Supplementary letter from National Grid
Further to your request for additional information
about the costs of managing intermittency with 40% of electricity
generated by renewable sources, I hope the following is helpful.
1. THE DEFINITION
OF THE
COSTS OF
VARIABILITY/INTERMITTENCY
There is scope for making different definitions
of the additional costs that result from the variability/intermittency
of certain renewable sources and, without care, this can lead
to double counting of some costs. The definitions used in my oral
evidence on 28 April and in this note are consistent with the
methodology published in an earlier paper[1]
which compared a scenario (A) in which electricity is produced
solely with conventional generation to a scenario (B) in which
a proportion of electricity is produced from wind power.
The key cost components were defined as follows:
| |
Scenario A: Conventional Only |
Scenario B: Conventional & Wind |
| 1. Generation capacity financing costs
| Conventional capacity (including security margin at peak)
| Conventional capacity (slightly reduced from scenario A due to wind capacity contribution but remaining sufficient for wind backup role) Wind capacity (turbine capital cost)
|
| 2. Generation fuel costs (also reflecting carbon emissions)
| To meet demand | To meet demand less wind energy contribution (potentially 40% of fuel)
|
| 3. Balancing costs (in which significant changes expected)
| Short-term reserve capacity & utilisation for diversified mix of: (i) demand forecast errors (ii) conventional generation breakdowns/unavailability Congestion/constraint costs (depending on network capacity constructed)
| Short-term reserve capacity & utilisation for diversified mix of: (i) demand forecast errors (ii) conventional generation breakdowns/unavailability (iii) wind variability Congestion/constraint costs (depending on network capacity constructed)
|
| 4. Network costs | Connections for conventional generation Infrastructure for conventional generation
| Connections for conventional generation Connections for wind generation Infrastructure for conventional & wind generation (some shared)
|
In the GB market, generation capacity (item 1) and fuel costs
(item 2) are funded through the wholesale electricity price. Balancing
(item 3) and network costs (item 4) are funded from National Grid's
use of system charges, some of which are passed to generators
(and so are also funded by the wholesale electricity price) with
the remainder passed to suppliers.
The earlier paper (which contains prices that need updating
to reflect recently observed inflation in fuel and equipment costs)
identified generation capacity costs (item 1) and fuel (item 2)
as the most material cost elements so that the overall economic
impact of accommodating wind in the electricity system is dominated
by the extent that fuel and associated carbon emission savings
offset the cost of the wind turbines. However, the impact of wind
on the need for conventional "backup" capacity, network
extension and balancing do result in additional costs which need
to be considered for a complete assessment.
In my oral evidence and subsequent detail provided below,
the estimates of additional balancing costs refer to item 3 only.
As noted below, however, the component arising from network constraints
will depend on the extent that additional network infrastructure
is established (item 4). At large wind penetration levels, wind
variability will be the dominant driver of short-term reserve
requirements.
2. Balancing Cost Estimates
Assuming that the electricity market continues to maintain
sufficient generation capacity to meet peak demand securely (including
sufficient backup capacity for low wind days), we estimate the
additional short-term balancing costs arising with wind providing
circa 40% of electricity in 2020 to lie in the range £500
million to £1,000 million per annum. (For reference, current
total balancing costs are circa £530 million per annum so
that these additional requirements represent a doubling or tripling
of current balancing costs).
On the basis that wind would be supplying approximately 140TWh
of the assumed annual electricity consumption in that year, these
balancing costs represent between £3.5 and £7 for each
MWh of wind produced. The cheaper end of this range represents
a business as usual scenario with reserves and balancing performed
as today and with the market prices of the various balancing services
remaining constant (despite the larger volumes required). The
higher end of the cost range includes network constraint/congestion
costs which might arise if there are delays in establishing network
capacity or if there is significant network capacity sharing.
As my colleague Nicola Pitts highlighted in her evidence, there
are opportunities for improved demand side response (perhaps better
facilitated by Smart meters) which could reduce such costs somewhat
from these business-as-usual estimates.
Given that these costs are incurred on 40% of the energy
produced, the cost per unit consumed would be in the range £1.4
and £2.8 per MWh consumed or 0.14p/kWh and 0.28p/kWh. This
would be a small part of a consumers electricity bill (circa 1.6%
to 3.2%). If we take an average domestic consumer electricity
bill to be £390 per year (source: EnergyWatch website), the
balancing cost for wind would represent an additional £6
to £12 per annum.
19 May 2008
1
"Total cost estimates for large-scale wind scenarios
in the UK", Dale, Milborrow, Slark & Strbac, Energy Policy
32 (2004) 1949-1956. Back
| 
