Information Technology Reference
In-Depth Information
simulated and recreated into a MATLAB class. Thus, at the end of every simulation ,
the Clearing House recall the Offline Statistical Module which carry out statistical anal-
ysis as well as visualization of the computational experiment results. Finally, it is worth
remarking that GAPEX allows direct generalization, as it is possible to create different
types of agents, thus allowing the design of extremely realistic agent-based models.
3
Agent-Based Modeling of the Italian Electricity Day-Ahead
Market
As discussed in previous Section, GAPEX is designed as a powerful and extensible
agent-based framework for electricity market modeling and simulation. Current ver-
sion of GAPEX allows one to simulate different power exchange protocols, but due to
its complex structure, in this paper attention is dedicated to the Italian power exchange.
It is worth remarking that a power exchange strongly differs from a stock market from
both structural and behavioral point of view. From the former, the power exchange
mechanism is a uniform double auction whereas the stock market one is continuous
time limit order book. Furthermore, energy is not a storable good (i.e.,
buy&hold
strat-
egy are not even possible) whose consumption is contemporary to the production and
is characterized by strong seasonality (i.e., daily, weekly and yearly). Moreover, from
the latter, the electricity sector is characterized by strong oligopoly (i.e., a limited and
basically time-invariant number of market traders) that repeat the same game on a daily
base. Theoretically speaking, such economic system seems perfect for an analytical
solution based on game theory, but the dimension of the game is so high that it practi-
cally impossible to study equilibria by means of traditional game theory. Despite a first
glance on analytical solutions, all these elements lead to an economic system that can
be effectively studied by means of a computational approach based on learning agents,
thus motivating the development of GAPEX framework for the implementation of the
model of the wholesale Italian Electricity Market. Making use of preliminary versions
of GAPEX, [7] described and implemented an agent-based model of power exchange
with a uniform price auction mechanism and a learning mechanism for the Gencos.
Moreover, [13] provided the first version of the Genoa Artificial Power Exchange and
compared the discriminatory and the uniform price auction mechanism with heteroge-
nous agents. Finally, [19] firstly attempted to create an agent-based model of the Italian
Electricity Market, with a reduced transmission network grid and a simplified descrip-
tion of GenCos. It is worth remarking that version presented and discussed in this pa-
per of both the GAPEX and the agent-based model of the Italian electricity day-ahead
market are characterized by significant extensions. Firstly, agent-based model incor-
porates now the exact procedure employed by the
Gestore Mercati Energetici S.p.A.
(hereafter GME) [9] thus overcoming the limitation of previously adopted formulation
that resulted a constrained ill posed optimization procedure. Furthermore, the cogni-
tive agents in the GAPEX make use of the Enhanced Roth-Erev reinforcement learning
algorithm (presented and discussed in Section 4), developed so to take into account pay-
offs of any sign. There are crucial features that allowed us to calculate the energy prices
based on scenarios that correctly emulate real power plants, real transmission limits
and real bids. In this Section we present the agent-based model of the Italian Electricity
