energies

Review
Technical Review of Existing Norwegian Pumped
Storage Plants
Livia Pitorac 1, *, Kaspar Vereide 1,2 and Leif Lia 1
1 Hydraulic Engineering Research Group, Department of Civil and Environmental Engineering,
Norwegian University of Science and Technology, 7031 Trondheim, Norway; [email protected] (K.V.);
[email protected] (L.L.)
2 Sira-Kvina Hydropower Company, 4440 Tonstad, Norway
* Correspondence: [email protected]; Tel.: +47-998-66-885




Received: 16 June 2020; Accepted: 15 September 2020; Published: 19 September 2020


Abstract: This paper presents a technical review of the existing pumped storage plants in Norway.
The power system is changing towards integrating more and more renewable energy, especially
from variable renewable energy sources, leading to new challenges for the security of supply, power,
frequency, and voltage regulation. Thus, energy storage options are a highly researched topic in the
current situation. Even though there are many energy storage technologies, most are optimal for
short term grid balancing, and few are capable of providing long term (weekly or seasonal) storage.
One exception is pumped storage, a mature technology capable of delivering both short term and long
term energy storage. In this paper, the ten existing pumped storage plants in Norway are presented,
several of which are capable of seasonal energy storage. The Norwegian knowledge and experience
with pumped storage plants technology is provided as a basis for future research within the field.
The review provides information about energy production and storage capabilities, construction
costs, specific costs per kW and stored kWh, electromechanical installation, technical specifications,
and operational experience with focus on the design of the tunnel system layout. The data presented
in this review are unique and previously unpublished. A discussion and conclusions regarding
the current situation, trends, and future outlook for pumped storage plants in Norway within the
European power market are provided.

Keywords: hydropower; pumped storage; hydropower tunnel systems; seasonal energy storage;
renewable energy

1. Introduction
Growing concerns regarding the climate change have led to a worldwide shift of focus from
fossil fuels to renewable energy sources (RES) in order to reduce the environmental impacts of energy
generation. In Europe, the goal of reducing greenhouse gas emissions is legislated through policy
frameworks which set targets for energy consumption coming from renewable sources, starting in
1997, when the European Union (EU) set the 2010 targets. In 2018, under the “Clean energy for all
Europeans” package, the EU set its 2030 targets on use of RES to at least 32% through the revised
Renewable Energy Directive [1].
The main renewable energy sources constructed now are wind and solar, which are volatile and
unregulated sources, where the fluctuations in energy production do not align with the fluctuations
in energy demand. One way to eliminate this problem is to build many more wind and solar farms
than necessary (backup power plants), in order to ensure that the energy demand is always covered.
Another solution for eliminating the problem is energy storage, reducing the need for backup power
plants. A very wide variety of energy storage technologies are currently available or under research,

Energies 2020, 13, 4918; doi:10.3390/en13184918 www.mdpi.com/journal/energies

Energies 2020, 13, 4918 2 of 20

with the main ones being batteries, mechanical energy storage, hydrogen, and pumped hydro [2].
Batteries are a common solution for energy storage, having the advantage that they can be installed in
any location, they have a quick energy release capability and a high round- trip efficiency varying
between 70 and 95%, depending on the type of battery [3]. However, batteries can only store relatively
small amounts of energy, making them a small-scale energy storage solution, suitable for power
frequency and voltage regulation or hourly energy storage to help meet the peak demand. The lifetime
of batteries is also limited compared with competing storage technologies. Mechanical energy storage
is a technology using kinetic and gravitational energy to store energy. Compressed air energy storage
(CAES) in the mechanical energy storage technology in which air is pumped in caverns or tanks during
low energy demand periods. It is a mature technology, used for decades, cheap, and unlike batteries,
it does not involve any use of toxic materials. The round-trip efficiency of CAES varies between 40
and 70% [4]. Disadvantages of CAES is that it requires a location with suitable geology, and moreover,
the air needs to be heated during the energy generation, involving the use of fossil fuels in the diabatic
method. Hydrogen energy storage is a technology in which electricity is converted into hydrogen
through electrolysis, hydrogen is stored and later transformed back into electricity when the demand
requires it. Despite a low round-trip efficiency of less than 50% [5], hydrogen energy storage has a high
storing capacity comparing to all other energy storage technologies, being able to provide seasonal
and annual energy storage, which led to an increased research interest into further developing it.
This paper focusses on the most mature and currently most applied electrical energy storage
technology, pumped hydro. Pumped hydro stores energy in the form of water in a reservoir by pumping
it during low demand periods and later releases it to produce energy, with the round-trip efficiency
reaching above 80% depending on site-specific conditions [6]. Pumped hydro is able to provide seasonal
energy storage [7,8], and is currently the world’s largest energy storage technology [9]. Currently,
the technology is superior in both stored energy volumes and in power capacity. Further development
is being researched, and underground pumped hydro is a promising new technology that may enable
construction to be independent of topography and in combination with thermal storage, drinking water
storage, or desalination [10]. To compare the costs of pumped hydro with competing technologies,
this paper presents a calculation of constructions costs, specific costs per power capacity, and storage
capacity. Previous studies have shown that pumped hydro has the lowest costs of currently existing
storage technologies [11,12]. As can be seen from calculations presented in this paper, the Norwegian
pumped storage plants (PSPs) have a low specific cost per kW and a very low specific cost per stored
kWh compared to what is presented for other PSPs [7,10]. This is owing to beneficial topography that
reduce the costs of storage reservoirs. A comparison of specific costs and a discussion of the trends for
future development are presented in the discussion.
This main contribution in this paper is a technical review of the existing PSPs in Norway.
The Norwegian power network is currently interconnected with Sweden, Denmark, the Netherlands,
Russia, and Finland, and there are two more connections under construction, with Germany and the
UK, respectively. The power grid is operated as a state-owned monopoly, but the majority of the power
generation facilities are publicly owned. Moreover, Norway is currently the world’s sixth largest
producer of renewable energy from hydropower, with approximately 125 TWh per year, according to
the International Commission on Large Dams committee (ICOLD, Paris, France) [13]. The country has
around 1600 hydropower plants (HPPs) producing about 95% of the total electricity in the national
grid. Norwegian hydropower reservoirs hold approximately 50% of the total energy storage capacity
in hydropower reservoirs in Europe [14]. However, only ten pumped storage plants (PSPs) exist with a
total capacity of approximately 1400 MW. In this context, Norway has a large potential for expanding
its pumped storage capacity and contribute with energy storage on a European scale.
Similar reviews have been published for the Austrian PSPs [15], and the US PSPs [16].
Previous reviews from Norway were published concerning the strategy for pumped storage
plants [17,18] and about the cost and prospect potential by the Norwegian Water Resource and
Energy Directorate (NVE) [19], but no technical review with descriptions of the existing PSPs has been
Energies 2020, 13, 4918 3 of 20

published so far. Lia et al. [17] briefly presents the current state of PSPs in Norway and discusses
the former and future strategies for PSPs development in Norway in the light of lack of national
political solutions for power exchange on a European level, at the time. Ever since, progress has
been done in the field, with two subsea cables are currently under construction, linking Norway with
Germany (NordLink) and with UK (North Sea Link), with expected completion date in 2020 and
2021, respectively.
The current review covers the round-trip efficiency, construction costs, specific costs per kW
and per stored kWh, tunnel system design, electromechanical installation, technical specifications,
and operational experience. The review has a special focus on the design of the tunnel systems and
how it influences the hydraulic transients. A discussion concerning the state-of-the-art for pumped
storage plants and the future of pumped storage in Norway is provided.

Pumped Storage in Europe

Europe has the goal of becoming the first climate-neutral continent by 2050. In 2018, out of
the 11,970 TWh gross energy consumption, 2270 TWh came from renewable energy sources [20].
The electricity generation from wind and solar power increased from 2% to 15% of the total electricity
production from all sources between 2004 and 2018 [20]. Current research shows that the importance of
energy storage increases significantly with the rise in variable renewable energy being included in the
power system [21]. Currently, Europe had a total installed capacity of pumped hydro of 55 GW [22].
A total of 206 GW of long term energy storage with 30 TWh storage capacity is predicted to be installed
in Europe in 2050, for the 89% renewable energy scenario [23]. For a 100% renewable energy Europe
scenario, the storage need is estimated to range from 80 to 400 TWh, with installed capacity between
500 GW to 900 GW [24].
Austria, Switzerland, Norway, and Sweden have the largest available energy storage capacities
in PSPs [25]. The countries with the highest pumped storage installed capacity are Italy (7685 MW),
Germany (6364 MW), Spain (6117 MW), France (5837 MW), and Austria (5596 MW) [22]. Currently,
Norway is 10th in Europe in terms of pumped storage installed capacity, with 1369 MW, leaving it
with a high pumped hydro development capability, as Norwegian reservoirs equivalate nearly 87 TWh
of energy storage [14], with 10–20 TWh of available capacity most of the time [26]. Previous studies
showed the technical potential of developing additional capacity in terms of PSP without the need for
constructing new reservoirs [27]. An estimation of the technical and economic potential in Austria
yield that the country already exploited 75% of its hydropower potential, leaving it with 14 TWh
maximum unexploited potential [28]. The Swiss Energy Strategy 2050 framework estimates an increase
hydro capacity of 2 TWh, a target that can be reached only be finding hidden hydro potential [29].
In the case of Sweden, there is a 35 TWh expansion potential, currently limited due to environmental
reasons, leaving it with a final potential of 6 TWh when taking into account the current technological
development, which could eventually be increased with 2–4 TWh by upgrading current facilities [30].
The data about hydro potential in different counties shows that Norway has the largest unexploited
storage capacity that can serve as support for further integration of the variable renewable energy
sources into the European power system.

2. Historical Development of PSPs in Norway

The historical development of hydropower and PSPs in Norway is closely related with its industry
development. All ten PSPs are located in the Central and West Norway (Figure 1). The first PSP in
Norway is the 11 MW Brattingfoss power plant set in operation in 1955. This PSP was constructed
for seasonal pumping in a hydropower scheme where the largest reservoir is on top of the scheme.
Between 1962 and 1979, another five PSPs were built in Norway, with an installed capacity ranging
from 35 MW to 270 MW.
Energies 2020, 13, 4918 4 of 20

Energies 2020, 13, x FOR PEER REVIEW 4 of 20

Figure1.1.Map
Figure Mapof
ofpumped
pumpedstorage
storageplants
plants(PSPs)
(PSPs)in
inNorway.
Norway.

The largest PSP is the 640 MW Saurdal PSP (320 MW pumping), set in operation in 1985, as part
The largest PSP is the 640 MW Saurdal PSP (320 MW pumping), set in operation in 1985, as part
of the Ulla-Førre hydropower scheme. Included in the same scheme is also the smaller Stølsdal PSP,
of the Ulla-Førre hydropower scheme. Included in the same scheme is also the smaller Stølsdal PSP,
with an installed capacity of 17 MW. The Ulla-Førre hydropower scheme has a total installed capacity
with an installed capacity of 17 MW. The Ulla-Førre hydropower scheme has a total installed capacity
of 2100 MW, representing 6.4% of the total output in Norway. It supplies over 4.5 TWh annual energy
of 2100 MW, representing 6.4% of the total output in Norway. It supplies over 4.5 TWh annual energy
production, representing 3.5% of the total Norwegian annual electrical energy consumption.
production, representing 3.5% of the total Norwegian annual electrical energy consumption.
With the power market deregulation in 1991, a decrease in the development of large hydropower
With the power market deregulation in 1991, a decrease in the development of large hydropower
plants occurred in Norway. This is also observed in development of pumped storage plants, with only
plants occurred in Norway. This is also observed in development of pumped storage plants, with
two pumped storage plants built in the new regime. Some new projects were licensed, but the
only two pumped storage plants built in the new regime. Some new projects were licensed, but the
construction start was postponed for an unknown period. One project, Illvatn pumped storage plant
construction start was postponed for an unknown period. One project, Illvatn pumped storage plant
was recently licensed, with an expected output of 48 MW and 113 GWh per year [31]. The investment
was recently licensed, with an expected output of 48 MW and 113 GWh per year [31]. The investment
decision has currently not been taken.
decision has currently not been taken.
The reason for the reduced hydropower and PSP construction after the deregulation is mainly
The reason for the reduced hydropower and PSP construction after the deregulation is mainly
that the market was saturated, and supply exceeded demand [32]. Before the deregulation, the power
that the market was saturated, and supply exceeded demand [32]. Before the deregulation, the power
prices were mainly set in regional long term firm power contracts based on long term marginal cost
prices were mainly set in regional long term firm power contracts based on long term marginal cost
for the producers. Combined with obligations for power producer to secure power supply in their
for the producers. Combined with obligations for power producer to secure power supply in their
specific
specificregion
regionand
andlimited flexibility
limited in the
flexibility market,
in the thisthis
market, incentivized investments
incentivized in overcapacity
investments [32].
in overcapacity
These are also explanations to why there has been few new PSPs constructed after
[32]. These are also explanations to why there has been few new PSPs constructed after the the deregulation,
and why mostand
deregulation, Norwegian
why mostPSPs are constructed
Norwegian for seasonalfor
PSPs are constructed storage andstorage
seasonal why there are no
and why short
there are
term
no short term PSPs. Another reason is that the Norwegian power system, based on hydropowerlarge
PSPs. Another reason is that the Norwegian power system, based on hydropower with with
reservoirs, has significant
large reservoirs, accessaccess
has significant to power and energy
to power reserves,
and energy resulting
reserves, in relatively
resulting low prices
in relatively for
low prices
system services
for system suchsuch
services as frequency reserves.
as frequency reserves.
3. Technical Review
3. Technical Review
This section presents a technical review of the ten existing PSPs in Norway. The data in this
This section presents a technical review of the ten existing PSPs in Norway. The data in this
chapter are obtained from each PSP owner through questionnaires and interviews, in addition to
chapter are obtained from each PSP owner through questionnaires and interviews, in addition to
original unpublished design reports, documentation, and construction or as-built drawings of the
original unpublished design reports, documentation, and construction or as-built drawings of the
tunnel alignment, powerhouse, electromechanical units, and efficiency curves.
tunnel alignment, powerhouse, electromechanical units, and efficiency curves.
Norwegian PSPs are most commonly designed for seasonal storage. Due to the topography in
Norwegian PSPs are most commonly designed for seasonal storage. Due to the topography in
Norway, with steep slopes and high plateaus, the larger reservoirs are located in the upper part of the
Norway, with steep slopes and high plateaus, the larger reservoirs are located in the upper part of
catchment; thus, most PSPs are used to pump water to the upper reservoir during the snow-melting
the catchment; thus, most PSPs are used to pump water to the upper reservoir during the snow-
season, for storage to be used during the low-flow season. The common practice is to utilize natural
melting season, for storage to be used during the low-flow season. The common practice is to utilize
existing lakes and increase the water level with dams for creating the storage volume. Tunnel systems
natural existing lakes and increase the water level with dams for creating the storage volume. Tunnel
systems connect the reservoirs to an underground powerhouse. It is common practice to have several
brook intakes along the headrace and tailrace tunnels to collect water from smaller secondary water
streams.
Energies 2020, 13, 4918 5 of 20

connect the reservoirs to an underground powerhouse. It is common practice to have several brook
intakes along the headrace and tailrace tunnels to collect water from smaller secondary water streams.

3.1. Overview of the Pumped Storage Plants

Table 1 presents an overview of the Norwegian PSPs. The ten PSPs have a cumulative capacity of
1369 MW. All schemes are open loop schemes with natural inflow and have in sum a gross energy
production of 2.6 TWh per year. Considering the 0.8 TWh consumption for pumping, the PSPs have a
net energy production of about 1.8 TWh per year.

Table 1. PSPs in Norway.

Turbine Pump Gross Annual Pump Net Annual Gross

Commission
Name Capacity Capacity Production Consumption Production Head
Year
(MW) (MW) (GWh) (GWh) (GWh) (m)
Aurland III 270 258 350 280 70 400 1979
Brattingfoss 11 11 33 3 30 118 1955
Duge 200 170 303 55 248 220 1979
Herva 35 31 142 24 118 257 1962
Jukla 40 41 76 22 54 230 1974
Nygard 57.5 52 138 49 89 450 2005
Saurdal 640 320 1285 333 952 465 1985
Stølsdal 17 6 61 10 51 103 1986
Tevla 50 42 125 18 107 164 1994
Øljusjøen 49 39 78 50 42 1 212 1974
Sum 1369 997 2591 844 1761 - -
1 The upper reservoir in Øljusjøen can be used for production both in Øljusjøen PSP and Borgund HPP. Thus,
the gross production does not simply represent the sum between the net production and the consumption, as part of
the available water is actually used for energy production in Borgund HPP.

Table 2 presents the energy storage capacities in the PSP reservoirs. The upper reservoirs are
much larger compared with the downstream reservoir for all PSPs. This is due to the fact that most of
the PSPs were designed for pumping of inflow during flood season and not pumping of the stored
water in the downstream reservoir. The total storage capacity is over 5 TWh in the upper reservoirs
and 0.85 TWh in the lower reservoirs. Two columns presenting the equivalent number of days of
operation to empty or fill the reservoirs are presented. On average, it takes over 90 days with operation
on full capacity to fill the upper reservoirs, and only 22 days to fill the lower reservoirs.

Table 2. Energy storage in Norwegian PSPs.

Upper Reservoir Lower Reservoir

Name
Mill. m3 GWh 103 GWh/MW Mill. m3 GWh 103 GWh/MW

Aurland III 448 440 1556 10 10 36

Brattingfoss 107 31 2480 8 2 218
Duge 1398 755 3879 926 500 2570
Herva 109 69 1747 22 14 389
Jukla 236 116 2124 31 15 272
Nygard 103 114 1715 43 47 761
Saurdal 3105 3331 4978 230 247 737
Stølsdal 2.4 1 31 1 0.5 37
Tevla 204 82 1650 5 2 43
Øljusjøen 161 84 1518 27 14 328
Sum 5873 5023 21,678 1303 851 5391
Energies 2020, 13, 4918 6 of 20

3.2. Construction Costs and Specific Costs

The construction costs for each of the ten PSPs are estimated based on today’s prices to compare
with other technologies and the individual PSPs. The Norwegian national cost base for hydropower
has been applied to calculate the costs [33]. This cost base is regularly updated and is based on
statistical construction costs for hydropower projects in Norway. Table 3 presents the estimated costs
for the Norwegian PSPs. When compared with other published PSP costs, the Norwegian PSPs have
a low specific cost per kW and a very low specific cost per stored kWh [7,10]. The specific cost per
stored kWh is based on the energy storage capacity of the upper reservoir. This does not consider
the limitation of the lower reservoir, but this is regarded as acceptable owing to the natural inflow to
both upper and lower reservoir. Note that for comparison with other storage technologies and PSPs,
these number do not reflect the fact that the Norwegian PSPs are open loop type with a significant net
power production in addition. They also do not reflect the fact that most of these PSPs are located
on top of a larger hydropower system and provide a significant value for the cascade of hydropower
plants downstream.

Table 3. Construction costs and specific costs.

Construction Costs Specific Cost per kW Specific Cost per kWh

Name
(mill. €) (€/kW) (€/kWh)
Aurland III 212 787 0.51
Brattingfoss 30 2834 1.14
Duge 300 1501 0.39
Herva 60 1721 0.99
Jukla 146 3654 1.72
Nygard 41 739 0.43
Saurdal 995 1555 0.31
Stølsdal 64 3760 121.6
Tevla 103 2079 1.25
Øljusjøen 79 1612 1.06

The main reason for the large variation in specific cost is the role of each power plant, as most of
these projects are included in larger hydropower schemes where the dams and tunnel systems benefit
additional hydropower plants. As an example, the water pumped in Stølsdal PSP is to a large extent
used for production in other HPPs located downstream in the scheme; thus, the energy production in
Stølsdal specifically is very low. Its contribution to the total energy production of the hydropower
scheme is not quantified in this paper owing to a large number of variables and uncertainties.
In Figure 2 it can be observed a correlation between the specific cost per kWh and the upper
reservoir capacity, and head, respectively. The data included in the graphs are from nine of the PSPs.
The data point from Stølsdal PSP is excluded due to the unnaturally high specific cost owing to the
strategic placement in the larger scheme of hydropower plants, as explained above.
The data show trends with decreasing specific cost with increasing upper reservoir capacity and
head. This shows that optimal pumped storage plants are with high head and large upper reservoir,
as could be expected. The decrease in specific costs seems to be logarithmic and converging with
increasing reservoir capacity, and almost linearly decreasing with the increase of head. It can be noticed
that the specific cost is almost halved when the head increases from 200 m to 400 m. At some point this
trend must break of as it cannot continue to zero, but the breaking point cannot be found from our
range of data.
Energies 2020, 13, 4918 7 of 20
Energies 2020, 13, x FOR PEER REVIEW 7 of 20

Energies2020,
Energies 2020,13,
13,xxFOR
FORPEER
PEERREVIEW
REVIEW 77ofof20
20

Figure 2. Cost correlation analysis for the Norwegian PSPs.

Figure 2. Cost correlation analysis for the Norwegian PSPs.
Figure2.
3.3. Tunnel System and Hydrulic
Figure 2.Cost
Costcorrelation
correlationanalysis
Transients analysisfor
forthe
theNorwegian
NorwegianPSPs.
PSPs.

3.3. Tunnel
A System
3.3.schematic
andand
layout Hydrulic Transients
3.3. TunnelSystem
Tunnel System and of a typical
Hydrulic
Hydrulic Norwegian PSP is presented in Figure 3. During turbining, water is
Transients
Transients
transported
A schematic from the
layout upper
of of reservoir
a typical to the turbine isthrough theinheadrace 3.tunnel and the penstock
AAschematic
schematic layout
layout typicalNorwegian
of aatypical NorwegianPSP
Norwegian
PSP
PSPisis presented
presented
presented
Figure
inFigure
in Figure 3.3.During
During
During turbining,
turbining,
turbining, waterwater
water
and continues
is transported to
from the lower
thetheupperreservoir
reservoir though
reservoirto the the tailrace
turbine tunnel.
through During
the headrace pumping,tunnel it flows
and theinpenstock
reverse
isistransported
transported from
from the upper
upper reservoir totothe
theturbine
turbinethrough
through the
the headrace
headrace tunnel
tunnel andthe
and thepenstock
penstock
from
and and lower
continues
andcontinuesreservoir
continues to tothe
tothe to the
lower
the lowerupper
reservoir reservoir.
reservoirthough
thoughthe Along
the the tunnel
tailracetunnel.
thetailrace
tailrace tunnel.
tunnel. various
During
During shafts
pumping,
pumping, can be observed,
it flows in such as
reverse
lower reservoir though During pumping, ititflows
flows inreverse
in reverse
surge
fromfrom tanks
from
lower lower
lower and
reservoir brook
reservoir
reservoir totointakes.
tothe
the the upper
upper
upper Surge tanksAlong
reservoir.
reservoir.
reservoir. are constructed
Along
Along the tunnel
thetunnel
the tunnel in order
various
various
various shafts tocan
shafts
shaftsreduce
can bethe
beobserved,
can
be observed, pressure
observed,suchas
such asstrain
such as
on the
surge penstock
surge
tanks
surge tanks
tanksandand from
and
brook brook
brook water
intakes. hammer.
intakes.
intakes.SurgeSurge Brook
tanks
Surge tanks
tanks are areintakes
constructed
areconstructed
constructed(knows in
in in also
order
orderorder as
to secondary
reduce
to reduce
to reduce the intakes)
pressure
the pressure
the pressure are
strain used
on
strainstrain
on to
on
the thethe
transport penstock
water
penstock
penstock from from
from
from water
smaller hammer.
waterhammer.
water catchments
hammer. Brook Brook
Brook intakesintakes
along the
intakes(knows(knows
tunnel
(knowsalso also
system,
also as
as assecondary
for
secondary an
secondary intakes)
extra
intakes) inflow.are used
are used
intakes) to
Commonly,
are toused to
these transport
typeswater water
ofwater
intakes fromare smaller catchments
unregulated. Thealongalong
rock trapthetunnel
tunnel system,for foran an extra inflow.Commonly,
Commonly,
transport
transport from from smaller
smaller catchments
catchments along the in
the Norwegian
tunnel system,tunnels
system, an is
for extra normally
inflow.
extra inflow. located before
Commonly,
the thesetypes
these
penstock, typesin ofintakes
of intakesto
order are
are unregulated.
unregulated.
protect the Therock
The rocktrap
mechanical trap ininNorwegian
Norwegian
components tunnels
tunnels
from isnormally
is
fallen normally
rocks located
located
in unlined before
beforetunnels.
these types of intakes are unregulated. The rock trap in Norwegian tunnels is normally located before
thepenstock,
the penstock,in inorder
orderto toprotect
protectthe themechanical
mechanicalcomponents
componentsfrom fromfallen
fallenrocks
rocksin inunlined
unlinedtunnels.
tunnels.
Normally,
the penstock, a Norwegian
in order toPSP is located
protect entirely underground,
the mechanical components featuring
from fallen D-shape
rocks tunnels
in unlined constructed
tunnels.
Normally,
Normally, aa Norwegian
Norwegian PSP
PSP isis located
located entirely
entirely underground,
underground, featuring
featuring D-shape
D-shape tunnels
tunnels constructed
constructed
using
Normally,the drill and
a Norwegian blast method.
PSPmethod. Due
is located to the good
entirely rock quality,
underground, lining
featuring is needed just along short sections
usingthe
using the drilland
drill andblastblast method. Due Due totothe
thegood
good rockquality,
rock quality, liningisisD-shape
lining neededjust
needed
tunnels
justalong
along constructed
short
short
where
using the
sections tunnel
the drill whereand crosses
theblast
tunnel weakness
method. zones,
Due
crossesweakness
weaknessto thethe rest
good
zones, of
therock the
restof tunnel
quality,
of thetunnel being
tunnellining left
being isleftunlined
needed
left unlined with
just
with local
along rock
local short
sections where the tunnel crosses zones, the rest the being unlined with local
bolting
sections
rock and
where shotcrete
bolting the
andtunnel where
shotcrete necessary
crosses
whereweakness (Figure
necessary 4). the
zones,
(Figure 4). rest of the tunnel being left unlined with local
rock bolting and shotcrete where necessary (Figure 4).
rock bolting and shotcrete where necessary (Figure 4).

Figure
Figure
Figure3.3.3.Example
Exampleof
Example oftypical
of typicalNorwegian
typicalNorwegian
Norwegian PSP
PSP
PSP layout.
layout.
layout.

Figure 3. Example of typical Norwegian PSP layout.

Figure 4. Typical drill and blast tunnel in Norwegian PSPs: without lining (left) and with lined
section (right).
 Energies 2020, 13, x FOR PEER REVIEW 8 of 20
Energies 2020, 13, x FOR PEER REVIEW 8 of 20
EnergiesFigure
2020, 13, 4918
4. Typical drill and blast tunnel in Norwegian PSPs: without lining (left) and with lined section 8 of 20
Figure 4. Typical drill and blast tunnel in Norwegian PSPs: without lining (left) and with lined section
(right).
(right).
AAunique
uniquecharacteristic
characteristic of of Norwegian
NorwegianPSPs PSPsisisthe
theuse
useand
and placement
placement of of rock
rock traps
traps in the
in the tunnel
tunnel
systems. A As previously
unique mentioned,
characteristic of the
Norwegian hydropower
PSPs is the usetunnels
and are mainly
placement of unlined
rock
systems. As previously mentioned, the hydropower tunnels are mainly unlined (Figure 4); thus, the traps in(Figure
the tunnel4); thus,
the systems.
risk
risk of fallen As
of fallen previously
rocksrocks
being mentioned,
being the
transported
transported hydropower
to theto tunnels
the turbine
turbine are
needs
needs to mainly unlined
to be mitigated.
be mitigated. (Figure 4); thus,
In addition,
In addition, the
in most
in most cases,
risk
cases, of road
the fallenestablished
rocks being transported
in the to the
tunnel turbine
during needs to beismitigated.
construction not In addition,
removed, in most
therefore cases,power
during
the road established in the tunnel during construction is not removed, therefore during power plant
the
plant road established
operation, initthe tunnel during construction is not
theremoved, therefore during power plant to the
operation, partsparts
of it of
erode erode
and and are flushed
are flushed towards
towards turbines.
the turbines. In order
In order to avoid
to avoid damage
damage to the
operation, parts of it erode and are flushed towards the turbines. In order to avoid damage to the
penstock
penstockand andthethe mechanical equipment,aarock
mechanical equipment, rocktrap
trapand
anda afine
fine trash
trash rack
rack areare placed
placed upstream
upstream the the
penstock and the mechanical equipment, a rock trap and a fine trash rack are placed upstream the
penstock,
penstock,two twoexamples
examples being
being shown
shown in in Figure
Figure5.5.
penstock, two examples being shown in Figure 5.

Figure 5. Rock trap and trash rack: view from upstream (left) and view from downstream (right).
Figure 5. Rock trap and trash rack: view from upstream (left) and view from downstream (right).
Figure 5. Rock trap and trash rack: view from upstream (left) and view from downstream (right).
In most of the PSPs (Figure 6), there are also several brook intakes located along both the headrace
In most of the PSPs (Figure 6), there are also several brook intakes located along both the
Intailrace
most oftunnel.
and headrace
the the PSPs If the(Figure
location 6), of
there
the are
brook also several
intake brook intakes
is favorable, this islocated
designed along both theas a
to function
and the tailrace tunnel. If the location of the brook intake is favorable, this is designed to
headrace
surge tankand as the tailrace tunnel. If the location of theisbrook intake is favorable, thistunnel
is designed to in
function aswell,
a surge otherwise,
tank as well, a separate
otherwise, surge tank
a separate built,
surge tank if is
necessary. The PSP
built, if necessary. systems
The PSP tunnel
function
Norway as a surge
are long, tank
varying as well, otherwise, a separate surge tank is built, if necessary. The PSP tunnel
systems in Norway arebetween
long, varying2 km and 17 km;
between 2 kmthus,andsurge tanks
17 km; are surge
thus, normallytanksconstructed
are normally in order
tosystems
reduce in Norway
the
constructed effect aretowater
of the
in order long,
reduce varying
hammer.
the effectbetween
Asofathe 2 kmhammer.
consequence,
water and mass17 km; a thus,
Asoscillations surge
consequence, tanks
occur are
in the
mass normally
system
oscillations which
constructed
result intointhe
occur inneed
the orderfor
system to areduce
which theinto
well-analyzed
result effect of
thedesignthefor
need water
ofathe hammer.
surge tank.
well-analyzed Asdesign
aInconsequence,
many surgemass
of theprojects, oscillations
the
tank. Inpreliminary
many
occur in thethe
design of thesystem
projects, surge which result
preliminary
tank sizedesign
is doneinto
of thethesurge
using needtank
thefor asize
Thoma well-analyzed
is stability
done using design
the Thoma
criteria of thestability
[34]. surge tank.
The In[34].
criteria
design of many
the surge
projects, the preliminary
of the surgedesign isof the surge tank size is done using the Thoma stability criteria [34].
tankThe
is adesign
quite straightforward tank processa quiteif straightforward
the surge tank process is the onlyif the surge
shaft in tank is the only
the system [35].shaft in
In Norway,
Thethe
design
system of[35].
the surge tank is
In Norway, thea tunnel
quite straightforward
systems are oftenprocess if the surge
more complex with tank is the
multiple onlyintakes
brook shaft in
the tunnel systems are often more complex with multiple brook intakes and unplugged adits along
theand
system [35]. Inadits
unplugged Norway,
alongthe thetunnel systemsthus,
main tunnels; are often more may
the design complex requirewith multiple
a more brook
refined intakes
analysis.
the main tunnels; thus, the design may require a more refined analysis. Commonly, the surge tank
andCommonly,
unpluggedthe adits
surgealong
tanktheis a main tunnels;surge
two-chamber thus,tank
the type
design may require
(Brattingfoss, a more
Duge, refined
Øljusjøen) or analysis.
a shaft
is a two-chamber
with upper
surge tank type(Herva, (Brattingfoss, Duge, Øljusjøen) or a shaft with upper expansion
Commonly, theexpansion
surge tankchamber
is a two-chamber Nygard,
surge tank Tevla).
typeIn(Brattingfoss,
one case, Jukla PSP,Øljusjøen)
Duge, an underground
or a shaft
chamber
closed (Herva,
surge Nygard,
tank filled Tevla).
with In one
pressurized case,
air Jukla
is PSP,
applied. an
Goodallunderground
with upper expansion chamber (Herva, Nygard, Tevla). In one case, Jukla PSP, an underground et al. [36] andclosed
Vereidesurge
[37]tank filled
present a with
pressurized
closed surge tank filled with pressurized air is applied. Goodall et al. [36] and Vereide [37] present a of
more air
detailed is applied.
description Goodall
of closed et al.
surge [36]
tank and
designVereide
in [37]
Norway. present a more detailed description
closed
more surge
detailed tank design inofNorway.
description closed surge tank design in Norway.

Figure 6. Cont.
 Energies2020,
Energies 2020,13,
13,4918
x FOR PEER REVIEW 9 of 20
9 of 20

Figure 6. Cont.
Energies 2020, 13, 4918 10 of 20
Energies 2020, 13, x FOR PEER REVIEW 10 of 20

Figure 6. The
Figure The tunnel
tunnelsystem
systemlayouts
layoutsofofthe
thepumped
pumpedstorage plants
storage in Norway.
plants in Norway.

3.4.
3.4.Electromechnical
Electromechnical Installation
Installation
An
Anoverview
overviewof
of the
the mechanical installationand
mechanical installation andelectrical
electrical equipment
equipment in in
thethe Norwegian
Norwegian PSPsPSPs
are are
shownininTables
shown Tables44and
and 5,
5, respectively.
respectively. Out
Outofofthe
theten
tenPSPs
PSPsininNorway,
Norway,seven
seven have
havereversible pump
reversible pump
turbines(RPTs)
turbines (RPTs)and
andthree
three have
have separate
separatepump
pumpandandturbine
turbineunits. Generalized,
units. Generalized,there are are
there fivefive
different
different
start-upprocedures
start-up proceduresfor
for pumping
pumping mode:
mode:
Energies 2020, 13, 4918 11 of 20

1. In air with pony motor, soft-starter, or frequency converter.

2. In water with electrical back-to-back start with a generator.
3. In water with mechanical back-to-back start with a turbine.
4. In water with frequency converter.
5. Direct start.

Table 4. Overview of mechanical equipment in Norwegian PSPs.

Turbine Type and Number of Start-Up Procedure in Pump Start-Up Time

Name
Units Pumping Mode (Minutes)
In air with an 11.4 MVA1
Aurland III 2 vertical RPT 9 min
frequency converter
1 horizontal Francis turbine In air, mechanical
Brattingfoss N.A.
and 1 pump back-to-back
In air with 5 MW pony
Duge 2 vertical RPT 15 min
motor
1 horizontal unit with
2 runners (pump and turbine)
Herva Mechanical back-to-back 10 min (first hours)
and 1 machine
(motor-generator)
Jukla 1 vertical RPT Pony motor of 4.3 MVA 11 min
In air with frequency
Nygard 1 vertical RPT 6.5 min
converter
2 vertical Francis turbines and
Saurdal Electrical back-to-back 7 min
2 vertical RPT
Stølsdal 1 Francis turbine and 2 pumps Direct start N.A.
In air with frequency
Tevla 2 vertical RPT N.A.
converter
Øljusjøen 1 vertical RPT Direct start 3 min
1 MVA = megavolt-ampere.

Table 5. Overview of electrical equipment in Norwegian PSPs.

Generator Output Motor Consumption Speed of Rotation Transformer

Name
(MVA) (MW) (MVA) (MW) (RPM) (kV/kV)
Aurland III 2 × 150 2 × 135 2 × 150 2 × 126 500 420/15.5
Brattingfoss 14 11 14 10.6 428 66/6.3
Duge 2 × 120 2 × 100 2 × 106 2 × 85 375 320/13
Herva 45 35 32 31 500 132/8
Jukla 44 40 48 41 500/375 67/12
Nygard 65 57.5 65 52.3 750 300/11.4
Saurdal 4 × 185 4 × 160 2 × 185 2 × 160 428 324/18.5
Stølsdal 20 17 N.A. 2×3 375 300/6.6
Tevla 2 × 30 2 × 24.8 2 × 30 2 × 21.1 500 132/8.8/4.4
Øljusjøen 55 49 50 38.6 428 300/7

For the first pump startup procedure, air is introduced in the pump with compressors, forcing the
water out of the spiral casing. The units are then started in pumping mode with the pump rotating in
air. The rotor is accelerated using a pony motor, soft starter, or frequency converter until the full speed
of rotation is reached. When the speed of rotation is set to synchronous speed, air is released, water is
admitted back inside the spiral casing, and the pump operation starts.
Energies 2020, 13, 4918 12 of 20

In the second procedure, the electrical back-to-back start, a pump unit is started using a nearby
generator
Energies 2020,unit. WithPEER
13, x FOR the REVIEW
two machines (generator and pump motor) being connected through 12 the
of 20
electrical system, both are excited with a current. The turbine runner starts to rotate; thus, the frequency
frequencywhich
increases, increases, which
triggers triggers
the motor the motorastowell,
to accelerate accelerate
until it as well, until with
synchronizes it synchronizes
the generator with the
speed
generator
of rotation. speed
When of therotation.
motor reaches When the thesynchronous
motor reaches theofsynchronous
speed rotation, it isspeed of rotation,
disconnected from ittheis
disconnected
generator, from theunit
the turbine generator,
is shut down,the turbine unit
and the is shut down,
pumping and the pumping commences.
commences.
Thethird
The thirdstart-up
start-upprocedure,
procedure,mechanical
mechanicalback-to-back
back-to-backstartstartisisimplemented
implementedwith withaamechanical
mechanical
connection between
connection between the theturbine
turbineand andpump.
pump. ThisThis can
can be
be achieved
achieved in in PSPs
PSPs where
where thethepump
pumpand andthethe
turbinerunners
turbine runnersare areon
onthethesame
sameshaft.shaft.In Inthis
thisprocedure,
procedure,first
firstthe
theturbine
turbineisisstarted,
started,accelerating
acceleratingthe the
entireunit
entire unitto tothe
thenominal
nominalspeed speedof ofrotation.
rotation.After
Afterthe
thesynchronization,
synchronization,the theinlet
inletvalve
valveof ofthe
theturbine
turbine
closeswhile
closes whilethe thepump
pumpvalve
valveopensopensininparallel,
parallel,and
andthe
thepump
pumpoperation
operationstarts.
starts.
Thefourth
The fourthstart-up
start-up procedure
procedure is the
is the mostmost modern
modern one,one,
usingusing frequency
frequency converters
converters to starttoinstart
water.in
water.
With thisWith this technique,
technique, the units the can units can be connected
be connected to the grid toeventheifgrid evenis ifnot
the unit the
at unit is not at the
the synchronous
synchronous
RPM, resulting RPM,
in a resulting in a gentler
gentler starting withstarting with less momentum.
less momentum. Several variantsSeveralexist
variants
suchexist such as
as full-size
full-size converters,
converters, part-size converters,
part-size converters, transistor,transistor, and thyristor-based
and thyristor-based technology. technology.
Thefifth
The fifthprocedure,
procedure, direct
direct startstart is a brute
is a brute connection
connection of the motor
of the pump pumptomotor to from
the grid the grid from
standstill.
standstill.
This methodThis method
is the simplestis the
andsimplest and most
most traditional traditional
one, one, the
in which both in which
grid andboththethe grid
unit haveandto the unit
sustain
ahave
highto sustainload,
start-up a high start-up
making load, making
it suitable just for it suitable
small units,just for small
where both theunits,
gridwhere both
and the the grid can
machinery and
the machinery
withstand it. Thiscan withstand
procedure it. This procedure
is characterized is characterized
by high starting torquebyand high
fullstarting
voltagetorque and full
and frequency
voltage
from the and frequency
beginning. from cases,
In some the beginning.
the pumpIn some
starts cases,
with the the
main pump
valvesstarts withand
closed, thefirst
main valves
opening
closed, and first opening when
when the normal operating pressure is reached. the normal operating pressure is reached.
Figure77shows
Figure showsthe theefficiency
efficiencycurves curvesin inturbine
turbinemode
modefor fortentenofofthe
theinstalled
installedturbines.
turbines.The Thebest
best
efficiencypoint
efficiency point(BEP)
(BEP)of ofeach
eachturbine
turbinevariesvariesbetween
between90.4%
90.4%andand93.5%,
93.5%,and andthethepower
powerspecific
specificspeed
speed
variesbetween
varies between0.33 0.33and
and0.880.88radians.
radians.The Thebest
bestefficiency
efficiencypoint
pointduring
duringpumping
pumpingisisknownknownfor for77ofofthe
the
RPTs/pumpsand
RPTs/pumps andititvaries
variesbetween
between87.4% 87.4%and and90.8%,
90.8%,with
withonly
onlytwo twopumps
pumpshaving
havingBEPBEPunder
under88.2%88.2%
andthe
and theother
otherfivefivehaving
havingthe theBEPBEPaboveabove90.6%.
90.6%.

Figure 7. Anonymized efficiency for a selection of the turbines in the described PSPs, function of unit
Figure
flow 7. Anonymized
(left) efficiency
and function of for a selection
power specific of the turbines in the described PSPs, function of unit
speed (right).
flow (left) and function of power specific speed (right).
The round-trip efficiency for the ten PSPs varies between 65% and 80%. The waterway head
loss inThe round-trip
each efficiency
PSP is calculated for the
using an ten PSPs varies
assumed between 65%number
Manning–Strickler and 80%.ofThe
M =waterway
33 for thehead loss
unlined
in each PSP is calculated using an assumed Manning–Strickler number of M = 33
tunnels, and M = 85 for the steel lined penstocks. The waterway is usually one of the main causes for the unlined
tunnels,
of energyand M =ranging
losses 85 for the steel
from lined
less thanpenstocks. The waterway
1% (Nygard) and up to is15%usually onedepending
(Duge) of the mainoncauses of
tunnel
energy losses
lengths. ranging electromechanical
The round-trip from less than 1%losses(Nygard) and the
including up transformation
to 15% (Duge) in depending on tunnel
the range from 20%
lengths. The round-trip electromechanical losses including the transformation in the
to 25%. The generator-motor and transformer efficiencies are assumed standard values of 98% and range from 20%
to 25%.
99%. TheThe
realgenerator-motor
turbine and pump and transformer
efficiencies efficiencies
as presented are anonymized
in the assumed standard
graphsvalues of 98%
above are usedand
in
99%. The real turbine and pump efficiencies as presented in the anonymized graphs above are used
in the calculations. These calculations assume operation of full capacity, which is conservative as this
generates the highest waterway head losses.
It is noted that all of the Norwegian PSPs are designed primarily for seasonal storage and
pumping of water during the spring and autumn high flow seasons. Most of the power plants were
not designed for frequent start–stop pump operations, which are reflected in the relatively time-
Energies 2020, 13, 4918 13 of 20

the calculations. These calculations assume operation of full capacity, which is conservative as this
generates the highest waterway head losses.
Energies
It 2020, 13, x that
is 2020,
noted FOR all
PEERof REVIEW
the Norwegian PSPs are designed primarily for seasonal storage and pumping 13 of 20
Energies 13, x FOR PEER REVIEW 13 of 20
of water during the spring and autumn high flow seasons. Most of the power plants were not designed
consuming start-up procedures, varying between 6.5 min and few hours, in the case of old ones, and
for frequent start–stop
consuming pump operations,
start-up procedures, varying which are reflected
between in the
6.5 min and fewrelatively
hours, in time-consuming start-up
the case of old ones, and
only going down to 2.3 min for the recently upgraded ones. The pump startup time is known for
procedures,
only goingvaryingdown tobetween 2.3 min6.5 formin
the and few hours,
recently upgraded in the caseThe
ones. of old
pump ones, and only
startup timegoing down
is known for
seven PSPs, out of which five have a startup below 10 min, meaning that they are be able to provide
toseven
2.3 min for out
PSPs, the of recently
which upgraded
five have aones.
startupThebelow
pump10startup time is known
min, meaning that they forare
seven PSPs,
be able out of
to provide
tertiary frequency reserves [38].
which fivefrequency
tertiary have a startup below
reserves [38].10 min, meaning that they are be able to provide tertiary frequency
reserves [38].
3.5. Particularities
3.5. Particularities
3.5. Particularities
Each of the ten PSPs have certain particular design features worth to mention. This section
Each of the ten PSPs have certain particular design features worth to mention. This section
presents
Eachsome theof
ofsome tenthe most interesting from a hydraulic pointworth
of view. In the Brattingfoss
sectionPSP, the
presents of PSPs have
the most certain particular
interesting from adesign features
hydraulic to mention.
point of view. In the This
Brattingfoss presents
PSP, the
headrace
some tunnel
of thetunnel crosses
most interesting a steep valley,
from avalley, similar
hydraulic to
point a narrow canyon. In this area, an overground
headrace crosses a steep similar to of view. Incanyon.
a narrow the Brattingfoss PSP,an
In this area, theoverground
headrace
suspended
tunnel crosses pipea connects
steep the
valley, upstream
similar to a reservoir
narrow with In
canyon. thethis
headrace
area, an tunnel as seen
overground in Figure
suspended pipe8.
suspended pipe connects the upstream reservoir with the headrace tunnel as seen in Figure 8.
Another
connects feature of the Brattingfoss PSP is the design of the unit, which has both the pump and turbine
Anotherthe upstream
feature of thereservoir
BrattingfosswithPSP
theisheadrace
the designtunnel
of theasunit,
seenwhich
in Figure 8. Another
has both the pumpfeature
and of the
turbine
runners connected
Brattingfoss PSP is to thethe sameof motor-generator
unit, which on the sametheshaft. A horizontal
turbine sketch of the unit can
runners connected to design
the same themotor-generator has
onboth
the same pump
shaft.andA horizontal runners
sketch connected
of the unit to
can
be
theseen in Figure 9.
besame
seen motor-generator
in Figure 9. on the same shaft. A horizontal sketch of the unit can be seen in Figure 9.

Figure 8. Pipe
Pipe section in Brattingfoss PSP.
Figure 8.8.
Figure Pipesection
sectionininBrattingfoss
BrattingfossPSP.
PSP.

Figure 9.
Figure Electromechanicallayout
9. Electromechanical layoutplan
planview
viewin
inBrattingfoss
Brattingfoss PSP,
PSP, planar view.
Figure 9. Electromechanical layout plan view in Brattingfoss PSP, planar view.
The Duge
The Duge PSP PSP features
features aa special
special intake
intake design
design at at the
the upstream
upstream reservoir.
reservoir. AsAs shown
shown in inFigure
Figure 10,
10,
there The
is a Dugeand
higher PSPafeatures
lower a special
intake. Bothintake
intakesdesign
can at the
be used upstream
in turbine reservoir.
mode, As shown
while during inpumping
Figure 10,
there is a higher and a lower intake. Both intakes can be used in turbine mode, while during pumping
there only
mode, is a higher and aintake
the upper
upper lower intake. Both intakes can be usedthat in turbine mode, while during pumping
mode, only the intake isisused.
used.TheThereason
reasonfor forthis
thisis is
that thethe pump
pump needs
needs a minimum
a minimum head head
in
inmode,
order only
to the
operate,upperwhichintake is
cannot used.
be The reason
fulfilled when for
the this is
water that the
level pump
in the needs
upstream a minimum
reservoir head
is low.in
order to operate, which cannot be fulfilled when the water level in the upstream reservoir is low. In
order to
In situations operate,
with which
low watercannot be
levels, fulfilled when the water level in the upstream reservoir is low. In
situations with low water levels, thethe gate
gate ofofthe
thelower
lowerintake
intakeisisclosed,
closed,and
andwater
waterisis pumped
pumped only only
situations
through the with
the upper low
upper one, water
one, creatinglevels, the
creating aa waterfallgate of
waterfall down the
down into lower
into the intake
the reservoir.is closed,
reservoir. Owing and
Owing to water
to the is pumped
the resulting
resulting energy
energyonly
through
through
loss, such the
such situationsupper one,
situations occur creating
occur very a
very seldom. waterfall
seldom. ItIt can down
can also
also be into
be noted the
notedthat reservoir.
thatthe
theLRWL Owing
LRWL is to
is below the
below the resulting
thelower energy
lowerintake,
intake,
loss,
loss,
the suchbeing
reason situations
that occur
the very
last seldom.
volume of It cancan
water alsobebereleased
noted that to the LRWL
power is below
plants located thedownstream
lower intake,
the reason being that the last volume of water can be released to power plants located downstream
the reason
through bottom being that the
outlets, last volume
in case
case severeofdraughts.
of severe water can be released to power plants located downstream
through bottom outlets, in of draughts.
through bottom outlets, in case of severe draughts.
Energies 2020, 13, 4918
x FOR PEER REVIEW 14 of 20
Energies 2020, 13, x FOR PEER REVIEW 14 of 20

Figure 10. Upstream intakes in Duge PSP.

Figure 10.Upstream
Upstreamintakes
intakesininDuge
Duge PSP.

The Herva PSP has two upstream reservoirs with different water levels; Hervavatn is regulated
The Herva PSP has two upstream reservoirs with different water levels; Hervavatn is regulated
between 1302 and 1287 masl, and Storevatn is regulated between 1270 and 1244 masl. The reservoirs
between 1302 and 1287 masl, and Storevatn is regulated between 1270 and 1244 masl. The reservoirs
are used alternatively depending on the available reservoir volume. This gives a better flexibility in
are used alternatively depending on the available reservoir volume. This gives a better flexibility in
operation, and a larger total storage capacity.
capacity.
operation, and a larger total storage capacity.
The Jukla PSP has several interesting features. It has four upstream reservoirs:
reservoirs: Juklavatn (1060
The Jukla PSP has several interesting features. It has four upstream reservoirs: Juklavatn (1060
to 950 masl), Dravladalsvatn (957 to 880 masl), Jukladalsvatn (1083 to 990 masl) and Langavatn (962
to 950 masl), Dravladalsvatn (957 to 880 masl), Jukladalsvatn (1083 to 990 masl) and Langavatn (962
to 927 masl). InIn addition,
addition, water
water from
from several
several brook
brook intakes
intakes and transfer reservoirs
reservoirs is diverted to
to 927 masl). In addition, water from several brook intakes and transfer reservoirs is diverted to
Downstream of
Dravladalsvatn using a series of diversion tunnels and channels. Downstream of the
the PSP,
PSP, the water
Dravladalsvatn using a series of diversion tunnels and channels. Downstream of the PSP, the water
diverted to
is diverted totwo
twodownstream
downstreamreservoirs,
reservoirs,Svartedalsvatn
Svartedalsvatn (860
(860 to to
834834 masl)
masl) andand Mysevatn
Mysevatn (855(855 to
to 775
is diverted to two downstream reservoirs, Svartedalsvatn (860 to 834 masl) and Mysevatn (855 to 775
775 masl),
masl), which
which serveserve
both both as intake
as intake reservoirs
reservoirs for pumping
for pumping modemode and
and for for a downstream
a downstream powerpower
plant.
masl), which serve both as intake reservoirs for pumping mode and for a downstream power plant.
plant.
Another Another particularity
particularity of Jukla
of Jukla PSP PSP is use
is the the use of closed
of closed surge
surge tank
tank (Figure11)
(Figure 11)for
forcontrolling
controlling the
Another particularity of Jukla PSP is the use of closed surge tank (Figure 11) for controlling the
The closed surge tank has a total volume of 5500 m33, 3with an absolute pressure
pressure transients. The
pressure transients. The closed surge tank has a total volume of 5500 m , with an absolute pressure
varying between 675 and 228 mWC, depending on which reservoirs are active [39]. Finally, Finally, owing to
varying between 675 and 228 mWC, depending on which reservoirs are active [39]. Finally, owing to
the large variation in water level between the different reservoirs, the generator has the possibility to
the large variation in water level between the different reservoirs, the generator has the possibility to
short-circuit some of the poles to allow operation at either
either 500
500 RPM
RPM or or 375
375 RPM.
RPM.
short-circuit some of the poles to allow operation at either 500 RPM or 375 RPM.

Figure 11.
Figure Jukla air
11. Jukla air cushion
cushion surge
surge tank
tank and
and powerhouse
powerhouse detail.
detail.
Figure 11. Jukla air cushion surge tank and powerhouse detail.
The Nygard PSP is used to pump water from Stølsvatnet (584 to 547 masl) up to Skjerjavatnet
The Nygard PSP is used to pump water from Stølsvatnet (584 to 547 masl) up to Skjerjavatnet
(964 toThe
944 Nygard PSP issummer
masl) during used toseasons
pump water
and is from
used Stølsvatnet (584 to 547during
for power production masl) up to Skjerjavatnet
winter. While the
(964 to 944 masl) during summer seasons and is used for power production during winter. While the
(964 to 944 masl) during summer seasons and is used for power production
PSP is part of the Modalen river, Skjerjavatnet is part of Eksingedalen river. Thus, in orderduring winter. While the
to use
PSP is part of the Modalen river, Skjerjavatnet is part of Eksingedalen river. Thus, in order to use
PSP is part of
Skjerjavatnet as the Modalen reservoir,
an upstream river, Skjerjavatnet
the original is part
lake of Eksingedalen
outlet was dammed, river.water
Thus,being
in order to use
diverted
Skjerjavatnet as an upstream reservoir, the original lake outlet was dammed, water being diverted
Skjerjavatnet as an upstream reservoir, the original lake outlet was dammed, water
through a tunnel to the PSP and further to Stølsvatnet in Modal river. In addition to the two reservoirs,being diverted
through a tunnel to the PSP and further to Stølsvatnet in Modal river. In addition to the two
through
water fromaeight
tunnel to the
brook PSPisand
intakes used.further to Stølsvatnet in Modal river. In addition to the two
reservoirs, water from eight brook intakes is used.
reservoirs, waterPSP
The Saurdal fromiseight
part brook intakes isthe
of Ulla-Førre, used.
largest hydropower system in Norway, including
The Saurdal PSP is part of Ulla-Førre, the largest hydropower system in Norway, including
The Saurdal PSP is part of Ulla-Førre,
Blåsjø, the largest reservoir in Norway, with about the largest
8 TWh hydropower system serving
of energy storage, in Norway, including
as upstream
Blåsjø, the largest reservoir in Norway, with about 8 TWh of energy storage, serving as upstream
Blåsjø, the largest reservoir in Norway, with about 8 TWh of energy storage,
reservoir. Blåsjø was created by rising the levels in three natural lakes; Førrevatn, Oddatjern serving as upstream
and
reservoir. Blåsjø was created by rising the levels in three natural lakes; Førrevatn, Oddatjern and
reservoir. Blåsjø was created by rising the levels in three natural lakes; Førrevatn,
Storvatn, which are located in two different river schemes. Saurdal has four units of 160 MW, whereof Oddatjern and
Storvatn, which are located in two different river schemes. Saurdal has four units of 160 MW, whereof
Storvatn, which are located in two different river schemes. Saurdal has four units
two of them are reversible pump turbines. Another characteristic to Saurdal is a U-tunnel section of 160 MW, whereof
two of them are reversible pump turbines. Another characteristic to Saurdal is a U-tunnel section
two the
along of them are reversible
headrace tunnel (seepump
Figureturbines.
12). As inAnother
the casecharacteristic to Saurdal
of Brattingfoss, is a U-tunnel
the headrace section
tunnel crosses
along the headrace tunnel (see Figure 12). As in the case of Brattingfoss, the headrace tunnel crosses
along the headrace tunnel (see Figure 12). As in the case of Brattingfoss, the headrace tunnel crosses
a steep valley, but the solution in this case was the construction of a U-tunnel as seen in the figure. In
a steep valley, but the solution in this case was the construction of a U-tunnel as seen in the figure. In
Energies 2020, 13, 4918 15 of 20

Energies 2020, 13, x FOR PEER REVIEW 15 of 20

a steep valley, but the solution in this case was the construction of a U-tunnel as seen in the figure.
In addition,
addition, thethe upstream
upstream surge
surge tank
tank in in Saurdal
Saurdal PSPisisalso
PSP alsoused
usedasasaabrook
brookintake,
intake, with
with aa maximum
3 /s. This makes Saurdal PSP a good example of using a brook intake as a surge
inflow of 2.1 m /s. This makes Saurdal PSP a good example of using a brook intake as a surge tank in
3

Norwegian PSPs.

Figure
Figure 12.
12. U-tunnel
U-tunnel in
in Saurdal
Saurdal PSP.
PSP.

The Stølsdal
The Stølsdal PSP
PSP is smaller
smaller in terms of reservoir capacities, capacities, but but it it features
features 19 shafts along the
headrace tunnel, mostly including brook
headrace tunnel, mostly including brook intakes, and intakes, and just one surge tank. The other
The other 18 18 shafts
shafts are
diverting water
diverting water from
from 22 22 brook
brook intakes
intakes to the headrace
headrace tunnel,
tunnel, out out of which, four collect collect water
water from
lakes with minor regulation. This This makes
makes the the Stølsdal
Stølsdal PSP
PSP the the most
most complex
complex PSP PSP in in Norway
Norway from from the
hydraulic point
hydraulic point of
of view. The
The maximum
maximum total total inflow
inflow inin the 22 brook intakes with a mean cumulated
discharge of 15 m33/s;/s; considering
considering that that the maximum turbine capacity is 22 m33/s,
the maximum /s, during
during snow snow melting
melting
rainfall, the water from the brook intakes cover over 70% of the turbine capacity.
or rainfall, capacity. The General water
general water
necessary minimum
level and necessary minimum head head areare controlled
controlled using
using two two upstream
upstream reservoirs,
reservoirs, Bjørndalsvatn
Bjørndalsvatn (708 (708
and 697 masl) and Sandsavatn (605 and 560 masl).
The Tevla PSP is used to prevent flood loss by pumping pumping water water toto the
the upstream
upstream Fjergen
Fjergen reservoir
reservoir
3 ),
during
during high
high inflow
inflow periods.
periods.The Theupstream
upstreamreservoir
reservoir hashas a larger
a largerstorage
storage capacity
capacity (204 mil.mil.
(204 m³),mas
opposed
as opposed to to
thethe
downstream
downstream reservoir
reservoirofofjustjust4.5
4.5mil.
mil.m³. 3
m Water
. Waterfrom
fromFjergen
Fjergenisisused
usedfor for production
production
in both the
theTevla
TevlaPSPPSPand andthe
the downstream
downstream thethe Meråker
Meråker HPP. HPP.The The headrace
headrace tunnel
tunnel in Tevlain Tevla is a Y-
is a Y-tunnel,
tunnel,
with one with one connecting
branch branch connecting the powerhouse
the powerhouse to the upstream
to the upstream reservoir,reservoir,
having a having 28 m2 ,
a crossofsection
cross section
of
and28the
m2,second
and thebranch,
secondwith branch, withsection
a cross a crossofsection 2
10 m ,oftransferring
10 m2, transferring
water from water fourfrombrookfourintakes.
brook
intakes. In addition,
In addition, on Y-branch,
on the first the first Y-branch,
there is onethere is one
more brook more brook
intake. Theintake.
surgeThetanksurge tank is
is located located
along the
along the main tunnel
main headrace headrace m2 ). (28 m2).
(28tunnel
The Øljusjøen
ØljusjøenPSP PSPhas hasinitially
initially been
been builtbuilt
for for pumping
pumping the water
the water in thein the upstream
upstream reservoir reservoir
during
during flood periods,
flood periods, for use
for further further use in
in other other downstream
downstream HPPs. Thus, HPPs. Thus,
water water is pumped
is pumped from a lower fromreservoir
a lower
reservoir
(Eldrevatn)(Eldrevatn)
and 16 brook and intakes
16 brooklocated
intakesalong
located
thealong
tailrace thetunnel
tailrace tunneltosystem,
system, to the upstream
the upstream reservoir
reservoir (Øljusjøane).
(Øljusjøane). Along theAlongheadracethe tunnel
headrace tunnel
system, system,
there is justthere is just intake,
one brook one brook
whose intake,
shaftwhose
is also shaft
used
is
asalso usedtank.
a surge as a surge tank.

4. Operational
4. Operational Experience
Experience
The operators
The operators of ofthe
thePSPs
PSPshave
havebeenbeeninterviewed
interviewed onon operational
operational experience
experience as aasfollow-up
a follow-up
for
for the technical questionnaires. During the interviews, each operator
the technical questionnaires. During the interviews, each operator was asked about possible was asked about possible
operational restrictions,
operational restrictions, rock
rock falls,
falls, tunnel
tunnel collapses,
collapses, sand
sand problems,
problems, problems
problems oror restrictions
restrictions during
during
tunnel dewatering,
tunnel dewatering, andand measures
measures taken
taken to to solve
solve or
or prevent
prevent anyany encountered
encountered problem.
problem. Some
Some specific
specific
topics of lessons learned are presented in this
topics of lessons learned are presented in this section. section.
Aurland III
Aurland III PSP
PSP has
has in
in 2016–2017
2016–2017 been been upgraded
upgraded withwith new
new starting mechanism for
starting mechanism for pumping
pumping
mode [40].
mode [40]. Previously,
Previously,thisthisPSP
PSP had
had a direct
a direct start
start with
with halfhalf of the
of the nominal
nominal voltage
voltage provided
provided by
by two
two coupling windings in the transformer that resulted in a high strain
coupling windings in the transformer that resulted in a high strain of the electromechanical of the electromechanical
equipment. Now,
equipment. Now, aa frequency
frequency converter
converter is is installed,
installed, reducing
reducing thethe strain
strain on
on the
the machinery
machinery during
during
start-up. The drawback that Aurland III encounters is that in pumping mode, the units
start-up. The drawback that Aurland III encounters is that in pumping mode, the units are not started are not started
both at the same time, a company decision in order to avoid the risk of failure. A frequency converter
or soft starter investment is considered for the Duge PSP as well [40].
Three of the ten PSPs have reported tunnel collapses. The Duge PSP experienced a tunnel
collapse in the 12 km long tailrace tunnel shortly after the first water filling. The tailrace tunnel is
Energies 2020, 13, 4918 16 of 20

both at the same time, a company decision in order to avoid the risk of failure. A frequency converter
or soft starter investment is considered for the Duge PSP as well [40].
Three of the ten PSPs have reported tunnel collapses. The Duge PSP experienced a tunnel collapse
in the 12 km long tailrace tunnel shortly after the first water filling. The tailrace tunnel is mainly
unlined, and the collapse occurred in a weakness zone of which the strength was overestimated.
The Saurdal PSP experienced a tunnel collapse in the tailrace tunnel after several years of operation.
Recently, a major tunnel collapse was discovered also along the headrace tunnel of the Saurdal PSP,
which could not be cleaned due to safety reasons; thus, construction of a by-pass tunnel is currently
under investigation. Finally, the Stølsdal PSP reported a tunnel collapse in one of the brook intakes.
At Duge PSP, the main reason for the tunnel collapse was not the fact that it was a PSP, as the collapse
occurred shortly after commissioning. For the two other PSPs one may speculate that the operation as
a PSPs might have increased the stresses of the rock mass surrounding the tunnels and might have
influenced the collapse.
The Duge PSP has reported a problem of rotor lifting, where the rotor of the unit is lifted of its
bearings when operated at too high load. This situation started after a refurbishment of the units in
2017. The reason is the very high submergence of the units (more than 40 m) combined with a long
tailrace tunnel (12 km) that results in a high pressure from downstream on the turbine. The power
plant is currently operating with a restriction of maximum 80% of installed capacity. Measures to
repair the units and return to normal operation are currently undertaken.
The Duge PSP has a very long tailrace tunnel. To dewater and inspect the tunnel, over 600,000 m3
of water has to be pumped more than 50 m in vertical elevation out of the tunnel. This is very costly
and has only been done once in the power plants lifetime (after the tunnel collapse). Recently, a tunnel
inspection of the 12 km long tailrace tunnel in Duge was successfully conducted with a remotely
operate vehicle (ROV). The tunnel inspection took less than 24 h and could be conducted without
dewatering the tunnel. The ROV was used to scan and film the inside of the unlined tunnel to document
the current condition after 50 years of operation. The condition was found to be good with only minor
and insignificant rock fall.
The Duge PSP currently experiences problems with sand and debris clogging filters and seals in
the turbine during pumping mode. This problem started after a change of downstream water level
restriction for pumping mode. Previously, pumping was not allowed below 655 masl in the lower
reservoir. In the last years, this restriction has been changed to 650 masl. It is likely that during the
40-year lifetime of this power plant, sand and debris has deposited in the lower reservoir close to the
intake and is now being sucked back into the tunnel during pumping on lower reservoir levels.
The Herva PSP reported limitations due to the design of the electro-mechanical equipment.
The unit consists of separate pump and turbine runners, connected to the same motor-generator
machine; the pump runner only being connected to the unit when running in pumping mode. Due to
a time-consuming coupling procedure of the pump runner, the pump is only operated once a year,
for a few weeks during flood season. During pump operation, the turbine runner is still coupled
to the motor-generator; thus, in order to reduce the friction losses, the water is evacuated from the
turbine spiral case. Such limitations would make it impossible for the Herva PSP to be used for
primary frequency control or secondary frequency control unless the pump runner is kept connected
all the time.

5. Discussion
The Norwegian power system is almost entirely based on hydropower plants with storage
reservoirs, with very small percent of variable energy sources, resulting in a robust power system with
sufficient energy storage and frequency reserves. As a result, all ten pumped storage plants in Norway
were not designed for system services or daily peak demand, but for pumping water during flood
season, in order to store it for the high demand periods. Another reason for the PSPs to be designed
for capturing the flood water is that in Norway, due to the topography, the larger reservoirs are mainly
Energies 2020, 13, 4918 17 of 20

located on plateaus on the top of the catchments. The Stølsdal, Nygard, and Duge PSPs have the upper
reservoir twice the size of the lower reservoir. For Aurland III and Tevla PSPs, the proportion between
the two reservoirs is around 1 to 45. This demonstrates that the main purpose of the pumping in most
of the schemes is to pump inflow and not water stored in the downstream reservoir. However, most
of the schemes have sufficient size of the lower reservoir to also allow daily and weekly pumping of
stored water to profit from variations in the power prices.
The specific costs per kW is low and the specific cost per stored kWh is very low when compared
with published number for PSPs in other countries [7,10]. This can be explained by the Norwegian
topography, which enables efficient construction of reservoirs with large volumes at high elevations.
For construction of new PSPs in Norway, the costs can be expected to be even lower, as these can
be constructed between already existing hydropower reservoirs. The costs can be further reduced
by upgrading existing hydropower plants and utilizing the existing tunnel systems. The costs will
then be reduced to only the pumping units and powerhouse expansion. The total existing available
hydropower storage in Norway is currently about 85 TWh. Previous studies have identified potential
to construct over 60,000 MW of pumped storage in Norway [18].
The technical review shows that there is currently only 1369 MW installed capacity in PSPs in
Norway. All PSPs have a head between 103 m and 465 m, placing them in high head hydropower plant
category. Five out of these are large PSP, having an installed capacity above 50 MW, and the others are
medium PSPs. There is no small PSP (below 10 MW), this being because they were designed to work
as support for various industrial factories around the country, which needed a large, rather constant
power supply. When looking into the power production versus the power consumption of each PSP,
it is observed that the net power production is positive in all cases. This is consistent with the fact that
all PSPs are open loop schemes, meaning that the water used for generation comes both from pumping
and from the catchment in which the PSP is located.
The typical Norwegian PSPs has three special design characteristics, comparing with PSP in other
parts of the world. First, the tunnels are commonly constructed using the drill and blast method
and are left unlined after commissioning, this being possible due to the good rock quality in the
Scandinavian mountain range. Even if unlined tunnels have a higher tunnel roughness than lined
tunnels, which leads to an increase in the major head loss, by not lining the tunnels, the cross section
is larger, compensating for the influence of the roughness. This design is applied for all ten PSPs
presented in this paper. Second, a rock trap solution is normally located inside the tunnel, upstream
the penstock or downstream the draft tube. Since erosion in the Norwegian rivers is usually relatively
small, a sand trap at the intake upstream the tunnel is not necessary. On the other hand, leaving
the tunnels unlined could lead to rock falls which, if not trapped, can be transported to the turbine
and damage it. Another possible source of debris that needs to be trapped is pieces of eroded road.
Typically, for facilitating transport within the tunnel during construction, inspection, or maintenance,
a road is built in the tunnel, parts of which can be eroded and transported to the turbine. In order
to prevent any damage by rock falls or eroded road parts, a rock trap is placed within the tunnel.
For this reason, a rock trap is located upstream the penstock or downstream the draft tube, respectively.
The third design characteristic of Norwegian PSPs is the typical high number of brook intakes. Due to
the topography, Norwegian hydropower tunnels are long, crossing under many small secondary
streams. For this reason, along the tunnels there are several brook intakes capturing the water from
the secondary streams and bringing to the system. Along the headrace tunnel of the PSPs in Norway,
there are on average between zero and three brook intakes, except for Tevla PSP and Stølsdal PSP,
which have 6 and 10 brook intakes, respectively. Both Tevla and Stølsdal PSP have side tunnels with
the only scope of connecting brook intakes to the main headrace tunnel. Interesting is the fact that
water is brought to the system through brook intakes along the tailrace tunnel as well, in the case of
PSPs, as opposed to HPPs. It can be noticed that there is no correlation between the size of the PSP and
the number of brook intakes. No correlation between the size of the reservoirs and number of brook
intakes could be observed either.
Energies 2020, 13, 4918 18 of 20

The most common unit used in Norwegian PSPs is the reversible pump turbine, found in seven of
the plants (Aurland III, Duge, Jukla, Nygard, Saurdal, Tevla, and Øljusjøen). The start-up procedure
for pumping is varied, but a common feature of most PSPs is that the pump is started in air. All of
the PSPs were designed for seasonal pumping during high flow periods during spring and autumn.
The majority of the PSPs therefore have time-consuming starting mechanisms to reduce costs and
electromechanical strain on the units. The operation of the pumps is still primarily for seasonal storage
during periods with high inflow. However, more frequent start–stops and operation also during low
flow periods can be observed. It is therefore becoming more attractive for the PSP owners to upgrade
the starting mechanism for enabling more frequent and rapid start–stop operations.

6. Conclusions
There currently exist 1369 MW installed capacity with an energy storage capacity of about 5 TWh
in the ten existing PSPs in Norway. The construction costs have been calculated with current prices
based on a national cost base with price statistics for hydropower in Norway. The results show that
the specific costs per kW is low and the specific cost per storage kWh is very low compared with
published numbers from PSPs in other countries. There is a large potential for construction of new
PSPs in Norway. The costs of new PSPs can be even lower as they can be constructed between existing
reservoirs and by upgrading already existing hydropower plants into PSPs.
The round-trip efficiencies range from 65% to 80% for the ten PSPs. Most of the PSPs in Norway
have long tunnel systems, which is one of the main causes of energy loss. The round-trip energy
loss from the tunnel system ranges from less than 1% (Nygard) to 15% (Duge). The round-trip
electromechanical losses including the transformer range from 20% to 25%. It is noted that the
electromechanical efficiencies may be improved by upgrading the units, while the tunnel system head
loss is usually not feasible to reduce.
All the Norwegian PSPs are open loop, with significant natural inflow to the reservoirs. This results
in operation primarily as a normal hydropower plant with shorter periods of pumping. The PSPs have
a significantly higher energy production compared with energy consumption. The operation of the
Norwegian PSPs is still mainly for seasonal pumping, but a shift towards more frequent start–stop
operation can be observed. Pumping during nighttime and no operation during daytime is becoming
more frequent.
In general, the Norwegian PSPs have time-consuming pump starting mechanisms, being designed
for seasonal storage. However, it is possible to upgrade the starting mechanism as it was done in
Aurland III PSP in 2016–2017. Such upgrading is becoming more attractive when the spread between
high and low power prices is increasing, and system services are priced higher.
Tunnel collapses have occurred in three out of ten PSPs. Two of them occurred after several
years of operation and may be related to the additional strain of pumping operation compared to
normal hydropower plants. The variation of the pressure on the rock mass around the tunnel is
significantly higher in PSPs as opposed to HPPs; thus, this needs to be accounted for during design of
tunnel systems.
The hydraulic system of each PSP in Norway is unique and with many interesting features. Several
technical solutions can still be regarded as innovative and the operational experience is worthwhile to
share with the research community.

Author Contributions: Conceptualization, L.P., K.V., and L.L.; data collection, L.P. and K.V.; data analysis, L.P.;
writing—original draft preparation, L.P. and K.V.; writing—review and editing, K.V. and L.L.; supervision, K.V.
and L.L.; project administration, L.L.; and funding acquisition, K.V. and L.L. All authors have read and agreed to
the published version of the manuscript.
Funding: This research is financed by and conducted as a part of the Norwegian Research Center for Hydropower
Technology (HydroCen).
Acknowledgments: The data is provided by the power plant owner and operators: Statkraft Energi, Sira-Kvina
kraftselskap, E-CO Energi, Nord-Trøndelag Elektrisitetsverk, BKK Produksjon, Hydro Energi, Østfold Energi
Energies 2020, 13, 4918 19 of 20

Conflicts of Interest: The authors declare no conflict of interest.

References
1. Directive (EU) 2018/2001 of the European Parliament and of the council of 11 December 2018 on the promotion
of the use of energy from renewable sources. Off. J. Eur. Union 2018, OJ L 328, 1–128.
2. Schaber, C.; Mazza, P.; Hammerschlag, R. Utility-scale storage of renewable energy. Electr. J. 2004, 17, 21–29.
[CrossRef]
3. Moseley, P.T.; Garche, J.; Elsevier, B. Electrochemical Energy Storage for Renewable Sources and Grid Balancing;
Elsevier B.V: Amsterdam, The Netherlands, 2015.
4. Abdin, Z.; Khalilpour, K. Chapter 4-Single and polystorage technologies for renewable-based hybrid energy
systems. In Polygeneration with Polystorage for Chemical and Energy Hubs; Academic Press: Cambridge, MA,
USA, 2018; pp. 77–131.
5. Steilen, M.; Jorissen, L. Chapter 10-hydrogen conversion into electricity and thermal energy by fuel cells:
Use of H2-systems and batteries. In Electrochemical Energy Storage for Renewable Sources and Grid Balancing;
Elsevier: Amsterdam, The Netherlands, 2015.
6. Kumar, A.; Schei, T.; Ahenkorah, A.; Caceres Rodriguez, R.; Devernay, J.M.; Freitas, M.; Hall, D.; Killingtveit, Å.;
Liu, Z. Hydropower. In IPCC Special Report on Renewable Energy Sources and Climate Change; Cambridge
University Press: Cambridge, MA, USA, 2011; pp. 437–496.
7. Deane, J.; Gallachóir, B.O.; McKeogh, E. Technico-economic review of existing and new pumped hydro
energy storage plant. Renew. Sustain. Energy Rev. 2010, 14, 1293–1302. [CrossRef]
8. Hunt, J.D.; Byers, E.; Wada, Y.; Parkinson, S.; Gernaat, D.E.H.J.; Langan, S.; van Vuuren, D.P.; Riahi, K. Global
resource potential of seasonal pumped hydropower storage for energy and water storage. Nat. Commun.
2020, 11, 947. [CrossRef] [PubMed]
9. Rogner, M.; Troja, N. The World’s Water Battery: Pumped Hydropower Storage And Clean Energy Transition; IHA:
London, UK, 2018.
10. Pikl, F.G.; Richter, W.; Zenz, G. Large-scale, economic and efficient underground eneegy storage.
Geomech. Tunn. 2019, 12, 251–269. [CrossRef]
11. Pumped Energy Storage: Vital to California’s Renewable Energy Future; San Diego County Water Authority:
San Diego, CA, USA, 2019.
12. Schmidt, O.; Melchior, S.; Hawkes, A.; Staffell, I. Projecting the future levelized cost of electricity storage
technologies. Joule 2019, 3, 81–100. [CrossRef]
13. ICOLD Committee on Dams for Hydroelectric Energy. Bulletin 163: Dams for Hydroelectric Energy; International
Commission on Large Dams (ICOLD): Paris, France, 2017.
14. Hydro in Europe: Powering Renewables; Eurelectric: Brussels, Belgium, 2011.
15. Pumped Storage Hydropower in Austria; Technischen Universität Graz: Graz, Austria, 2018.
16. Compendium of Pumped Storage Plants in the United States; American Society of Civil Engineers: New York, NY,
USA, 1993.
17. Lia, L.; Vereide, K.; Strypet, L.F.; Kvaal, B. The new strategy for PSP in Norway-medium sized projects in
existing power schemes. In Proceedings of the Hydro 2016 Conference, Montreux, Switzerland, 10–12 October
2016; Aqua-Media International Ltd.: Sutton, UK, 2016.
18. Charmasson, J.; Belsnes, M.; Andersen, O.; Eloranta, A.; Graabak, I.; Korpås, M.; Helland, I.P.; Sundt, H.;
Wolfgang, O. Roadmap for Large-Scale Balancing and Energy Storage from Norwegian Hydropower. Opportunities,
Challenges and Needs Until 2050; CEDREN: Trondheim, Norway, 2017.
19. Pumpekraft i Norge; The Norwegian Water Resources and Energy Directorate (NVE): Oslo, Norway, 2011.
20. Eurostat. European Office of Statistics. Available online: https://ec.europa.eu/eurostat/ (accessed on
15 July 2020).
21. Scholz, Y.; Gils, H.C.; Pietzcker, R. Application of a high-detail energy system model to derive power sector
characteristics at high wind and solar shares. Energy Econ. 2017, 64, 568–582. [CrossRef]
22. International Hydropower Association. Hydropower Status Report. Sector trends and Insights; IHA: London,
UK, 2020.
Energies 2020, 13, 4918 20 of 20

23. Cebulla, F.; Naegler, T.; Pohl, M. Electrical energy storage in highly renewable European energy systems:
Capacity requirements, spatial distribution, and storage dispatch. J. Energy Storage 2017, 14, 211–223.
[CrossRef]
24. Droste-Franke, B.; Carrier, M.; Kaiser, M.; Schreurs, M.; Weber, C.; Ziesemer, T. Improving Energy Decisions
Towards Better Scientific Policy Advice for a Safe and Secure Future Energy System; Springer: Berlin, Germany, 2015.
25. Heide, D.; Greuner, M.; von Bremen, L.; Hoffmann, C. Reduced storage and balancing needs in a fully
renewable European power system with excess wind and solar power generation. Renew. Energy 2011,
36, 2515–2523. [CrossRef]
26. The Norwegian Water Resources and Energy Directorate (NVE). Norway and the European Power Market.
Available online: https://www.nve.no/norwegian-energy-regulatory-authority/wholesale-market/norway-
and-the-european-power-market/ (accessed on 15 July 2020).
27. Solvang, E.; Charmasson, J.; Sauterleute, J.; Harby, A.; Killingtveit, Å.; Egeland, H.; Andersen, O.; Ruud, A.;
Aas, Ø. Norwegian Hydropower for Large-Scale Electricity Balancing Needs; SINTEF Energy Research: Trondheim,
Norway, 2014.
28. Harreiter, H. Empowering Austria: The electricity strategy of Austrian electricity companies. Wasserwirtschaft
2017, 107, 23–27. [CrossRef]
29. Müller, A.; Münch-Alligné, C.; Nicolet, C.; Denis, V.; Avellan, F. Pushing the envelope: Switzerland’s
approach to unlocking hidden hydropower potential. In Hydro; Hydropower and Dams: Porto, Portugal,
2019.
30. Byman, K. Future Electricity Production in Sweden; The Royal Swedish Academy of Engineering Sciences
(IVA): Stockholm, Sweden, 2016.
31. Hydro. Illvatn pumpekraftverk Luster kommune. Konsesjonssøknad med konsekvensutredning; Norges Vassdrags-og
Energidirektorat: Oslo, Norway, 2010.
32. Bye, T.; Hope, E. Deregulation of electricity markets-The Norwegian experience. Econ. Pol. 2005, 40, 5269–5278.
33. Kostnadsgrunnlag for Vannkraft (Cost Base for Hydropower Plants). Norwegian Water Resources and Energy
Directorate; The Norwegian Water Resources and Energy Directorate (NVE): Oslo, Norway, 2016.
34. Thoma, D. Zur Theorie des Wasserschlosses bei Selbsttätig geregelten Turbinenanlagen; Walter De Gruyter:
Munchen, Germany, 1910.
35. Anderson, A. Surge shaft stability with pumped-storage schemes. J. Hydraul. Eng. 1984, 110, 687–706.
[CrossRef]
36. Goodall, D.; Kjørholt, H.; Tekle, T.; Broch, E. Air cushion surge chambers for underground power plants.
Int. Water Power Dam Constr. 1988, 11, 29–34.
37. Vereide, K. Hydraulics and Thermodynamics of Closed Surge Tank for Hydropower Plants; Norwegian University
of Science and Technology (NTNU): Trondheim, Norway, 2016.
38. Network Code on Load-Frequency Control and Reserves; ENTSO-E: Brussels, Belgium, 2013.
39. Vereide, K.; Tekle, T.; Nilsen, T.K. Thermodynamic behavior and heat transfer in closed surge tanks for
hydropower plants. J. Hydraul. Eng. 2015, 141, 06015002. [CrossRef]
40. Breiland, S. Reconstruction of the Pumped Hydropower Plant, Duge; University of South-Eastern Norway:
Notodden, Norway, 2019.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).