getting killed, through the construction of sloping side wall
gutters
 Pumped Storage Hydropower Developments 423

Figure 15 Okinawa project: Intrusion prevention net.

6.15.4.1.5(iii) PR activities aimed at construction personnel concerned

In order that as many construction personnel concerned as possible may be interested in rare animals, pocket booklets and
pamphlets containing photographs of rare animals, precautions for environmental conservation to be taken while executing
work, and what to do when a rare animal is encountered were distributed; posters calling attention to protection of rare animals
were put up; and lecture meetings concerning rare animals were held. The photographs of the rare animal species – birds,
amphibians, and reptiles – contained in the pocket booklets were 16 in number and were in color. A protected-animal monitor
was appointed, who, carrying out patrols, checked whether the rare animal protection measures were functioning effectively and
whether handling of rare animals was appropriate. At the end of the pamphlet was included a rare animal sighting report form for
mandatory reporting in case of encountering a rare animal.

6.15.4.1.6 Permanent measures

6.15.4.1.6(i) Construction of sloped-wall roadside ditches
Ordinarily, roadside ditches are U-shaped and if a small animal were to fall in it, it would have trouble climbing out, and in the hot
sun of Okinawa, the animal would be fried to death by the heat absorbed by concrete. There had been a number of cases in the
prefecture of chicks falling into U-shaped ditches and being killed by the heat.
Therefore, ditches with sloping sidewalls enabling animals to climb out by themselves were constructed. The ditches have
sloping walls on their mountain side and vertical walls on their road side so that animals will be guided toward natural ground, the
ditches thus playing a role in promoting traffic safety (Figure 16).

6.15.4.1.6(ii) Restoration and revegetation of the disposal yard, and biotope creation
Regarding the construction site, it was considered necessary to regain as quickly as possible a natural environment in which rare
animals could live, through the restoration of the site to its original state. Particularly, regarding the disposal yard provided adjacent
to the regulating pond where 210 000 m3 of water had been accommodated in the beginning, it was thought that the conventional
concept of a disposal yard could be abandoned, positioning it as a place (biotope) for creating an environment in which different
species of wildlife (birds, insects, plants, etc.) can coexist (Figure 17).

Figure 16 Okinawa project: Slope-type side ditch.

424 Design Concepts

Figure 17 Okinawa project: Central pond in disposal yard.

In order to recreate vegetation similar to the Itajii (the village tree of Kunigami Village, Lithocarpus edulis) forest in the
surroundings, about 20 varieties of trees, beginning with priority-status trees such as Itajii and Adeku (Syzygium buxifolium),
fruit-bearing trees such as Shimaguwa (Morus australis) and Sharinbai (Umbellata), and pioneer trees such as Akamegashiwa
(Mollotus japonicus) and sendan, which grow rapidly to form windbreaks for Itajii and others and aid in their growth, were selected,
and around 30 000 trees were planted intermixed in the disposal yard of approximately 45 000 m3. Saplings of the trees planted are
almost nonexistent commercially, and seeds collected from standing trees were sown in pots and grown for 2–3 years and then
transplanted. If left as bare ground after planting, there would have been outflow of red soil whenever it rained; so the surface of red
soil was covered with bark and wood chips at flat areas and coconut fiber mats at slopes, which, at the same time, held down
evaporation and improved the environment for the growth of trees.
Before development, there were gullies at this construction site and a riverine environment existed. In order to restore this
environment, ponds and waterways of various sizes were arranged in a well-balanced manner to create habitats for small animals
(aquatic organisms, insects, etc.) thereby aiming for biodiversity. The ponds and waterways were lined, so that small animals could
inhabit gaps between stones. The stones used for lining were of different sizes, large and small, for variation. Coconut fiber mats
were used to cover the surfaces of small ponds and waterways to prevent outflow of red soil (Figure 18).

6.15.4.1.7 Results of the mitigation measures

Environmental impact surveys have been conducted regularly since 1990 to assess the impacts of construction work on the
surrounding environment. The outline of the environmental monitoring in 1999 is given in Table 7.
The condition of growth of the principal varieties of trees at the disposal yard, a part of the results, is shown in Figure 19. The
numerical values used were obtained from trees growing in areas of 10 � 10 m at ten different locations around the regulating pond.
(The data for 1997 were to a great extent affected by a slowly moving typhoon resulting in very long hours of strong wind and rain,
and, therefore, were considered unsuitable for comparisons and omitted.)
At the disposal yard biotope, saplings have grown satisfactorily, as can be seen in Figure 20. Tadpoles and newts can be seen
swimming, and dragonflies flirting about in and around water. Because the yield of fruits from trees and the population of insects,
on which animals can feed, have increased, rare birds and boars began to appear. As an unexpected secondary effect, aquatic animals
were found to be laying eggs on the coconut fiber mats, which had been spread out at the bottom of water bodies; thus the mats
were fulfilling the role of aquatic plants.

Figure 18 Okinawa project: Cover by coconut fiber mats.

 Pumped Storage Hydropower Developments 425

Table 7 Okinawa project: Outline of environmental monitoring

Monitoring
Environmental monitoring item frequency

Terrestrial monitoring
Vegetation (before, after typhoon) 10 points around the plant Twice a year
Animals Mammals, birds, reptiles, amphibians, Insects, Line census, fixed point survey
soil fauna, aquatic organisms
Noise, vibration 4 points around plant
Regulating pond water Water quality, electrical conductivity 1 point in regulating pond Continuous
quality Transparency, pH, DO, COD, SS, n-hexane extract Once a year
Marine monitoring
Water quality Water temperature, salt content, transparency, pH, DO, 4 points in frontal sea area Twice a year
COD, SS,
n-hexane extract, coliform group number, T-n, T-P
Bottom sediment COD, total sulfides, particle-size distribution, loss on 5 points in frontal sea area
ignition
Organisms Tideland organisms, Eggs–fry, Zoo/Phytoplankton 2–3 points in frontal sea area
Marine algae and grasses 3 traverse lines in frontal sea area
Coral 3 traverse lines, 5 points in frontal
sea area
Fishes and other nekton 1 traverse line in frontal sea area

3.5
Height (m) . Branch spread (m)

3.0 Av.Trunk Diameter

Av.Tree Height
. Trunk diameter (cm)

2.5
Av.Branch Spread
2.0

1.5

1.0

0.5

0.0
Summer 95' Summer 96' 97' Before, after Before, after
typoon 98' typoon 99'
Figure 19 Okinawa project: Environmental impact assessment results.

Figure 20 Okinawa project: Central part of the disposal yard.

Regarding the measures for preventing red soil from running out, there were no outflows even when typhoons had occurred, and
it was confirmed that muddy water had not been produced.
Because rare species of flora and fauna were discovered at the site and this was shown on television, many people from the
general public started arriving to visit the site. The fact that rare animals and plants can be seen within a short period of time, without
having the need to go deep into the surrounding mountains and forests, has won attention. Studies are being conducted at the
426 Design Concepts

Kunigami village office to see whether this feature can be taken advantage of and the site can be used as a place for education of
children and for general environmental education.
Figure 20 is a photograph showing the condition at the central part of the disposal yard.

6.15.4.1.8 Reasons for success

The Study Committee for the Protection of Invaluable Assets was formed by local environmental specialists before the start of
construction; studies were conducted on environmental conservation measures and their effects; and construction was performed
taking these into consideration. These are thought to be the primary reasons for the success of the environmental protection
measures.
In particular, the creation of a biotope at the disposal yard with a view of recovering natural habitats can be given high marks.
The soil at the disposal yard was inorganic, and there prevailed conditions of severe wind and salt damage; so studies regarding tree
planting had been taken up from an early stage of planning. Based on such studies, surveys of salt damage and a survey of the types
of trees favorable to nesting of birds were carried out. These were useful in deciding the types of trees to be planted and the method
of planting.

6.15.4.2 Goldisthal Pumped Storage Power Plant

6.15.4.2.1 Introduction
The 1060 MW Goldisthal pumped storage plant on the Schwarza River is the biggest hydroelectric project in Germany and the most
modern one in Europe. Construction of the project began in September 1997, and the plant started commercial operation in
October 2004 [20].
The Goldisthal project is unique in that two of the four vertical Francis pump–turbine units feature variable-speed (asynchro­
nous) motor–generators. This arrangement provides several benefits including power regulation during pumping operation,
improved efficiency at partial-load conditions, and enhanced dynamic control of the power delivered, for stabilization of the grid.

6.15.4.2.2 Developing the Goldisthal project

In 1965, scientists within the German Democratic Republic performed a ranking study, which identified the current site on the
Schwarza River in Thuringia State as the best location for a large pumped storage hydroelectric project. Beginning in 1972 geologists
with Baugrund Dresden of Germany performed intensive geological investigations and began work on infrastructure such as the
transformer substation and bypass road. In 1975, Schachtbau Nordhausen GmbH of Germany dug several 4 km-long investigation
tunnels to explore the site geology (Figure 21).
Work on the Goldisthal project was suspended in 1980, primarily for economic reasons. The increase in energy demand in the
country was not as high as expected, and financial problems delayed the construction schedule for the plant.
Nine years later, as a result of the political unification of Germany and restructuring of the East German power supply,
investigations to develop Goldisthal resumed. In 1990, Vereinigte Energiewerke AG (VEAG) began the planning and permission
procedures to develop Goldisthal. The utility proposed to install four turbine–generator units with a total capacity of about

Figure 21 Goldisthal project.

 Pumped Storage Hydropower Developments 427

1000 MW. VEAG determined that the most advantageous arrangement for this plant involved combining conventional pumped
storage (synchronous) technology (represented by two single-speed motor–generators) with variable-speed (asynchronous) tech­
nology (represented by two variable-speed motor–generators).
Variable-speed motor–generators allow operation of the pump–turbine unit over a wider range of head and flow, making them
economically advantageous for a pumped storage facility. VEAG conducted considerable investigations to determine the number of
variable-speed units to be installed. The company eventually decided to install two variable-speed units and two conventional units.
The decision was based on a variety of reasons. First, calculations indicated that the company will need about 200 MW of
controlled power for the pumping operation, which was within the control range of two variable-speed units. Second, asynchronous
machines are not capable of restarting the system during a power outage – they need support from the grid to begin operation.
Consequently, conventional units would be required in the event of a power outage. Third, because no company in Europe had any
experience in operating large variable-speed units, VEAG considered installing variable-speed units alone as too risky.
VEAG received the building permit for Goldisthal in 1996. Construction began in September 1997 with the work on the access
tunnel to the underground powerhouse and transformer caverns, as well as such surface infrastructure work as providing site access.

6.15.4.2.3 Main features of the project

• Upper reservoir with a usable capacity of 12 million m3. This is the largest artificial reservoir in Germany, covering 55 ha under
full storage conditions. The rockfill dam impounding this reservoir is 3370 m long.
• Two 6.2 m-diameter headrace tunnels with steel-armored lining, with a total length of about 870 m.
• Underground powerhouse cavern containing four 265 MW vertical Francis pump–turbines, four motor–generators, and auxiliary
systems. This cavern is accessed via a 1 km-long tunnel from the operation building complex.
• Underground transformer cavern containing four unit transformers, the 10 kV transformers of the internal electrical supply,
switchgear, and starting frequency converters.
• Two 8.2 m-diameter tailrace tunnels with concrete-armored lining, each 275 m long.
• Lower reservoir with a capacity of 18.9 million m3, impounded by a rockfill dam (Figure 22).

6.15.4.2.4 Choosing variable-speed machines

The most important innovation at the Goldisthal project is the first ever application of variable-speed motor–generators of this size
in a hydro plant in Europe. Essentially, turbines have one optimum operating point in terms of head, flow, unit size, and speed. But

Upper reservoir Static heads Lower reservoir

Active storage volume 12 × 106 m2 Extreme maximum 334.00 m Gross storage volume 18.9 × 106 m2
Max. storage height 29.5 m2 Maximum 325.00 m Dead storage volume 2.9 × 106 m2
Max. reservoir surface area 55 ha Normal 301.65 m Active storage volume 12 × 106 m2
Dam volume 5.4 × 106 m2 Minimum 280.70 m Max. reservoir surface area 78 ha
Crest length 3.370 m Extreme minimum 279.20 m Dam height 67 m
Dam volume 0.7 × 106 m2
Electromechanical equipment
Crest length 220 m
4 units
Max. W.L. 874.00 (2 with adjustable speed) 333 (+4/−10%) Secondary dam
Nom. output per unit 266 MW Total storage volume 0.7 × 106 m2
Nominal speed 333 rpm Max. reservoir suface area 8 ha
Min. W.L. 849.30 Nom. turbine discharge 103 m2 s−1 Dam height 26 m
2 −1 6 2
Nom. pumping discharge 80 m s Dam volume 0.2 × 10 m
Centre line unit 493.50 m asl Crest length 120 m
Transformer cavern
Length 122 m
Width 15 m
Height 17 m
2
Excavation volume ca. 32.000 m
Max. W.L. 568.50
Headrace tunnels
2 pressure tunnels (steel-lined) Ø 6.20 m Min. W.L. 549.00 (540.00)
Excavation volume ca. 65 000 m2
Length:
5.8°

North Tunnel 817 m

5.4°

60 m
N=2

South Tunnel 828 m

S=2

Manifold tunnels
2 branches (M1/M3) 47 m each 8.4°
2 branches (M2/M4) 40 m each
ca. 861 m ca. 418 m

0 50 100 150 200 m

Machine cavern Tailrace tunnels
Length 137 m 2 pressure tunnels Ø 8.20 m
Scale
Width 25 m Excavation volume ca. 57.000 m2
Height 49 m Length:
Excavation volume ca. 152 000 m2 North Tunnel 279 m
Elev. access tunnel 508.40 m asl South Tunnel 276 m
Elev. cavern foundation 482.50 m asl Manifold tunnels
4 branches 67 m each

Figure 22 Schemes of Goldisthal project.

428 Design Concepts

Figure 23 Goldisthal project: Variable-speed unit (at left) and one synchronous unit.

when these units are coupled with a variable-speed motor–generator, operating speed can be varied over a certain range of the
nominal synchronous speed of the turbine–generator unit. As head and flow vary, the unit is able to increase or decrease its speed to
operate closer to its peak efficiency for that unique set of conditions.
Each half of the 1060 MW Goldisthal pumped storage project features one variable-speed unit (Figure 23) and one synchronous
unit. The variable-speed units allow for efficient operation at a wider head range, during both turbine and pumping operations.
The difference between synchronous and asynchronous machines lies in the construction of the rotor. While classical synchro­
nous generators have salient poles, variable-speed generators have a three-phase winding on the rotor. And whereas the
synchronous rotor is energized by a direct current (DC) to create a rotating magnetic field, the asynchronous rotor is energized
by a low-frequency AC. A direct frequency converter in the rotor circuit is used to control the frequency. If the frequency is changed,
so too is the speed of the unit.
The rotor can be retarded or accelerated, opposite the stator field, from 90% to 104% of the synchronous speed. The variable
frequency of the asynchronous generators at Goldisthal ranges from 5 Hz opposite to the stator field of 333 rpm (which provides
300 rpm) to 0.01 Hz (which is nearly the rated speed of the unit) to 2 Hz in addition to the stator field (which provides 340 rpm).
Asynchronous motor–generators provide several advantages, including

• More flexibility in their operation;

• Higher efficiency over a wide range of operation at partial-load conditions;
• A wide range of controllable and optimized power consumption in pump operation;
• Additional and faster features for grid control, such as fast power outlet regulation;
• Better use of the reservoir because higher water level variations can be allowed; and
• Better contribution to grid stability because of the high moment of inertia of the rotating masses.

Asynchronous machines make it possible to regulate power not only in the turbine mode but also in the pumping mode. The range
of control at Goldisthal amounts to 190–290 MW.
The power plant at Goldisthal is arranged to be split in half, with each side being a mirror image of the other. Each half of the
plant contains one synchronous and one asynchronous machine working together at one headrace tunnel. This allows operators to
take half of the plant off line at any one time for maintenance while the other half continues operating.

6.15.4.2.5 Operation to date

The Goldisthal plant was commissioned in October 2004. The variable-speed machines operate an average of 19 h a day in both
primary and secondary regulation. When operating under conditions of partial load, these units have an efficiency advantage of
about 10% as compared with the single-speed units.
An automatic controller on the two variable-speed machines constantly calculates and adjusts the units for optimal production,
based on the momentary head and the power output required.
The asynchronous machines can be started more quickly than the synchronous units because no fixed rotation speed is necessary
for synchronization of the variable-speed units. Starting from 95% of full synchronous speed, the frequency converter regulates its
parameters to the current speed and releases the unit to synchronization.
The operator expects overall maintenance expenditures to be lower for variable-speed units because of the smaller starting and
brake load operation, which is helped by the starting frequency converter. However, periods between maintenance of the
 Pumped Storage Hydropower Developments 429

variable-speed machines are expected to be somewhat shorter than those of the conventional machines. These units are inspected
every 4 weeks, as against every 6 weeks for the synchronous units. Moreover, inspection of the variable-speed units requires 10 h, as
compared with 8 h for the synchronous machines.
The higher frequency of inspection and the longer time required result to a large extent from the larger number and size of the
auxiliary systems associated with the asynchronous generator. For example, because the rotor of the asynchronous machine needs
more power and the voltage and current are much higher than in a synchronous rotor, the slip ring system is much bigger.
With regard to ancillary services, the asynchronous units at Goldisthal have been quite valuable. Because of the large capacity of
the units, a large regulation range is available. This is used daily for grid frequency control. The asynchronous machines can be
regulated from 40 MW up to 265 MW, while the synchronous machines can only be regulated from 100 to 265 MW. Thus, the
asynchronous machines provide 60 MW more for regulation. This allows the operator to take advantage of the lower basic power
output of 40 MW, saving water to be used for later generation.
In addition, the asynchronous machines have the ability to respond very quickly. If fast power is needed in the grid, the
asynchronous machines can retard their speed and supply additional braking energy to the grid (for a few seconds). In early
November 2006, parts of Europe experienced a long blackout. In the eastern part of Germany, frequency on the grid was 50.6 Hz,
whereas it normally is 50 Hz. The operator used the Goldisthal units to take energy out of the grid, and the asynchronous units were
used for regulation in pumping operation.
On average, about 70% of the working ability of the power plant is used each day. That results in daily production of
5500–7500 MWh during turbine operation.

6.15.4.3 Tianhuangping Pumped Storage Power Plant

6.15.4.3.1 Introduction
East China Electric Power’s Tianhuangping pumped storage hydroelectric project is the biggest of its type in Asia. It provides
valuable cover for demand surges in the central coastal region, including fast growing Shanghai. It is located in Anji County in
Zhejiang, about 175 km from Shanghai, and has a total installed capacity of 1800 MW [21] (Figure 24).

6.15.4.3.2 Two pump storage reservoirs

There are two pump storage reservoirs about 1 km apart, which have a difference in elevation of 590 m. Both have a storage volume
of 8 million m3. The lower reservoir is located on the Daxi Creek branch of the Xitiao river and the upper reservoir is an artificial
basin cut deep into the mountains. Two 7 m-diameter conduits from the upper reservoir each branch into three 3.2 m-diameter
pipes to power the six 306 MW turbines in the underground powerhouse. The maximum generating head is 607.5 m.
The lower reservoir is formed by a CFRD dam 95 m high, with an overall storage capacity of 8.6 million m3, a catchment area of
24.2 km2, and an average annual runoff of 27.6 million m3. The designed maximum pool level of the lower reservoir is 344.5 m. The
construction of the CFRD dam commenced in October 1993 and was completed in 1997. The upper reservoir is approximately
600 m higher, in a natural valley basin where the only opening is closed by a 72 m high rockfill dam, so as to maintain the 2315 m
long crest constantly at an elevation of 908 m. Four smaller saddle dams have been constructed to form the reservoir. The maximum
depth of the reservoir is 50 m. The slope and bottom of the upper reservoir have an asphalt concrete lining, which is best suited to

Figure 24 Tianhuangping project view.

430 Design Concepts

Figure 25 Tianhuangping project: Upper reservoir.

absorb the settlements and deformations of the dams without becoming permeable to water. The lining area amounts to
104 000 m2 for the reservoir bottom and 182 000 m2 for the reservoir slope, with a slope inclination of 1V:2–2.4H (Figure 25).
The substructure consists of a drainage layer made of crushed rock, the thickness of which is 90 cm on the slope and 60 cm on the
bottom. Bituminous emulsion was sprayed to stabilize the substructure surface and to achieve a better bond with the asphaltic
binder layer. The asphaltic binder layer is 10 cm thick on the slope and 8 cm on the bottom. The thickness of the impervious asphalt
concrete layer is 10 cm in both cases. In order to protect the asphalt concrete against aging as a result of ultraviolet radiation
associated with oxygen in the air, the slope and the bottom are provided with an asphalt mastic seal coating. In the curve at the
junction of the slope and the bottom and at the connections to the concrete structures, a 5 cm-thick protective layer of asphalt
concrete is applied, together with a polyester mesh reinforcement. Domestic bitumen products, of which several types were carefully
examined, had been ruled out due to their excessive paraffin content, which impairs the bonding characteristics of bitumen
(Figure 26)

6.15.4.3.3 Benefits of the project

The new plant plays a vital role in stabilizing the entire east China power grid, improving the quality of the power supply in east
China and ensuring the safe operation of the nuclear power stations in the surrounding areas (Figure 27).
From September 1998, when the first unit was put into operation, to the end of August 2008, the Tianhuangping power station
had generated 20.767 billion kWh electricity during peak times and absorbed 25.865 billion kWh during off-peak times. The overall
efficiency was 79.2%, which had exceeded the value of designed efficiency (74%); and the station occupied the world’s leading
position among all the pumped storage stations of its kind. The power station played an important role in peak shaving, frequency
modulation, and emergency duty in the network system, and effectively improved energy efficiency and overall efficiency of the
network system. As of August 2008, the power station had modulated system frequency during emergencies on more than 50
occasions and had played an important role in its mission of supplying power during the peak periods of every year, for instance,
during summer time and the time of major meetings.

Figure 26 Tianhuangping project: Lower reservoir.

 Pumped Storage Hydropower Developments 431

Figure 27 Tianhuangping project: Underground powerhouse.

6.15.4.4 Coo-Trois Ponts Pumped Storage Power Plant

6.15.4.4.1 Introduction
The Coo-Trois Ponts (Coo) Pumped Storage Plant is an underground facility located in the Ardennes section of Belgium. The total
rated capacity of the plant is 1164 MW provided by six single-stage, reversible pump–turbines. Coo was constructed in two stages.
The initial three units totaling 474 MW (Coo I) were put online in 1971. An additional three units totaling 690 MW (Coo II) were
put online in 1980. The overall head for the facility is 275 m. The unit hydrograph (UH) ratio varies between Coo I and Coo II from
slightly under 2 to slightly over 2. The initial stage of Coo included construction of an upper reservoir, power tunnel, underground
powerhouse, and tailrace tunnel. At the first stage itself the powerhouse was constructed to the full size required to accommodate
the second-stage expansion also. The second stage included construction of an additional separate upper reservoir, power tunnel,
and tailrace tunnel. The original design envisaged the three pump–turbine units installed during the second stage to be identical in
size to those in the first stage. Revised turbine designs, however, allowed for larger-capacity units to be installed within the
powerhouse, which originally had been designed for the smaller Stage 1 units. The overall layout of the project features is shown
in Figure 28 [22].

6.15.4.4.2 Main features of the project

The upper reservoirs consist of two asphaltic-lined systems with a total storage volume of 8 450 000 m3. This represents approxi­
mately 5 h of generation at full output. The intake structures at both Stage 1 and Stage 2 upper reservoirs are vertical and include a
vortex suppression plate (Figures 29 and 30).
The underground structures at Coo are located in the massif of Coo, a Paleozoic formation composed of phyllites,
quartzo-phyllites, and quartzites. These formations posed numerous problems during construction of the underground facilities.
The rock formation revealed expansive characteristics and required special treatment during construction. This treatment included
extremely long rock bolts, a temporary concrete tinning, and a final steel lining for all high-pressure water passages. The first- and
second-stage waterways consisted of a 6.6 m-diameter, steel-lined, vertical shaft, which connected to the power tunnel section with a
70% slope ending in a manifold upstream of the underground powerhouse. The first stage high-pressure section had an overall
length of 183 m. The second-stage tunnel was entirely separate from the initial-stage pressure tunnel and had larger waterway
openings to accommodate the larger-capacity units. This tunnel started with a vertical shaft, which extended to a sloping power
tunnel. The vertical-shaft tunnel has a diameter of 8 m and an overall length of 354 m. Flow to the units is controlled by spherical
valves with a diameter of 2.65 m. The transformers are sited outdoors in a valley adjacent to the powerhouse. The powerhouse
cavern is lined with concrete. Due to the expansive character of the rock and numerous fractures, the entire arch of the cavern
required a concrete lining. A sequenced excavation plan for the arch and the cavern area was instituted to allow for safe construction.
The multiple-stage construction resulted in two concrete-lined tailrace tunnels, each 245 m long with a 9 m diameter. The
submergence of the runners is approximately 18 m. A 210 m-long tunnel provides access to the downstream area.
Two embankments were constructed at either end of an ancient oxbow in the Ambleve River to create the lower reservoir. The
embankments were constructed of earth and rockfill with an asphaltic-concrete lining. The construction of these two embankments
created an essentially closed system, although the lower reservoir area can be dewatered by means of a dewatering conduit located at
432 Design Concepts

Figure 28 Layout of Coo project.

Figure 29 Intake structure of Coo 1.

the downstream end of the oxbow. The two lower-reservoir embankments are approximately 10 and 30 m high, respectively
(Figure 31).

6.15.4.4.3 Generating equipment

The generating and pumping equipment for the facility was designed to go from standstill to full generating load in 2.5 min and
from standstill to pumping in 6.5–7 min. Thanks to upgrades, those initiation times have been substantially reduced. The units are
started in the pumping mode using pony motors (Figure 32).
The project currently produces approximately 800 000–1 000 000 MWh per year in the generating mode and utilizes
1 000 000–1 333 000 MWh in the pumping mode. The facility is also utilized for synchronous condensing and spinning reserve
(Table 8).
 Pumped Storage Hydropower Developments 433

Figure 30 Intake structure of Coo 2.

Figure 31 Coo project: Upper and lower reservoirs view.

Figure 32 Coo project: Turbine view.

6.15.4.4.4 Special features of interest

The overall Coo-Trois-Ponts project was a staged development. The original planning allowed for the use of two upper
reservoir areas with separate high-pressure and tailrace tunnels. The initial construction of the powerhouse included excavation
of the total cavern area with sufficient placement of concrete to provide support for the second-stage development. A refined design
434 Design Concepts

Table 8 Coo Project: Performance of turbines

Max head & power Min head & power

Coo 1 Coo 2 Coo 1 Coo 2

Head Generating m 275.35 273.56 235.05 230.85

Pumping m 275.35 273.56 235.05 230.85
Capacity Generating MW 158 230 60 80
Pumping MW 135 190 145 200
Discharge Generating m3 s−1 60 100 23 36
Pumping m3 s−1 51 82 55 87

of the second-stage pump–turbine units allowed for an increase in size and capacity within the constraints of the original cavern and
powerhouse excavation limits.
Although the lower reservoir is interconnected with the Ambleve River, the system has many aspects of a closed system.

References

[1] Federal Energy Regulatory Commission, (FERC) http://www.ferc.gov/industries/hydropower/gen-info/regulation/pump.asp

[2] Price A (2000) Recent developments in the design and applications of a utility-scale energy storage plant. Proceedings of the Electrical Energy Storage Systems Applications and
Technologies International Conference 2000.
[3] Schoenung S (2001) Characteristics and technologies for long- vs. short-term energy storage. Sandia National Laboratories. Report SAND2001-0765. March 2001.
[4] Herr M (2002) Economics of Integrated Renewables and Hydrogen Storage Systems in Distributed Generation. London, UK: Imperial College of Science, Technology and

Medicine, University of London.

[5] Butler P, Miller J, and Taylor P (2002) Energy storage opportunities analysis. Phase II final report: A study for the DOE energy storage systems program. Sandia Report

SAND2002-1314. Unlimited Release. May 2002.

[6] Boyes J (2000) Energy storage systems program report for FY99. Sandia Report, SAND2000-1317, 2000.
[7] Baxter R and Makansi J (2002) Energy Storage. The Sixth Dimension of the Electricity Value Chain. Pearl Street Inc.
[8] Gandy S (2000) A Guide to the Range and Suitability of Electrical Energy Storage Systems for Various Applications, and an Assessment of Possible Policy Effects. Master Thesis,
Imperial College of Science, Technology and Medicine.
[9] Schoenung S (2002) Characteristics of energy storage technologies for short- and long-duration applications. Proceedings of the Electrical Energy Storage Systems Applications
and Technologies International Conference 2002 (EESAT2002), San Francisco, CA, 15–17 April.
[10] Swaminathan S and Sen R (1997) Electric utility applications of hydrogen energy storage systems. DOE/GO/10/70-T19, US Department of Energy.
[11] Ter-Gazarian A (1994) Energy Storage for Power Systems. Peter Peregrinus Ltd.
[12] Schoenung S (2001) Characteristics and Technologies for Long vs. Short-Term Energy Storage. Report SAND2001-0765, March 2001. Sandia National Laboratories.
[13] Gordon S and Falcone P (1995) The emerging roles of energy storage in a competitive power market: Summary of a DOE workshop. Sandia Report. SAND95-8247.
[14] Tanaka T and Kurihara I (1998) Market potential of utility-purpose energy storage in Japan up to the year 2050. Proceedings of the Electrical Energy Storage Systems Applications
and Technologies International Conference 1998 (EESAT98), Chester, UK, 16–18 June.
[15] www.voithhydro.com/…/VSHP090009_Pumped_Storage_72dpi.pdf
[16] Houdeline JB, Lavigne S, Mora P, et al. (2005) Single Stage Reversible Pump/Turbine – From Design to Experience. Austin, TX: Waterpower XIV.
[17] Houdeline JB, Verzeroli JM, Kunz T, and Schwer A (2010) Reversible Pump/Turbine and Generator-Motor Design The Last PSP Achievements Experience in Asia. IWPC.
[18] Electric Power Development Company (EPDC) www.jpower.co.jp/english/index.html
[19] MITI Ministry of Economy, Trade and Industry http://www.miti.gov.my/cms/index.jsp/
[20] Beyer Th. http://www.hydroworld.com/index/display/article-display/351208/articles/hydro-review-worldwide/volume-15/issue-1/articles/goldisthal-pumped-storage-plant­
more-than-power-production.html
[21] Chinese National Committee on Large Dams (2010) Current Activities. Dams Construction in China. IWHR.
[22] Energy Division of the American Society of Civil Engineers (1995) Hydroelectric Pumped Storage Technology: International Experience.
[23] Lejeune A and Pirotton M Exploitation de la Centrale de Coo. LHM, University of Liege.