IEEE Power & Energy Magazine - July/August 2021 - 83

demand or decreasing generation, the shift from other fuels
(e.g., gas) to electricity in an MES proved to be much cheaper
than the curtailment of renewables. For instance, based on
market prices for electricity and gas in 2019 in Italy, the investigated
CHP could have paid for negative flexibility provision
a price that was, on average, up to 26.8% of the day-ahead
market price. In contrast, the curtailment of renewables to
provide the same negative flexibility would have caused generation
losses. Depending on the rules in place for renewables
support and the flexibility market, the reduction of generation
would be evaluated in a range from no additional costs to the
full market price. In the context of an efficient framework of
market operation, network tariffs, and taxation, an MES could
thus reduce the need to curtail renewable generation and
simultaneously reduce its carbon emissions. For example, an
investigation of published auction results indicates that MESs
already have a tangible impact on ancillary service markets
in Germany.
Results from the project highlight the impact of uncertainty
on MESs and aggregators that participate in the market. Specifically,
while the provision of flexibility generally leads to
additional revenues and cost savings, the quality of the price
forecasting has a strong influence on the revenues. Fair and
transparent rules for distributing portfolio revenues among
participating units are therefore key. The aggregator should
implement a dynamic revenue sharing method and differentiate
between earnings for reserve provision and energy provision.
The provision of internal backup capacity should also be considered
as an internal service and be rewarded even if the capacity
is not activated. As another interesting point, in frequency
reserve markets with low penalties (e.g., three times the market
price) for not meeting service requirements and where the
activation probability is low, as in the Austrian mFRR market,
a flexibility provider might be tempted to ignore backup rules
since the additional revenues from increased bidding volumes
can exceed penalty payments.
Innovative Market Designs
for Synergy Maximization
Electricity, gas, and heat systems are economically decoupled,
with little harmonization. They have separate markets
with divergent timeframes and different characteristics. In a
recent European Union " Strategy for Energy System Integration "
communication, the European Commission stressed the
need for efficient markets, where prices reflect all the costs
of energy carriers as an important pillar of a climate-neutral
economy. One way to achieve this is to redesign energy markets
so that synergies between different carriers are accounted
for. Several market designs can be put forward to better reflect
the interaction between carriers. The authors' work specifically
focused on heat, gas, and electricity, but the concepts
could easily apply to other carriers. The proposed designs
range from a single-carrier market, with separate, sequential
day-ahead markets for different carriers (decoupled), to
a fully integrated multicarrier market, in which dependencies
july/august 2021
between carriers are explicitly considered in products and
clearing processes. We also proposed designs with single and
multicarrier markets at local and global levels.
In the decoupled market design, there are separate (single)
day-ahead markets for different carriers, with each run by its
market operator. Market timings are such that participants
can readjust their positions for the next markets, considering
the clearing outcome of the previous ones. In contrast, in the
integrated market design, there is one market for different
carriers run by a unique operator that matches all the orders
from local and global levels. While one could argue about
the implementation challenges of such an approach, it can be
usefully considered as a benchmark for the greatest social
welfare, which most closely approaches centralized multienergy
models already presented in the literature. Possible
coordination mechanisms based on a decomposition of such
a single-operator, integrated multi-energy market design are
discussed at the end of this section.
Novel " conversion orders " introduced into the integrated
market design facilitate considering the interactions between
carriers during market clearing. Conversion orders enable buying
energy from or selling energy to one carrier, depending on
the realized market prices of another carrier at the same time
instance. They are a tool for market participants to mitigate
risks related to the price forecasts of different carriers and a
means to couple the markets of different carriers. A conversion
order is specified by the amount of energy a market participant
is willing to convert at a given time step of the market clearing
horizon, a given efficiency, and a desired (conversion) price
spread. As opposed to the existing complex orders in European
day-ahead markets (such as block and linked orders), conversion
orders impact both the constraints of the market clearing
algorithm and its objective function.
Our simulation results showed that, in the context of the
integrated market design, the global maximization of social
welfare could be achieved, while in the context of the decoupled
market design, imperfect forecasts typically result in a
loss of profit and technical infeasibilities for conversion technologies
and subsequently less overall economic efficiency.
However, in the integrated market, participants would have
to share more information about their underlying portfolios
(e.g., conversion efficiency and the price of conversion).
Finally, since an integrated design involves only one market,
a completely different market organization is necessary.
Table 4 gives an overview of the differences between
decoupled and integrated market design approaches. As a
successive step, we explored alternative integrated designs.
This was done to preserve the current organizational structure,
with separate market operators for each carrier, while
enabling the incorporation of multicarrier orders. These
integrated decentralized market designs (Table 4) are based
on different mathematical decompositions of the integrated
market design clearing algorithm. We analyzed two integrated
decentralized designs and proved that, under mild
assumptions, they converge to the same solutions as the fully
ieee power & energy magazine
83
IEEE Power & Energy Magazine - July/August 2021

Table of Contents for the Digital Edition of IEEE Power & Energy Magazine - July/August 2021
