Page 318 - Decision Making Applications in Modern Power Systems

Decision-making-based optimal generation-side Chapter | 11 279

RES and the demand side into account, while exploiting the controllability
of certain network components.
The main problems that need to be taken into account are as follows:
1. Probabilistic security: Probabilistic variants of deterministic SC-OPF
problems need to be developed, providing enough flexibility to quantify
the trade-off between security and economic system operation.
2. Production and generation-side reserve scheduling: Within a security-
constrained probabilistic framework standard, day-ahead planning pro-
blems such as production and reserve scheduling need to be revisited.
3. Exploiting demand response for reserve provision: In an uncertain envi-
ronment, demand-side resources should be taken into account a decision
mechanism to provide ancillary services while reducing the cost that
would occur is reserves were solely purchased from the generating units.
4. Exploiting component controllability: Corrective control actions offered
by certain network components could result in a more economic opera-
tion of the network, especially in the cases where the level of uncertainty
is increasing.
5. Development of new algorithms and tools: To address the problem of tak-
ing optimal decisions in the presence of uncertainty, new algorithms for
stochastic scheduling with guaranteed performance need to be developed,
and the (probabilistic) properties of the obtained solutions should be
reinterpreted.

11.3 Decision-making application to reserve scheduling

Due to the ever increasing installed capacity of RES, for example, wind and
photovoltaic, which are ever changing and are weather dependent, it is nec-
essary to revisit certain operational concepts, such as (N 2 1) security and
reserve scheduling. In this framework the power required to balance the sys-
tem is compensated by each generator with a fixed percentage, that is, fixed
distribution vector; hence, the reserves of each generator are then determined
by the worst-case value of the power mismatch. Here, the required reserves
that the systems operator needs to purchase via the probabilistic approach
can be determined but do not optimally distribute them to the generating
units.
The aim of this section is to optimally allocate the reserve requirements
to the generators.
In today’s different electricity markets, the goal is to minimize the gener-
ation dispatch and the reserve costs, while satisfying the network constraints.
The generation dispatch is determined by the energy market, while the net-
work constraints are determined by transmission market, so that the network
security is guaranteed, for example, the N 2 1 security criterion. Usually, a
reserve capacity of units is predetermined and their optimality in dispatch is