Memorandum by Sir Donald Miller
INTRODUCTION
1. Government, in setting out their policy
on renewable energy, have on several occasions stated that the
targets are "subject to the costs being acceptable to consumers";
White Paper Cm 5671 P16 Section 1.22 lines 7-8 is one such example.
However no steps have been taken to assess what these are likely
to be and to consult consumers on what they consider acceptable.
This paper concentrates on the costs of wind energy and particularly
the costs to consumers, wind being the technology for which the
costs are most clearly established. It first examines the various
areas where costs are incurred before attempting to quantify each
to arrive at a total figure. It then adds some more general comment,
based on established engineering criteria and long experience
of power generation, on the likely relative costs of other renewable
sources such as marine and dispersed local generation.
WIND ENERGY
2. It is frequently claimed that the output
from wind developments will supply so many houses. In reality,
as Fig 1 published by National Grid shows, the output is so intermittent
and unpredictable that it cannot be relied on as a supply. An
electricity system cannot store energy in significant quantity
so that its operation requires there be an exact balance at all
times between the demand and generation output. Thus any source
of generation which cannot be relied on to produce power on demand
requires to be backed up by conventional fossil fuelled generation.
The burdens of this are met entirely by consumers and comprise
(a) the capital costs of providing this spare generating plant
but also (b) the increased operating costs required to ensure
that the system can compensate for the short term and unpredictable
swings in output from wind farms.
3. Dealing first with (a), the capital costs
of support plant, it is sometimes claimed that wide dispersement
of installations throughout the country will greatly mitigate
the requirement for such back-up but experience does not support
this. An examination of Danish performance shows that on 29 December
2000, with 1,860MW of wind turbine capacity installed and an output
of 1,715MW, this was reduced to 951MW over one hour under storm
conditions. Since the period of study is necessarily short, this
is unlikely to correspond to the worst conditions, but nevertheless
it represents a power swing of 41% of wind output per hour.
4. Similarly, in Germany the grid operator
E.ON* in their report Wind Year Overview 2003 describe
a reduction from 4,300 MW to 660MW (a drop of 3,640MW) over six
hours with a peak rate of about double that or 30% of output per
hour.
5. National Grid in their evidence to a
House of Lords Select committee gave as their criteria for when
the intermittency of wind power would incur additional costs for
the UK system as:
"When there is a potential loss over one
hour greater than 3% of peak demand. 3 % of the UK peak demand
of 60,000 MW amounts to 1,800 MW and the system must be capable
of withstanding this loss at all times including the hours when
wind output is falling rapidly The UK grid system is designed
and operated to cater for the instantaneous loss of a major generating
station or a double circuit high voltage line and thus has already
to accommodate a maximum power loss of 1,250MW. If we take the
maximum short term wind power reduction at say 35% of wind output
we can readily calculate the capacity of installed wind generation
beyond which, according to National Grid, significant additional
operating cost will be incurred on the system. This calculation
below shows that this occurs once the installed wind turbine capacity
exceeds:
6. Already there some 2,500 MW of wind turbines
in the UK in operation and some 19,000 MW including those in the
planning process. In Scotland alone, see Fig 1, some 2,000MW are
in operation or under construction and as much as 8,000MW, including
those for which applications have been made for planning consent,
so that we are already in line for very significant system support
costs. Discussing costs, the Royal Academy of Engineering, comment
in their Annual Review 2002-03 on the Government's Energy target
of generating 20% of our electricity by 2020 from renewables say
"this is over-optimistic and fails to address the fundamental
difficulty of all renewable sources—they are intermittent."
*E.On, owner of UK generator Powergen, is a
major player in the UK wind market.
7. An analysis by the Grid arm of the large
German utility E.On has demonstrated that this additional back
up thermal generating plant needs to be 80% of the installed wind
turbine capacity, a figure which is supported by the Irish Grid,
The Royal Academy of Engineers, by OFGEM and more recent work
in a BERR/SEDC study. Indeed other work based on probability theory
both here and in Germany suggests that if the reliability of our
supplies is to be maintained at traditional levels something like
90% back up will be required for the high levels of wind power
envisaged.
8. Turning to (b), the need for operational
support, since output from wind generators can vary dramatically
over a short time any reduction must be compensated for by quick
response conventional generating plant. However fossil fuelled
plant, such as a large coal fired generation, takes typically
10 hours to start and load up from cold so that it becomes necessary
to run sufficient back up plant at less than rated load whenever
the wind turbines are in operation. The minimum load at which
a 600MW coal fired generator can be operated is typically some
200MW or one third of rated full load so that to match the wind
turbine output at any time on the system a large number of fossil
fuelled units will require to part loaded with the loss of efficiency
this entails. Experience suggests that the total amount of this
de-loading on the system should be some 60% of the wind turbine
output at any time and that this will need to be done on a regional
basis, if large swings in the power transfers are to be avoided.
Operation in this inefficient mode incurs further additional costs
which are met by the consumer. It also increases the quantities
of CO2 released compared with operation at full load, so reducing
the gain in CO2 emissions by at least 20% compared with that given
by a simple calculation based on the wind energy output.
9. Nor does this represent the whole of
the financial costs incurred as a result of an excessive reliance
on wind energy. There is also the cost of transmitting the energy
to the consumer. It is clear that based on present planning the
vast majority of wind installations, will be mainly in Scotland
and to a lesser extent in Wales and the West of England, and all
remote from where the demand is required in the South and South-East
of England. It may therefore be instructive to look in more detail
at the situation in Scotland. National Grid are planning on an
installed wind farm capacity of 8,000MW in Scotland by year 2013-14
which compares with existing conventional generating capacity
of 8,900MW and a maximum system demand of some 5,700 MW. The present
capacity of the transmission system from Scotland to the north
of England is some 2,200MW so that there is at present no outlet
for additional capacity either in Scotland or over the interconnectors
to England without displacing existing Scottish generating capacity.
Recognising this limitation, Government and Ofgem (the electricity
regulator) have authorised the Scottish Grid Companies to spend
an initial £800 million on reinforcing the HV transmission
system to the North of England to cater for exports from wind
farms in Scotland. £350 million of this initial expenditure
is for the construction of a new EHV line along the major tourist
route of the A9 in the Highlands from Beauly, north of Inverness,
to Glasgow with the remainder for reinforcing the cross Border
links to the North of England. This will increase the capacity
of the transmission to England to some 3,300MW, sufficient only
to cater for the expected wind capacity for this year (2008).
The extra cost this year of this transmission expenditure, which
would not be required but for excessive installations of wind
energy is equivalent to £35/MWhr of energy produced and alone
is comparable with the price of bulk power from conventional generation
10. Even the heavy expenditure on the transmission
system referred to above will not be sufficient to cater for the
large amounts of wind power projected for installation in Scotland.
The Authoritative source for information on the UK Grid System
is the 1,000 page statutory Seven Year Statement published
annually by the three transmission companies, National Grid, Scottish
Power and Scottish and Southern. This most recent Statement (2007)
contains an exhaustive analysis of the effect of increasing wind
energy in Scotland on the required power flows to England and
is depicted in chart form in Fig 3. The National Grid figures
show an increase from wind power of 3,625MW (from 8,827 to 12,493)
over the seven years to 2013 with export flows (the red line in
Fig 3) increasing by 3,094MW (from 2,653 to 5,746 MW). This is
all from proposed increases in wind installations in Scotland.
It should be noted that these increases in wind output are not
balanced by reductions in conventional generating plant, even
supposing this were technically possible on the system. Indeed
heavy expenditure is now being undertaken to fit flue gas desulphurisation
at Longannet and Cockenzie power stations to extend their working
lives beyond 2015 and further licences can be expected to extend
the lives of Scotland's nuclear plants. National Grid's analysis
also shows that these same levels of import are required to maintain
supplies in England. This situation is illustrated pictorially
in Fig 4 where by 2013 over 5,000MW is shown being exported from
Scotland and transported throughout England to the SE, even affecting
flows to the continent over the sub sea cables.
11. It should be noted that the flows described
above are those used by the Grid companies for system planning
and investment purposes and are for conditions at maximum demand
and with wind power outputs set at 60% of installed capacity.
Thus the export of 5,746 MW in 2013-14 means that the whole of
the wind generation in Scotland (8,000 MW installed capacity)
will be exported even at periods of high system load on the Scottish
system. At lower loads, because of the relatively lower operating
costs of other generation in Scotland (nuclear and high merit
coal) compared with England, the high levels of wind power outputs
exported to England will be maintained or even increased.
12. The costs of all this transmission for
these massive wind power developments proposed for Scotland could
not be better illustrated than by the Commentary of National Grid
"Upon completion of these planned reinforcement programmes,
the Scottish-England boundary continues to show insufficient transfer
capacity indicating further reinforcement may be required"
and proposing substantial further expenditure in Central Scotland
as well as two additional double circuit high power 400KV Transmission
lines from Kilmarnock in the west and Eccles in the east across
the border as well as substantial reinforcement of the transmission
system all the way to the SE of England-the costs of which will
all be met by the electricity consumer.
13. It is worth noting in passing that without
any wind farm installations Scotland already generates some 60%
of its electricity from non CO2 emitting sources (nuclear and
water) one of the highest world wide after France with its 80%
nuclear and Norway with significant water power resources.
14. Having established that there would
be very heavy expenditure on the transmission system it is next
necessary to examine where these costs fall. While it is true
that all costs are eventually met by electricity consumers some
paths are more direct than others. For transmission costs, the
charging system is operated by National grid on behalf of the
three transmission owning companies. In the case of "Use
of System Charges", only a proportion (currently 27%) of
the cost of the transmission capacity that their installations
require is met by the generators with 73% being met by electricity
consumers by payments to the Distributors through their tariffs.
In addition generators of less than 100MW capacity currently qualify
for a "Small Generator Rebate" amounting to some £4.5/MWHR.
Tariffs published by National Grid quote a cost per annum to genertaors
of £ 13.5/KW of generation capacity connected in the South
of Scotland (increasing to over £ 22/KW in the North) these
representing 27% for their present estimate of costs. While these
costs can be expected to find their way indirectly through to
consumers the remaining 73% ie £36.5/KW for generation connected
in central Scotland, is charged to consumers through their tariffs
as a recoverable cost by the Grid Operators from the Distributors.
15. It is accepted that even though the
developer is relieved of some of the above charges the costs of
wind power would make it uncompetitive in the UK market. The Royal
Society of Engineering, The UK's premier Engineering Institution
published an assessment of generation costs in 2005 and a further
updated study has more recently been published by PB Power one
of the UK's leading Firms of Consulting Engineers Firms. The latter
shows (Fig 5) onshore wind, (even when virtually ignoring system
support and transmission costs) at £55/MWhr against a market
price of some £40/MWhr for bulk power in recent years and
an even lower cost for new nuclear. In order to make the construction
of wind farms attractive to developers the Government has decreed
that wind power developers are paid a subsidy (Renewables Obligation
Certificate) currently £35/Mwhr escalated with RPI as well
as being relieved of the Carbon Tax Levy of £ 1.4/MWhr. These
costs are also paid by consumers through their electricity bills.
However because the buy out payments by Distributors who are short
of the required ROC certificates are redistributed to the ROC
holders, the prices at auction are considerably higher than the
nominal ROC values.
16. The implications of these various factors
for the costs of electricity from wind power located in Central
and in Northern Scotland to the Consumer are therefore:
|
| Item |
| Cost £/MWhr
|
| | Central
| Northern |
|
| Auction price | |
| |
| (Energy +ROC +CCL) |
| 93.3 (average of last 4 auctions)
|
| Capital charges for back-up plant (80%) |
"" |
| |
| ""
| 22.6 (as R Academy of Engineering)
|
| Additional running costs for part loaded plant.
| "" |
| |
| Transmission charges (73% not met by developer)
| | 14.9 |
24.2 |
| | £130.8/MWhr
| 140.1/MWhr |
|
17. These figures of £130/MWhr and £140/MWhr
are some three to four times the cost of bulk power in the electricity
market (Fig 5) or power from a new nuclear station. The direct
subsidy to the developer alone in the auction price at some £
50/MWhr is more than the market value of the electricity and this
is before adding the additional costs incurred by the grid operator
or the additional transmission charges and system losses which
are a direct charge on the consumer.
18. It should be noted that for new generating plant,
whether replacement or additional, located in the South of England
where it is needed, capital investment in transmission would be
minimal. There would be no requirement for Ofgem's recently authorised
£800 million to cater for wind energy in Scotland, or the
further heavy expenditure which is foreseen by National Grid to
transmit this energy all the way to SE England.
19. To put this matter in perspective, the total increase
in annual costs of 8000MW of wind energy in Scotland by 2013-14
would reach the startling figure of £2,000 million in that
year alone. By 2020, if the cost of the Government target of 40%
wind in Scotland is met, the cost of the ROC subsidy alone, all
met by the consumer, would aggregate to the startling figure of
some £4,000 millions with the figures for the UK some £
30,000millions. (See the estimate of aggregate ROC subsidies by
2020 amounting to given in the House of Lords by the Parliamentary
Secretary to the DTI on 5 May 2004 and the statement by the National
Audit Office that subsidies would be running at £1,000 millions/annum
by 2010).
20. The continuance of these subsidies is at variance
with claims that the present support is to be regarding as launch
costs and that wind turbines will become more economic with time.
Engineers know that unlike conventional methods of generating
electricity, the weight and thus the cost of wind turbines goes
up faster with size than does the output. So the potential for
cost reductions is very limited indeed.
21. Even if it is accepted that some premium in the cost
of energy from renewable sources is justified because of reductions
in CO2 emissions these levels of cost go far beyond any foreseeable
level of carbon pricing. Surely too, if we are to accept such
financial penalties it should only be against the assurance of
a fully effective policy but this does not appear to be the case.
At present there are some 19,000 MW (compared with some 25,000
MW of renewables to meet the 2020 target) wind power installations
in prospect or under investigation in the UK alone. If built there
would be practically no where on high ground that one would not
be dominated by wind installations. As to effectiveness, a simple
calculation shows that we would need to install some 10 times
this capacity each year even to compensate for the additional
CO2 emissions from new coal fired power stations being commissioned
in China.
22. The inevitable conclusion is that this massive concentration
on wind must be about the worst energy investment yet devised
either for its ineffectiveness or its cost/benefits for consumers.
It is surely time that consumers were made aware of the massive
subsidies being awarded to developers and the costs being incurred
in his name and the views of consumers taken into account. It
raises the question as to what level of costs is required to trigger
the Government criterion of "provided the costs are acceptable
to the consumer" but I submit that if these costs were more
widely known they are far beyond what would be acceptable to the
majority of consumers, particularly as with rising energy costs,
fuel poverty is a growing issue for many households and when concern
is being expressed about the competitiveness of British Industry
OTHER RENEWABLE
ENERGY SOURCES
AND DISPERSED
GENERATION
23.It is too easy when looking at the power of the waves
to believe that here is a massive energy source which could be
tapped for commercial needs and recent official statements have
tended to reflect this view. Objective professional engineering
opinion, as typified by the 2003 Report of the Royal Academy of
Engineers and the recent report of PB Engineers (Fig 5) takes
a markedly less optimistic position and one which objective engineering
assessment would support.
24. The fundamental difficulty with many, if not all,
all renewable energy sources is that the energy density is low
and therefore the costs of development are inherently high; it
is not for nothing that high energy density, coal, oil, gas, nuclear
and high head water power have been the favoured sources for energy
generation and will continue to provide the lowest cost power.
The energy density in a moving fluid (wind or water) is a function
of the density of the fluid times the cube of the velocity. But
the velocity of a fast moving tide is some 6 knots at best over
a few hours at springs (much lower for most of the time) compared
with some 25 knots for a brisk wind so that even with the greater
density of water the energy density on average is not that much
different from wind. After taking into account the much larger
forces acting on a marine turbine and the harsh environment of
a marine installation, the prospects are far from promising. Nor
it is likely that any new technique will overturn this conclusion;
we have had screw propellers and paddles, the basis of the present
marine tidal prototypes for some two hundred years during which
there has been no shortage of development and no-one is holding
out the prospect of a better means of transferring energy between
an electrical generator shaft and a marine current.
25. Most engineering opinion would put the prospects
for wave power development even lower than for tidal, not least
because of the uncertain forces to which such devices are subjected
and the logistic problems of anchoring and servicing any substantial
installation. With installations on any scale there must also
be a high risk of danger to navigation. These disadvantages seem
to be recognised in the recent decisions of Government to determine
a buy out price (admittedly at the development stage) of £177/MWhr
for Tidal and no less than £196/MWhr for Wave.
26. Turning to micro level generation for application
on a domestic scale a fundamental difficulty is that of the economics
of scale. It is a good rule of thumb for most plant such as that
for generating electricity that a doubling of the unit size allows
a reduction in the capital cost by one third per kW of output.
Thus in moving from a 600MW generator to multiple units of say
6kw there will a massive capital cost penalty and also a not insignificant
reduction in thermal efficiency of generation?
This can of course be compensated for to a degree if the
exhaust heat which would otherwise go to waste can be used for
space heating but this requires a good match between the electricity
and the heat demands at all times throughout the 24 hours and
the year. At best this could only be achievable and that on a
short term basis by use of an associated energy storage system.
27. It is frequently asked why large central heating
systems on a town scale have been so widely employed in say Sweden
and Finland and not in this country. There is of course the obvious
difference that they have longer and colder winters than Britain
with its maritime climate but detailed investigations of this
in the past have shown that the district heating is heavily subsidised
by the electricity sales to the detriment of the latter.
28. More recently we have seen increasing demands for
a "buy back" tariff for electricity sold to the grid
from small installations to match the buying price. It should
be recognised that this is simply a call for another form of subsidy
since the value of the output from any generator which cannot
guarantee to provide output over periods of maximum demand is
nothing more than the value of the fuel displaced. Nor of course
does the small generator bear any of the costs of providing the
electricity system, the expenses of which have to be recovered
within the prices for energy sales to consumers. To the extent
therefore that any such tariff exceeds the fuel price this would
simply amount to an additional tax on the generality of electricity
consumers
CONCLUSION
29. The question has to be asked whether the present
concentration on renewables is an appropriate response to the
possibility of global warming being down to anthropomorphic CO2.
The science of this seems far from being firmly established, and
until it is, any measures to lessen its possible impact should
I suggest be regarded as more in the nature of an insurance policy
with premiums rated accordingly. To accept a charge of as much
as three times the current price of bulk power would seem altogether
excessive in such circumstances especially when we have to hand
alternative means of electricity generation which are fully proven,
vastly more economic and capable of being introduced on a much
shorter time scale.
3 June 2008
| 
