IEEE Signal Processing - July 2018 - 99

■

efficient energy-trading experience by matching the market
participants' sell and buy orders in near real-time granularity. In market operations, the constraint of energy generation
influences the thresholds of a maximum and minimum
allocation of energy. Different market-time horizons can
exist in the market operation (e.g., to cover various stages
of the electricity market), and the market operation should
be able to produce efficient allocations at every stage.
The pricing mechanism: The objective of a pricing mechanism is to efficiently balance energy supply and demand,
and it is implemented as a part of the market operation.
Examples of pricing mechanisms include auctions with an
individual or uniform clearing price. Pricing mechanisms
for P2P energy trading have a significant difference compared to that of the traditional energy market. In particular,
with traditional energy, a large part of energy prices consists
of taxes and surcharges, whereas in a P2P trading market,
taxes and surcharges are absent due to the zero marginal
cost of renewable energy. Regardless, pricing needs to
reflect the state of energy within the P2P energy network,
e.g., a higher surplus should lower the price of P2P energy
trading and vice versa.
The automatic energy-management system (AEMS): The
purpose of an AEMS is to secure the supply of energy for a
market participant while implementing a specific bidding
strategy. To do so, an AEMS has access to the real-time
supply and demand information of its market participants,
and, based on these data, an AEMS forecasts the generation and consumption profile and develops the bidding
strategy. The AEMS of a rational user would always buy
energy on the microgrid market when it falls below its
maximum price limit. Individual agents' intelligent bidding
strategies employ varying prices at different times and are
expected to be a core components of active P2P energy
markets in the future.
The regulation: Regulation is a key feature of P2P energy
trading, and it determines how these markets fit into the
current energy policy, i.e., governmental rules decide what
market design is allowed, how taxes and fees are distributed, and in which way the market is integrated into the traditional energy market and energy supply system. As a
result, governments can either support P2P energy markets
to accelerate the efficient utilization of renewable energy
resources and to decrease environmental degeneration by
regulatory changes, or they can discourage the implementation of such markets if these result in negative impacts on
the current traditional energy system.

Brooklyn TransActive P2P project
Now, we focus on an existing pilot project on P2P energy trading built in Brooklyn, New York. This discussion of a real P2P
energy network will provide the reader with an idea of how
P2P energy trading is being envisioned for future energy markets. Local P2P energy trading, absent any utility involvement,
is yet to be covered by regulations that decide how such markets fit into the current energy policy [9]. The choice of the

BMP for this discussion is motivated by the breadth of the
project as well as by the successful implementation of trading
techniques shown in their recent pilot demonstration. The
BMP consists of a microgrid market in Brooklyn, New York,
and is run by LO3 Energy. The participants of the BMP are
located across three distribution grids, including Brooklyn
Borough Hall, Park Slope, and Bay Ridge. As shown in
Figure 2, the BMP trading network consists of physical and virtual layers. In the physical layer, the BMP uses the traditional
grid to supply physical energy flow; however, it also has a
physical microgrid network among a limited number of housing blocks, which is comprised of 10 # 10 housing blocks that
can be decoupled from the main grid in case of an emergency.
The virtual layer, which is completely separated from the physical layer, is implemented on top of the existing physical grid
infrastructure, provides the technical infrastructure for the local
electricity market, and is based on a Tendermint protocolbased private blockchain called the TransActive blockchain
architecture [9]. While each peer must have a blockchain
account to participate in the P2P energy trading, a meter is
installed in the house of each peer that communicates with the
corresponding blockchain account and transfers energy generation and demand data from the TransActive meter.
The energy trading in the BMP is mostly done automatically by an AEMS and only requires several preferences of its
market participants. The participants must use a mobile application (the BMG app), through which they can choose their
preferences on the source of energy and price limits for the
AEMS to conduct the energy trading. An example diagram
of the mobile application is shown in Figure 3. Although the
participants can change their preferences at any time, it is
also possible for the participants to choose and set one preference for all future use without any further interaction with the
mobile application. Once the preferences are submitted by the
participants, the energy trading between two participants (one
consumer and one prosumer) takes place following a step-bystep process [9].
■ Step 1: The buy and sell orders of a consumer and a prosumer, respectively, are submitted to the market by their
AEMSs. Any buy or sell order consists of a quantity and
a price.
■ Step 2: The market mechanism is a closed-order book with
a time-discrete double auction in 15-min intervals. In the
double auction, 1) consumers repeatedly bid up to their
maximum price for their preferred energy sources, 2)
prosumers bid the minimum price that they request for selling their generated energy on the market, 3) the highest
bidder is allocated first and lower bidders are allocated following a merit-order dispatch, and, finally, 4) the last allocated bid price represents the market-clearing price for that
particular time slot.
■ Step 3: Consumers that cannot undercut the clearing price
are supplied by other sources.
■ Step 4: Financial transactions are carried out between the
allocated market participants of that particular time slot
according to predefined payment rules.

IEEE Signal Processing Magazine

July 2018

Table of Contents for the Digital Edition of IEEE Signal Processing - July 2018