Environmental Engineering Reference
In-Depth Information
On the other hand, there are also times of low uncertainty and high wind power gen-
eration. In these times, the traders of wind energy have to do something extraordinary to
get rid of the energy. Although it may still be possible to bring the energy into the market
by using dumping prices, it may not cover the marginal costs anymore. However, once
a certain percentage of the total consumption is exceeded, other methods will have to be
applied. These include energy pooling, export, methods to increase the demand and other
trading strategies.
4.
Optimisation Schemes
In this section we describe a number of optimisation schemes that are either already in
operation, or are likely to become part of the near future's energy trading at the markets.
4.1.
Pooling of Energy
The optimisation problem when looking at pools of energy is relatively complex. There-
fore, an example will be used in the following to illustrate the optimisation targets and
possibilities of this scheme.
Let us introduce the Balance Responsible Party (BRP) as the party that ascertains that
the wind together with other generation units in a pool follows the planned schedule for
the pool. The BRP may not be a generator, but some party that has the required tools to
effectively predict and manage the different pool members in an optimal way seen from the
pool members point of view.
The first milestone for the BRP is to bring together a sufficient amount of uncorrelated
wind power in the pool. The lower the correlation on timescales greater or equal than an
hour, the lower the forecasting error and the higher the energy price.
Low correlation means that the area aggregation over individual sites gives a smoother
output signal, which is easier to forecast. The same smoothing is then implicitly applied to
the forecast. The net result is that the phase on the shortest timescale of the individual wind
farms is invisible on the aggregated power generation.
A large volume of wind power also helps to level out negative impact from problems due
to restrictions on the grid or reduced turbine availability, although these may only appear as
small errors in a large pool.
One of the key parameters for a BRP is to secure a permanently high level of uncertainty
of when and how much energy the pool is delivering to the network. As discussed earlier,
this is because the wind power trading becomes vulnerable to speculation from stronger
market participants, if it can easily be derived how much wind energy needs to be traded on
the market. This is not a trivial task for the BRP, because there are also other constraints,
which limit the possible ”confusion” level introduced to the market.
Hence, the major tools that the BRP needs apart from the wind farm capacity are:
•
A storage unit
•
An ensemble prediction system for the difference between demand and wind power
•
A short-term prediction system for wind power and demand
Search WWH ::
Custom Search