An Overview of Local Electricity Market Designs

Jovana Forcan Miodrag Forcan

Department of Mathematics and Informatics Department of Power Systems
2022 18th International Conference on the European Energy Market (EEM) | 978-1-6654-0896-7/22/$31.00 ©2022 IEEE | DOI: 10.1109/EEM54602.2022.9921020

Faculty of Philosophy Faculty of Electrical Engineering

University of East Sarajevo University of East Sarajevo
East Sarajevo, BiH East Sarajevo, BiH
[Link]@[Link] [Link]@[Link]

Abstract— The increasing amount of energy production from Distributed generation (DG) capacity is expected to
distributed energy resources leads to a re-examination of the significantly increase in the near future according to the
existing electricity market designs. Local electricity markets International Energy Agency [1]. At the same time, there is a
could enable more flexibility in managing the energy uncertainty clear trend of decreasing feed-in tariffs for distributed
of the increasing number of prosumers. In this paper, an overview generation (DG) [2], even down to a level lower than the retail
of local electricity market designs is given. The role of local electricity price [3], [4]. According to previous circumstances
electricity markets within the structure of wholesale and retail DG is incentivized to increase the utilization of generation
electricity markets is elaborated. The most recent literature locally [5], thus contributing to establishing the local electricity
review on local electricity markets is given. This paper also
markets (LEM). The main players participating in LEM are
discusses the mathematical frameworks that are commonly used
in the design of local electricity markets. The special objective is
expected to be microgrids, prosumers, consumers, and retailers.
to analyze the potential application of game theory-based Distribution system operator (DSO) is often envisioned to have
approaches. The commonly used optimization and equilibrium the MO role in LEM [6]. The conceptual EM architecture
models are studied and compared. In order to reveal the roles and envisioning the role of LEM within the architecture of the
benefits of different local electricity market participants, the existing wholesale and retail electricity markets is presented in
selected local electricity market designs are computationally Fig.1.
implemented and mutually compared. The case studies are LEM starts from separating retailers from end-consumers
performed on the example of the conventional model of
and incorporating the microgrids and prosumers as market
benchmark test power network. Based on the numerically
obtained results the specifics of each market design are
players. The microgrids are envisioned here as the groups of
elaborated. prosumers and consumers coordinated by the microgrid
operators. Microgrid operator is envisioned to participate in
Index Terms-- Distribution network, Game theory, Local LEM on behalf of the microgrid. Retailers are not considered to
electricity market, Prosumer. be an active part of the LEM. They role is to purchase the
electric energy from the wholesale EM which is necessary to
I. INTRODUCTION fulfill the energy requirements of the remaining participants
(consumers and prosumers that are not LEM participants).
Deregulation of the electricity market (EM) is becoming
However, retailers are subject to the results of the LEM, which
more prevalent in countries around the world since it is usually
could make their actions to change in the wholesale EM.
followed by greater availability of renewable energy options.
Additional analysis is required to get insight into single or
EM design should consider the market power of different
multiple retailer action change as the result of LEM. In It is not
players to establish fair and sustainable trading. Transmission
expected that microgrids and prosumers alone could enable
level EM designs have been already well-established and used
100% of energy demands in distribution system. so-called
by many countries in Europe and North America. The most
prosumer-centric LEM it is expected that prosumers would
commonly, generating companies (GenCos) and retailers
have a certain market power, thus behaving as price makers.
participate and trade in wholesale EM under the supervision of
MO (DSO?) would be responsible for establishing the LEM
market operator (MO) which is responsible for market clearing.
mechanisms enabling fair trading and incentivizing DG
MO tasks are usually performed by independent system
participation.
operator (ISO) or transmission system operator (TSO). The
electricity price in wholesale EM is often variable on hourly This paper investigates the potential LEM designs and gives
basis and it could be also dependent on location (nodal pricing). an overview of the recently envisioned solutions. The rest of the
Retailers often operate in a way to offer the fixed electricity paper is organized as follows. Section 2 presents the state-of-
price (retail electricity price) to consumers or to encourage them the-art literature review on LEM and discusses the commonly
to participate in demand response programs. used mathematical frameworks in market design. Game theory

978-1-6654-0896-7/22/$31.00 ©2022 IEEE

Authorized licensed use limited to: UNIVERSITY COLLEGE CORK. Downloaded on December 10,2024 at [Link] UTC from IEEE Xplore. Restrictions apply.
mathematical framework is elaborated in Section 3. Section 4 A. Non-strategic prosumers (price-taker model)
presents the results of the numerical analysis applied to selected In this LEM model prosumers have zero or little market
LEM designs with the main goal of revealing the advantages power and simply accept the market prices as given (no
and disadvantages of different solutions. Finally, the paper is strategic behavior) [5]. A typical example is the family of
concluded in Section 5. Lagrangian decomposition methods, widely used as a market
mechanism in LEMs [9]. The main objective of LEM design is
a social welfare maximization. There could be a single price for
all LEM participants (market clearing price), or locational
marginal pricing (LMP) principle could be applied. The most
common three pricing strategies are identified in [5]:
community internal price, central market coordinator, and
iterative bilateral contract. The main drawback of this LEM
model is the lack of verification of prosumers’ satisfaction and
market stability [5]. Lagrangian methods are not necessarily
suitable for objectives such as fairness, profit maximization,
and equilibrium discovery for non-convex or strategic settings
[9].
The most recent examples of LEM designs based on the
Lagrangian decomposition methods are available in [10-12]. In
[10], a community manager (CM), as a non-profit market entity,
is proposed to be responsible for coordination of prosumers in
achieving optimality conditions. The CM is envisioned to be
recognized as a trusted third party (TTP) in [11]. An integrated
pricing mechanism considering social welfare maximization
and network voltage management is proposed in [12].
B. Strategic prosumers (price-maker model)
Game theory represents a very suitable mathematical
framework to capture strategic behavior of individual rational
prosumers in LEM [5]. In cases where the perfect competition
assumption is relaxed, game theory is the formal framework for
conducting market analysis [9]. The use of game-theoretic
approach for LEM has been recently reviewed in [13]. LEM
could be designed using both, non-cooperative and cooperative
game-theoretic approach.
Non-cooperative game theoretic schemes focus on finding
the market equilibrium [5]. The finding of Nash equilibrium
Figure 1. The conceptual EM architecture envisioning the role of LEM
guarantees that no player has the incentive to change their
II. LITERATURE REVIEW ON LEM strategy while the strategies of other players remain fixed [14].
To solve for Nash equilibrium, the papers [15], [16] propose
Three potential prosumer-centric market models are iterative communication between prosumers and LEM
identified in [7]: prosumer-to-grid, community-organized operator. In [17], it is shown that Nash equilibrium could be
prosumer groups, and peer-to-peer. Prosumer-to-grid model achieved together with the maximum social welfare. The
envisions the grouping of prosumers in interconnected or
framework of Stackelberg games is proposed for LEM in [18].
islanded microgrids. Goal-oriented prosumer clusters represent
the basic structural element in the model envisioning However, the Nash equilibrium is inefficient for maximizing
community-organized prosumer groups potentially leading to the sum of prosumers’ profits [19].
formation of the virtual power plants (VPP). The least The focus of cooperative game-theoretic approach is on the
structured peer-to-peer model relies on direct interconnection allocation of the total profit among the participants, who are
of prosumers selling and buying services from each other. The assumed to be profit driven in forming coalitions [20], [5]. The
potential LEM designs and clearing approaches are reviewed in main criteria for evaluating the profit allocation among
[8]. A recently published paper [9] presents a review of models, prosumers participating in coalitions are the fairness and the
solution concepts and algorithmic techniques related to LEM.
level of incentive [21]. The cooperative game-theoretic
According to economic objectives, LEM models can be divided
approach is proposed for prosumers selling surplus PV
on system-focused (maximize the collective economic outcome
of the participants) and individual-focused (analyzing the generation in [22]. The Shapley value criteria is selected as a
economic outcome of individual market participants) [5]. payoff allocation method. A multi-timestep LEM scheme based
on cooperative game theory is proposed in [23], [24].

Authorized licensed use limited to: UNIVERSITY COLLEGE CORK. Downloaded on December 10,2024 at [Link] UTC from IEEE Xplore. Restrictions apply.
Prosumers can behave strategically even if they are not The study is performed on IEEE 33 node distribution test
aware of the LEM mechanism using artificial intelligence (AI) feeder [30], and decision horizon is set to be 1 hour. We have
algorithms [9]. The modeling framework proposed in [25] assumed that there are 6 prosumers located at nodes 10, 18, 22,
define GenCos as agents capable of learning their optimal 25, 26, and 33. Tab. 1 shows the parameters for generation
strategies by utilizing experiences acquired from repeated costs (ai and bi), generation capacities and load demand for
interactions with the market clearing process. Deep every prosumer. Generation cost parameter ci is set to 0 for
reinforcement learning is applied in previous work. The same every prosumer. Electricity price coefficients 4 and 5 are set
modeling framework could be applied to prosumers in LEMs. to be 0.001 $/(kW)2 and 5 $/kW, respectively. According to the
given IEEE test feeder system’s data sheet, the total load
III. GAME THEORY MATHEMATICAL FRAMEWORK
demand is 4850 kW. Power loss is calculated using a well-
Game theory is a mathematical discipline used to study known Newton-Raphson method.
problems of conflict among interacting decision-makers [26].
A n-person game can be described as three-tuple (N, Xi, ) TABLE I. Parameters describing prosumers
Prosumer Demand ai bi Capacity
where N = {1,2,…,n}, is a set of players, Xi is the set of
: → ℝ is payoff function,
number and node (kW) ($/kW2h) ($/kWh) Cpi (kW)
strategies of player i and Prosumer 1 (10) 60 0.015 0.75 400
expresses the profit that a player obtains corresponding to the Prosumer 2 (18) 90 0.02 1 500
strategy of players [27]. All players, when acting together, can Prosumer 3 (22) 90 0.02 1 400
take a collective action, which is a vector = , ,…, . Prosumer 4 (25) 420 0.00875 0.5 500
⊆ × ×… ×
Prosumer 5 (26) 60 0.01 1 500
Given a collective action set and
= , ,…, , =
Prosumer 6 (33) 1200 0.00875 0.875 300
vector of collective action
, ,…, ∈ , | ≡ , ,…, , , ,…,
A. Nomenclature
of the collective action set can be interpreted as a collection of
– generation cost of the ith prosumer ($/kW2h)
6 – generation cost of the ith prosumer ($/kWh)
actions when the ith agent switches from the strategy xi to yi

7 – generation cost of the ith prosumer ($/h)

while the remaining agents are playing xj, j = 1,2,…,i – 1, i +
1, …, n. Nash equilibrium is the concept by which player
cannot benefit by changing only his own strategy while the 8 – revenue function of ith prosumer ($)
other players keep theirs unchanged. A point ∗
= 9 – cost function of ith prosumer ($)
, ,…,
∗ ∗ ∗
is Nash equilibrium point if for each i: : – total active power generation of i-th prosumer (kW)
∗
= ∗
! "
| ∗ . :;<, – distributed generator power output of the ith prosumer
(kW)
:1=>?@, – solar power output of the ith prosumer (kW)
At x* no player can improve his individual payoff by his own

:A B, – wind power output of the ith prosumer (kW)

action. Nikaido-Isoda function transforms Nash equilibrium

:B, – load demand of the ith prosumer (kW)

problem into an optimization problem. It is defined as
Ψ , = ∑& [ | − ].
The optimum response function, the result of maximizing the :>=11 – total active power loss (kW)
Nikaido–Isoda function, at the point x is :CD > D/ – power supplied by the utility company (kW)
) = arg max/∈0 Ψ , , ,) ∈ . E – payoff function of the utility company ($)
The optimization problem can be solved iteratively using 4 – electricity price coefficient $/(kW)2
relaxation algorithm until it converges to Nash equilibrium 5 – electricity price coefficient $/kW
point. The termination condition for the relaxation algorithm is
max/∈0 Ψ 1 , < 3. For the chosen small value ε the Nash
B. LEM model case studies
equilibrium is obtained when the condition is satisfied, where The objective function for prosumer is defined in the
following way:
max F = 8 − 9 .
s is the iterative step of the relaxation algorithm.
More about these concepts can be find in [27]-[28]. (1)
The fixed electricity price p = 0.15 $/kW is set for price-
IV. NUMERICAL ANALYSIS taker model. The price-consumption curve is defined for Nash-
Using the methodology presented in [29] we have analyzed Cournot LEM model in the following way:
the examples of prosumer centric LEMs, where the prosumers G ?1H = 5 − 4 ∙ ∑ : , (2)
and consumers are considered as participants. Prosumers can
satisfy the self-load by purchasing electricity from the utility or where : = :;<, +:A B, +:1=>?@, .
generating the electricity from its own local generators such as
solar panels, small wind turbines, and diesel generators. It is assumed that every prosumer i has the following power
Prosumers are selling their electricity surplus to other capacity share: PDG,i,max = Cpi/5, Pwind,i,max = (Cpi - PDG,i)/4,
participants [29]. The aim of utility is to maintain the system Psolar,i,max = 3(Cpi - PDG,i)/4.
reliability and security. The utility makes a profit by selling the
electricity to consumers if prosumers’ generated electricity
cannot meet the needs of consumption. Trading analysis for
participants is performed using price-taker model and Nash-
Cournot game theory model.

Authorized licensed use limited to: UNIVERSITY COLLEGE CORK. Downloaded on December 10,2024 at [Link] UTC from IEEE Xplore. Restrictions apply.
1) Price-taker LEM model (Case 1) + G ∙ :>=11 + G ?1H ∙∑ & :. (13)
The revenue function and generation constraints can be In this case the prosumers’ revenue is calculated using the
expressed as follows: price pnash, while the cost for prosumers and utility’s payoff is
8 = G ∙ :;<, +:A B, +:1=>?@, ), (3) calculated using the price p. For calculating consumers’ cost
0 ≤ :;<, ,L ≤ :;<, ≤ :;<, ,L?M ∀O
both prices pnash and p are used.
C. Results and discusion
0 ≤ :A B, ≤ :A B, ,L?M ∀O (4) Tab. 2 shows the implementation results obtain by analyzed
0 ≤ :1=>?@, ≤ :1=>?@, ,L?M ∀O. LEM models. For every prosumer i, the table gives distributed
generator power output :;<, , total power output : and the
The cost function is defined as follows:
profit. It is also shown in Tab. 2 how much electricity utility
9 = ∙ :;<, + 6 ∙ :;<, + 7 + G ∙ :B, . (5) needs to produce to meet consumption of the grid, and the
utility’s profit.
The generation costs of solar panels and wind turbines are
TABLE II. SIMULATION RESULTS
assumed to be negligible in this analysis.
The payoff of utility is defined as follows: Case 1 Case 2a Case 2b

E = G ∙ :CD > D/ ,
PDG,1 - - -
(6) Prosumer 1 P1 320 320 320
Payoff 39.0 46.63 48.4
subject to the constraint of power flows: PDG,2 - - -
∑ & : + :CD > D/ − :>=11 = :D=D?> .
Prosumer 2 P2 400 400 400
(7) Payoff 46.5 55.6 58.24
PDG,3 - - -
The total cost of consumers is defined in the following way: Prosumer 3 P3 320 320 320
9Q= 1CLR@1 = G ∙ :D=D?> − ∑ & :B, " + G ∙ :>=11 . (8) Payoff 34.5 41.25 43.9
PDG,4 - - -
Prosumer 4 P4 400 400 400
2) Nash-Cournot LEM model (Case 2)
Payoff -3.0 -3.59 8.74
In the game theoretic approach, there are 6 prosumers and PDG,5 - - -
the utility, such that utility is not equal player with prosumers. Prosumer 5 P5 400 400 400
The utility must supply the grid with the remaining electricity Payoff 51.0 60.98 62.74
PDG.6 - - -
since the total capacity of prosumers does not allow them to
Prosumer 6 P6 240 240 240
generate enough power to meet the total load demand. Payoff -144.0 -172.19 -136.95
Prosumers compete as independent players with the aim to Putility 3019.4 3019.4 3019.4
Utility
maximize they own profits and satisfy the constraints. Payoff 452.91 541.56 452.91
Consumers Cost 476.91 570.25 537.98
a) Uniform price (Case 2a)
The revenue function for prosumer i can be expressed as According to results, in all considered cases, the value of the
follows: cost function for diesel generator of each prosumer is too high,
8 =G ?1H ∙ :;<, +:A B, +:1=>?@, ). (9) so it is not in prosumers’ best interest to produce electricity
using diesel generator as this reduces their profits (PDG = 0).
The cost function is defined as follows: Model from Case 1 gives the smallest cost for consumers, as
9 = ∙ :;<, + 6 ∙ :;<, + 7 + G ∙ :B, . (10)
the market price is the lowest. Prosumers 1, 2, 3 and 5 achieved
?1H the lowest profit in Case 1. For prosumers 4 and 6 model from
The payoff of utility is defined as follows: Case 1 gives better payoff than the model from Case 2a. This
E= G ∙ :CD > D/ .
is because prosumers’ cost in Case 2a is calculated using the
?1H (11) price G ?1H while in Case 1 their cost is calculated by using the
The total cost of all consumers is defined in the following price p. For prosumer 4 only the model from Case 2b gives the
way: positive payoff. The prosumers achieved the largest profits in
9Q= 1CLR@1 = G ?1H ∙ :D=D?> − ∑ & :B, " + G ?1H ∙ :>=11 . (12) Case 2b. This is because prosumers’ revenue is calculated
using the price G ?1H and their cost is calculated by using the
b) Non-uniform price (Case 2b) price p. All models give the negative payoff for prosumer 6 as
The revenue function for prosumer i is defined by it has the small generation capacity and high demand.
expression (9) subject to (4). The cost function is defined by With all three models, prosumers produce the same amount
(5). The payoff of utility is defined by expression (6). The total of electricity, as in all three cases they produce the maximum
cost of consumers is defined in the following way: electricity from solar panels and wind turbines, while they all
decide not to use diesel generators. Compared to model from
9Q= 1CLR@1 = G ∙ S:D=D?> − T :B, − T : U + Case 2b in which prosumer 4 achieves the largest profit, when
& & using a model from Case 1 and Case 2a, prosumer 4 needs to

Authorized licensed use limited to: UNIVERSITY COLLEGE CORK. Downloaded on December 10,2024 at [Link] UTC from IEEE Xplore. Restrictions apply.
purchase the electricity from other prosumers or utility to significantly rise. The high electricity prices could attract the
satisfy its load demand. new prosumers to join the market, consequently lowering the
price in environment with more competition. However, in the
Fig. 2 shows the convergence of prosumers’ power outputs first phase of LEM implementation with small number of
to an equilibrium condition in Case 2a and Case 2b. After 23 prosumers it could be convenient to partially regulate the
iterations the termination condition is satisfied, and the system electricity price until the competition environment is achieved.
reaches a Nash equilibrium point. Solar and wind power
outputs coincide in Case 2a and Case 2b. Fig. 3 shows the price REFERENCES
G ?1H at each iteration step. The price starts from 0.3 and then [1] Market analysis and forecast from 2019 to 2024, Fuel report, Oct. 2019,
it gradually converges to 0.1794 $/kW, in both, Case 2a and [Online]. Available: [Link]
Case 2b. [2] H. Klinge Jacobsen and S. T. Schroder, “Curtailment of renewable
generation:Economic optimality and incentives,” Energy Policy, vol. 49, pp.
663–675, 2012.
[3] [Link]. (2019, Feb.) Feed-in tariffs: get money for generating
your own electricity. [Online]. Available:
[Link]
[4] H. Wirth, “Recent facts about photovoltaics in Germany,” Fraunhofer ISE,
vol. 1, no. 100, pp. 92, 2017. [Online]. Available:
[Link]
[Link]
[5] L. Han, “A scalable cooperative game theoretic framework for coalitional
energy management”, Ph.D. dissertation, Mansfield College, Univ. of Oxford,
2000.
[6] M. Ampatzis, P. H. Nguyen and W. Kling, "Local electricity market design
for the coordination of distributed energy resources at district level," in Proc.
2014 IEEE PES Innovative Smart Grid Technologies, Europe, pp. 1-6, doi:
10.1109/ISGTEurope.2014.7028888.
[7] Y. Parag and B. Sovacool, “Electricity market design for the prosumer era”,
Figure 2. The convergence of prosumers’ power outputs to an equilibrium Nat Energy 1, 16032, March 2016, [Link]
condition [8] M. Khorasany, Y. Mishra, and G. Ledwich,” Market framework for local
energy trading: a review of potential designs and market clearing approaches”,
IET Gener. Transm. Distrib., vol. 12, no. 22, pp. 5899-5908., 2018,
[Link]
[9] G. Tsaousoglou, J. S. Giraldo, and N. G. Paterakis, “Market Mechanisms
for Local Electricity Markets: A review of models, solution concepts and
algorithmic techniques”, Renewable and Sustainable Energy Reviews, vol.
156, pp. 111890, 2022, [Link]
[10] F. Moret and P. Pinson, "Energy Collectives: A Community and Fairness
Based Approach to Future Electricity Markets," IEEE Transactions on Power
Systems, vol. 34, no. 5, pp. 3994-4004, Sept. 2019, doi:
10.1109/TPWRS.2018.2808961.
[11] J. L. Crespo-Vazquez, T. AlSkaif, Á. M. González-Rueda and M.
Gibescu, "A Community-Based Energy Market Design Using Decentralized
Figure 3. The convergence of the electricity price to an equilibrium Decision-Making Under Uncertainty," IEEE Transactions on Smart Grid, vol.
condition 12, no. 2, pp. 1782-1793, March 2021, doi: 10.1109/TSG.2020.3036915.
[12] A. Paudel, M. Khorasany and H. B. Gooi, "Decentralized Local Energy
The payoff of the utility in Case 1 and Case 2b is smaller Trading in Microgrids With Voltage Management," IEEE Transactions on
than in Case 2a although it needs to produce the same amount Industrial Informatics, vol. 17, no. 2, pp. 1111-1121, Feb. 2021, doi:
of electricity in all three cases. This is because its payoff in 10.1109/TII.2020.2980160.
Case 2a is calculated by using the price G ?1H . The power [13] W. Tushar, C. Yuen, H. Mohsenian-Rad, T. Saha, H. V. Poor and K. L.
Wood, "Transforming Energy Networks via Peer-to-Peer Energy Trading: The
losses in distribution network in Case 1, Case 2a and Case 2b Potential of Game-Theoretic Approaches," IEEE Signal Processing
are 249.38 kWh as prosumers and the utility produced the same Magazine, vol. 35, no. 4, pp. 90-111, July 2018, doi:
amount of electricity in all three cases. 10.1109/MSP.2018.2818327.
[14] P. Berenbrink, T. Friedetzky, L. A. Goldberg, P.W. Goldberg, Z. Hu, and
V. CONCLUSION R. Martin,“Distributed selfish load balancing,” SIAM Journal on Computing,
Selecting the most suitable design for LEM is not an easy vol. 37, no. 4, pp. 1163–1181, 2007.
[15] H. Wang and J. Huang, "Incentivizing Energy Trading for Interconnected
task. The prosumer-centric LEM schemes seem to be the first
Microgrids," IEEE Transactions on Smart Grid, vol. 9, no. 4, pp. 2647-2657,
choice to select according to analyzed literature since the July 2018, doi: 10.1109/TSG.2016.2614988.
prosumers are most likely to be incentivized to participate in [16] N. Liu, X. Yu, C. Wang and J. Wang, "Energy Sharing Management for
LEM. Game theory is shown to be a very suitable mathematical Microgrids With PV Prosumers: A Stackelberg Game Approach," IEEE
framework for describing the strategic behavior of prosumers. Transactions on Industrial Informatics, vol. 13, no. 3, pp. 1088-1098, June
Numerical analysis of price-taker and Nash-Cournot models 2017, doi: 10.1109/TII.2017.2654302.
reveal the consequences of the potential transition towards [17] K. Anoh, S. Maharjan, A. Ikpehai, Y. Zhang and B. Adebisi, "Energy
prosumer-centric LEMs. If market power is allocated to Peer-to-Peer Trading in Virtual Microgrids in Smart Grids: A Game-Theoretic
Approach," IEEE Transactions on Smart Grid, vol. 11, no. 2, pp. 1264-1275,
prosumers without proper regulation the electricity prices could

Authorized licensed use limited to: UNIVERSITY COLLEGE CORK. Downloaded on December 10,2024 at [Link] UTC from IEEE Xplore. Restrictions apply.
March 2020, doi: 10.1109/TSG.2019.2934830. Coalitions With Energy Management Using Cooperative Game Theory,"
[18] J. Lee, J. Guo, J. K. Choi and M. Zukerman, "Distributed Energy Trading IEEE Transactions on Power Systems, vol. 34, no. 1, pp. 303-313, Jan. 2019,
in Microgrids: A Game-Theoretic Model and Its Equilibrium Analysis," IEEE doi: 10.1109/TPWRS.2018.2858540.
Transactions on Industrial Electronics, vol. 62, no. 6, pp. 3524-3533, June [25] Y. Ye, D. Qiu, J. Li and G. Strbac, "Multi-Period and Multi-Spatial
2015, doi: 10.1109/TIE.2014.2387340. Equilibrium Analysis in Imperfect Electricity Markets: A Novel Multi-Agent
[19] A. Ghosh, V. Aggarwal and H. Wan, "Strategic Prosumers: How to Set Deep Reinforcement Learning Approach," IEEE Access, vol. 7, pp. 130515-
the Prices in a Tiered Market?,", IEEE Transactions on Industrial Informatics, 130529, 2019, doi: 10.1109/ACCESS.2019.2940005.
vol. 15, no. 8, pp. 4469-4480, Aug. 2019, doi: 10.1109/TII.2018.2889301. [26] H. Singh, "Introduction to game theory and its application in electric
[20] E. Baeyens, E. Y. Bitar, P. P. Khargonekar and K. Poolla, "Coalitional power markets," IEEE Computer Applications in Power, vol. 12, no. 4, pp. 18-
Aggregation of Wind Power," IEEE Transactions on Power Systems, vol. 28, 20, Oct. 1999, doi: 10.1109/67.795133.
no. 4, pp. 3774-3784, Nov. 2013, doi: 10.1109/TPWRS.2013.2262502. [27] J. Contreras, M. Klusch and J. B. Krawczyk, "Numerical solutions to
[21] W. Saad, Z. Han, H. V. Poor and T. Basar, "Game-Theoretic Methods for Nash-Cournot equilibria in coupled constraint electricity markets," IEEE
the Smart Grid: An Overview of Microgrid Systems, Demand-Side Transactions on Power Systems, vol. 19, no. 1, pp. 195-206, Feb. 2004, doi:
Management, and Smart Grid Communications," IEEE Signal Processing 10.1109/TPWRS.2003.820692.
Magazine, vol. 29, no. 5, pp. 86-105, Sept. 2012, doi: [28] J.B. Krawczyk, and S. Uryasev, “Relaxation algorithms to find Nash
10.1109/MSP.2012.2186410. equilibria with economic applications”. Environmental Modeling &
[22] W. Lee, L. Xiang, R. Schober and V. W. S. Wong, "Direct Electricity Assessment vol. 5, 63–73, 2000, [Link]
Trading in Smart Grid: A Coalitional Game Analysis," IEEE Journal on [29] N. Zhang, Y. Yan, and W. Su, “A game-theoretic economic operation of
Selected Areas in Communications, vol. 32, no. 7, pp. 1398-1411, July 2014, residential distribution system with high participation of distributed electricity
doi: 10.1109/JSAC.2014.2332112. prosumers”, Applied Energy, vol 154, pp. 471-479, 2015,
[23] L. Han, T. Morstyn and M. McCulloch, "Constructing Prosumer [Link]
Coalitions for Energy Cost Savings Using Cooperative Game Theory," in [30] IEEE “Power Systems Test Benchmarks,” DOI:
Proc. 2018 Power Systems Computation Conference (PSCC), pp. 1-7, doi: 10.5281/zenodo.4030695. [Online]. Available:
10.23919/PSCC.2018.8443054. [Link] IEEE33Bus
[24] L. Han, T. Morstyn and M. McCulloch, "Incentivizing Prosumer

Authorized licensed use limited to: UNIVERSITY COLLEGE CORK. Downloaded on December 10,2024 at [Link] UTC from IEEE Xplore. Restrictions apply.