Network integration of wind power - Wind Power Integration: Connection and System Operational Aspects - page 112

The first relates to the island of Ireland. The plan here is to achieve about 40%
of energy by consumption from renewables (mostly wind). A crude estimate is that
the load factor of demand on the island is 60% and the load factor of wind gen-
eration is about 30%. To achieve the consumption target of 40% of energy from
renewables, it follows that a capacity of about twice 40% of wind generators must
be installed. (The average power consumed is 60% of maximum demand. A wind
energy penetration of 40% implies average wind generation of 40% of average
demand, or 0.4 0.6 (24%) of maximum demand. If the wind sector has a gen-
eration load factor of 30%, wind capacity needs to be (0.4 0.6)/0.3 or 80% of
maximum demand.) The situation worsens as the generation load factor decreases.
It may be that as penetration increases there is a need to use less windy sites
because hill tops are either already used or protected for visual reasons. If the
renewables mix includes a significant proportion of smaller wind turbines, they will
be closer to ground level and subject to much increased surface roughness effects
which reduce the wind incident intensity and therefore the generation load factor.
However, Ireland's load ranges from about 2,000 to 8,000 MW over the year.
So if 80% of 8,000 MW is needed as installed renewable capacity as estimated
above (6,400 MW), then that capacity cannot be used to supply load for a sig-
nificant part of the time. Furthermore, the above calculation assumed the generators
could be accepted whenever there is sufficient wind. This is clearly not the case.
Added to this is that to maintain system inertia, fault level and voltage stability on
the island, something like 50% of the instantaneous demand must be met by tra-
ditional generating plant with inertia and the ability to provide reactive power
during faults.
The overall effect is that the estimate of high level use of the transmission
system needs to be revised downwards as the percentage penetration grows. The
assessment also needs to factor-in the probability of transfers to and from the island
on interconnection with Britain. Both links with GB are HVDC cables and there-
fore have no inertia contribution (much like DFIG wind farms). To be realistic, this
level of study needs to be done using dispatch models, but the sales opportunities
for the wholesale energy in Britain and therefore the prices are almost impossible to
gauge in the lengthy timeframe needed to develop transmission circuits and sub-
stations. It is therefore something of a leap of faith that the high penetration can be
managed, and as a result acceptance of network development has been slower than
is necessary if the target is to be met.
The second example refers to countries with a high penetration of nuclear
power (which is low carbon plant but with limited operational flexibility). The
ownership and market arrangements are important. Supposing that the nuclear
power is retained in the ownership of the utility and treated as base load or must-run
plant, then other generation must be reduced when the wind is strong. However, if
market rules require that the nationalised generation is used for balancing and the
use of privatised generation for balancing is an action of last resort, the choice is
mostly to attempt to manage nuclear generation to balance wind as an action
'within the market day'. Operators then may have to decide whether to reduce
nuclear or curtail wind. It is worth reminding readers that this discussion is relevant