Energy 313 (2024) 133715

Contents lists available at ScienceDirect

Energy
journal homepage: www.elsevier.com/locate/energy

Game-theoretic optimization strategy for maximizing profits to both

end-users and suppliers in building rooftop PV-based microgrids
Jianing Luo a , Karthik Panchabikesan b, Kee-hung Lai c , Timothy O. Olawumi d,
Modupe Cecilia Mewomo e, Zhengxuan Liu f,*
a
School of Mechanical Engineering, Southwest Jiaotong University, Chengdu, China
b
Florida Solar Energy Center, Cocoa, FL, USA
c
Faculty of Business, The Hong Kong Polytechnic University, Hong Kong, China
d
School of Computing, Engineering and Built Environment, Edinburgh Napier University, Scotland, United Kingdom
e
Department of Construction Management and Quantity Surveying, Faculty of Engineering and the Built Environment, Durban University of Technology, Durban, South
Africa
f
Faculty of Architecture and the Built Environment, Delft University of Technology, Julianalaan 134, 2628 BL, Delft, Netherlands

A R T I C L E I N F O A B S T R A C T

Handling editor: Isabel Soares Rooftop photovoltaic (PV) with battery storage offers a promising avenue for enhancing renewable energy
integration in buildings. Creating microgrids with backup power from closely spaced solar buildings is widely
Keywords: recognized as an effective strategy. Nevertheless, a notable gap exists between the preferences and priorities of
Rooftop photovoltaic (PV) systems electricity consumers residing in these solar-powered buildings and the interests of microgrid investors. The
Microgrid
electricity consumers focus on decreasing the levelized cost of energy, while the microgrid investors focuses on
Game theory
achieving high net profit. This study proposes a novel game theory-based microgrid optimal design approach for
Renewable power generation
Solar building designing power generations of the microgrid system and PV installation with battery storage on the building
Economics roofs, considering the different requirements and interests of electricity consumers and microgrid investors. The
Bi-level optimization design optimization is framed around the Nash Equilibrium of the Stackelberg game, incorporating a bi-level
optimization cycle that addresses the conflict and cooperation of electricity consumers and microgrid in-
vestors. A win-win situation can be yielded using the developed optimal design approach compared to con-
ventional optimal design approaches. The results demonstrate a significant improvement, with the microgrid
power generation yielding a large net profit (up to 0.08 USD/kWh) and concurrently reducing the levelized cost
of energy by approximately 14 %.

Nomenclature (continued )
Abbreviations
Abbreviations
Etot Total electricity consumption (kWh)
COP Coefficient of Performance Pt Power purchased at time step t
DR Demand Response Pele,t Electrical load at time step t
LCOE Levelized Cost of Energy Pr Electricity unit price (USD/kWh)
PV Photovoltaic ProfitMG Profit of aggregator
UG Utility grid Pren,t Renewable power generation at time step t
Notations PL Partial load
APV PV areas (m2) PPV Power generated from solar radiation (kW)
Cini Initial cost (USD) Rad Solar radiation per area (kW/m2)
Cmai,sbs Maintenance cost (USD) Δt Timestep
Copt Operating cost (USD) Tamb Ambient temperature (◦ C)
Cap Capacity (kW) UPbpg Unit price of the backup power generator (USD/kW)
(continued on next column) (continued on next page)

* Corresponding author.
E-mail address: [email protected] (Z. Liu).

https://doi.org/10.1016/j.energy.2024.133715
Received 15 June 2024; Received in revised form 12 October 2024; Accepted 3 November 2024
Available online 4 November 2024
0360-5442/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
J. Luo et al. Energy 313 (2024) 133715

(continued ) (maximum up to $ 0.1664 million) under 30 % demand response

Abbreviations participation. As for the application of the system flexibility, Swami-
Tref PV panels reference temperatures (◦ C)
nathan et al. [16] optimized the capacities of the renewable power
Yk Total years of its life cycle generator considering the impacts of the building energy flexibility,
Subscripts where savings of initial cost and operational costs can be further real-
amb Ambient ized. Tomin et al. [17] proposed a microgrid design approach consid-
bpg Backup power generation
ering the impacts of flexible renewable power generation, achieving a
bat Battery storage
ec Electricity consumers potential reduction of up to 40 % in the levelized cost of energy (LCOE).
MG Microgrid The literature shows that the developed optimal design approaches
PV Photovoltaics can effectively provide environment-friendly and cost-effective solu-
rpc Renewable power generation components tions for microgrid investors. However, a notable gap exists as these
Greek letters
studies commonly assume alignment between the demands and interests
ηbpg Backup power generator efficiency
ηPV Overall efficiency of PV panels
of electricity consumers and the overall microgrid system, which di-
εope Power generation efficiency verges from the actual situation. In practice, from the electricity con-
sumers’ perspective, electricity consumers solely focus on having a
reliable power supply with relatively lower electricity prices [18].
1. Introduction However, to our best knowledge, only a few studies have optimized the
microgrid system considering the different requirements and interests of
China has set a commendable goal of achieving carbon neutrality microgrid investors and electricity consumers, respectively.
before 2060, and to confront the pressing issue of carbon emissions, a Addressing these challenges requires a solution that acknowledges
major focus is placed on power generation. Notably, 70 % of China’s the differing needs of microgrid stakeholders, and game theory emerges
power generation is attributed to conventional thermal power plants, as a promising tool to achieve this [19]. Game theory provides a theo-
making it a primary contributor to carbon emissions [1]. Tackling this retical framework for conceiving social situations among competing
challenge necessitates urgent attention to two crucial scientific prob- players and delves into mathematical models of conflict and cooperation
lems: augmenting the penetration of renewable energy in power gen- between different decision-makers [20]. Currently, game theory has
eration and ensuring a reliable power supply [2]. These challenges have been adopted to conduct energy trading in the smart grid system con-
garnered substantial responses from both academic and industrial sec- cerning the grid-building interactions in Refs. [21–27]. For instance,
tors, signifying a collective commitment to finding viable solutions [3, Nwulu and Xia [28] proposed a game-theoretic energy-economic model
4]. to decrease the fuel costs for the grid side and offer an incentive to the
As the prospective electrical system, a microgrid consists of distrib- electricity customers for the demand side to offset the impact of power
uted power generators, energy storage systems, and interconnected supply interruption. Dabush et al. [22] developed a game-theoretic
loads [5]. This innovative system holds significant promise in advancing model to examine the viability of installing PV systems on rooftops of
renewable energy penetration for electricity generation, fostering eco- affordable housing buildings, where the economic benefits of the
nomic benefits, and enhancing system reliability simultaneously [6]. building electricity consumers and the utility grid are quantified. While
Alfergani et al. [7] employed multi-objective optimization to enhance many existing game theory-based energy system design works concen-
the power capacity of the microgrid. Zhang et al. [8] has developed a trate primarily on proposing design solutions to decrease the operation
hybrid robust-stochastic multi-objective optimization approach for cost and/or proposing the operation schedule to achieve peak shifting
integrating cooling, heating, hydrogen, and power-based microgrids, and enhance the reliability of the power grids. Few studies began to
achieving a potential reduction in operational costs by up to 15.44 %. focus on the issue of the profit distribution for the power generation
Lou et al. [9] developed an optimization approach for the microgrid investors and electricity consumers when distributed renewable power
system based on power load analysis considering the impact of micro- generation is used in the energy system. This critical aspect represents
grid locations on system performance. Due to the abundant solar energy an evolving area of research that merits further exploration and atten-
resources in Lhasa, this approach allowed for a reduction in the area tion in the broader landscape of game theory applications in the energy
required for solar thermal collectors by more than 50 % compared to sector.
Xi’an. The proper optimal design of microgrids proves instrumental in In the future energy systems, an increasing number of electricity
realizing these objectives by leveraging presumed/quantified power customers are expected to adopt distributed power generators, such as
generations and electrical loads. The optimization of microgrid power PV systems on building roofs [29]. This shift will usher in a new mode
generators and battery storage capacities serves to mitigate carbon where the microgrid’s power generation and the utility grid’s power
emissions from power generation and decrease the operation costs. This supply collaborate to ensure a reliable power source. The effect of
optimization process integrates a typical annual load profile and optimizing distributed power generation capacity for electricity con-
renewable power generation based on typical annual weather data [10, sumers on the microgrid backup power generation capacity, crucial for
11]. Jung and Villaran [12] introduced an innovative optimal design microgrid investors, cannot be overlooked. It is imperative to address
method that utilizes a typical daily power-load profile derived by the optimal design concerning their different requirements and interests
condensing the annual load and power generation data into a 24-h of the power generation investors and electricity consumers. Developing
format. This approach enhances economic benefits and minimizes the reasonable design schemes has become critical, necessitating a prior
computational complexity associated with the design process. understanding of the diverse needs of multiple stakeholders.
To further unlock the potential of microgrids in terms of system To address these research gaps, this study proposes a game theory-
economics and environmental friendliness, new technologies, such as based microgrid optimal design approach, aiming to effectively
demand response (DR) and system flexibility, are actively incorporated consider the practical optimization objectives from two different roles (i.
into microgrid optimal design [13]. For instance, Tsao et al. [14] e., electricity consumers and microgrid investors). The main objective of
developed a microgrid optimal design approach by leveraging block- this study is to effectively consider the multiple requirements of
chain technology to provide demand response programs, where the different stakeholders (i.e., investors of the microgrid power system with
profitability of the designed microgrid was increased up to 1.68 %. equipped backup power generation and demand-side electricity con-
Gamil et al. [15] developed a microgrid optimal design approach sumers with the voluntary installation of renewable energy generation)
considering different percentages of demand response participation. The in the microgrid system, where the former focus on achieving high net
results showed that the operation cost can effectively be reduced profit and the latter aims to decrease the LCOE and ensure a reliable

2
J. Luo et al. Energy 313 (2024) 133715

power supply. The main contributions and innovations of this study are 2.2. Requirements and interests of electricity consumers
outlined as follows:
Electricity consumers require a stable power supply and decreased
• A bi-level design optimization is developed based on the Nash LCOE. The LCOE, as a quantitative electricity utilization indicator, is
Equilibrium of the Stackelberg game concerning the conflict and used to calculate the power utilization cost of traditional energy projects
cooperation of electricity consumers and microgrid investors. such as thermal power, hydropower, and gas power, and later expanded
• Diverse requirements and interests of electricity consumers and to the renewable energy field. It is quantified as shown in Eq. (1), which
microgrid investors are methodically quantified and considered consists of the initial cost (Cini,ec ), the operating cost (Copt,ec ), and the
throughout the optimal design process. maintenance cost (Cmai,ec ) and Etot , which is the total of electricity
• A design solution is proposed to create a win-win scenario by consumption.
harmonizing the requirements and interests of both electricity con- The initial cost as shown in Eq. (2) is the average annual initial cost
sumers and microgrid investors within the optimal design that electricity consumers must pay for renewable power generators,
framework. battery storage, and renewable power components. The initial costs of
the renewable power generation components as shown in Eq. (3) include
The proposed innovative game theory-based microgrid optimal the hardware cost of the system controller and soft cost (e.g., engi-
design approach contributes strategically to navigating the intricate neering, construction, commissioning, and regulatory costs), as well as
landscape of stakeholder needs, effectively accommodating the diverse the additional electric infrastructure costs [30,31]. Where, CPV,ec , and
objectives of electricity consumers and microgrid investors in the CBat,ec are the initial cost of PV panels and battery storage. Yk represents
microgrid system. the total years of its life cycle, and PPV represents the power generation
from PV panels.
2. Structure of microgrid systems and their requirements and The operational cost as shown in Eq. (4) is the sum of electricity
interests purchased from the microgrid system and the utility grid. Where, PtMG is
the power purchased from the microgrid, and PtUG is the power pur-
2.1. Typical microgrid system chased from the utility grid. PrMG and PrUG are the electricity unit price
of the microgrid and utility grid, respectively.
Due to the limited impact of individual building on the utility grid, a The annual maintenance cost is assumed to be 1 % of the total
microgrid system is developed as an aggregator within a specific area average annual initial cost as shown in Eq. (5) based on [32], which
encompassing closely spaced spatially distributed electricity consumers. investors have to pay per year to provide maintenance for the major
This system, depicted in Fig. 1, serves to ensure a reliable power supply facilities of the system per year.
by incorporating backup power generators. In the shown setup, the ( )/
adopted microgrid system equipped with backup power generation is LCOEec = Cini,ec + Copt,ec + Cmai,ec Etot (1)
proposed to cope with the risk of power outage due to the uncertainties ( ) /
of the dynamic electrical load and the intermittent and uncontrollable Cini,sbs = CPV,ec + CBat,ec × 1 Yk + Crpc,ec (2)
renewable power generations. As the major electricity consumers on the
demand side, the buildings are anticipated to install PV modules Crpc,ec = max(PPV ) × 50 + 1900 (3)
equipped with the battery storage system to decrease the burden of the ( ) ( )
utility grid and the electricity consumption operation cost. Copt,ec =
∑8760
PtMG × Δt × PrMG +
∑8760
PtUG × Δt × PrUG (4)
t=1 t=1

∑( / )
Cmai,ec = 1% × Cfac,k × 1 Yk (5)

2.3. Requirements and interests of the microgrid investors

Regarding investor requirements, the microgrid needs to ensure the

power supply to the electricity consumers and decrease electricity pur-
chases from the utility grid. Microgrid investors primarily focus on
maximizing their net profit derived from microgrid power generation.
This net profit is calculated by deducting the relevant power generation
costs from the revenue generated by selling electricity to consumers, as
quantified in Eq. (6) [19]. The power generation costs encompass the
initial costs of backup power generators and the operation cost of power
generation, as quantified in Eq. (7) and in Eq. (8) respectively.
( )
∑8760
ProfitMG = t=1
P t
MG × Δt × PrMG − Cini,MG − Cbpg,MG (6)

/
Cini,MG = CMG,bpg × 1 Ybpg (7)
( )
∑8760 ( / )
Cbpg,MG = t=1
PtMG ηbpg × Δt (8)

Fig. 1. Components of a simple microgrid system.

3
J. Luo et al. Energy 313 (2024) 133715

3. Game theory-based optimal design of the microgrid-building 3.2. Basic scheme for identification of nash equilibrium
system
3.2.1. Nash equilibrium of leader – optimization of microgrid power
3.1. Establishment of the stackelberg game between the electricity generator capacity
consumers and microgrid investors Fig. 3 outlines detailed steps of Optimization A, specifically the
method for designing the optimal microgrid power generator capacity.
In the game theory approach, each player aims to maximize their The capacity of the microgrid power generator is optimized within its
own welfare in a game by adopting effective measures, even though designated search range, guided by the power generation requirements.
there may be partial conflicts among players. In this established Stack- The power generation system is directed based on these requirements.
elberg game, three key elements are included (i.e., ESG = {σ,S,U}), i.e., Concerning the dynamic efficiency of the power generation, the opera-
players, their design solutions, and their benefits. Each player i (i ∈ σ ) tion cost for the power generation is calculated in the microgrid power
determines their design solution si (si ∈ S) to maximize the benefit ui generation model. These factors are incorporated into the objective
(ui ∈ U). The final solution is obtained when the game is in the Stack- function to calculate the net profit of the microgrid. Then, the obtained
elberg Nash Equilibrium, which is shown in Eq. (9). The obtained design net profit of the microgrid power generation is used in the optimizer to
solution s* is a Nash equilibrium when each player cannot increase one’s judge whether the design solution is the best choice. The optimization
own expected payoff by changing the design solutions while the other process will be repeated until the optimal design solution (i.e., microgrid
players keep theirs unchanged [33]. power generator capacity) is found.
( )
U(s* ) ≥ U si , s*− i (9) 3.2.2. Nash equilibrium of followers – optimization of PV areas and battery
Fig. 2 illustrates a schematic diagram of the proposed game-theoretic storage capacity
optimal design for microgrid power generation and solar building sys- Fig. 4 shows detailed steps of Optimization B (mentioned in Fig. 2),
tems. In this game-theoretic optimization, two different requirements that is, the optimal design method of PV areas and battery storage ca-
for the two different interested parties are considered. This first party is pacity for the solar building system. The PV area and the battery storage
microgrid power generation system, where the objective is to maximize capacity are determined via optimization within their search ranges.
the net profit of the microgrid. The second party is solar building system, The weather data is fed into the solar building model, and the electrical
where they prefer to decrease the levelized cost of the energy. Due to loads are quantified according to the thermal comfort requirements and
this, it consists of two optimizations, i.e., microgrid power generation usage profiles. The power generation from the PV modules is quantified
optimization and solar building system optimization. The former, acting based on the design PV area and weather data. Considering the power
as the leader, aims to design the power generator capacity to pursue the balance, the electricity purchase from the microgrid and utility grid is
maximum net profits as one type of power supply. It should be used in calculated according to electrical loads and renewable power genera-
the second optimization. The latter, functioning as a follower, aims to tions. Based on the power generation capacity of the microgrid, the
design the PV area and battery storage capacity to decrease the LCOE electricity purchased from the microgrid’s power generation system and
according to the capacity of the microgrid power generators. Subse- utility grid are quantified, respectively. This partial cost, as operation
quently, based on the optimal design solution from the latter optimi- cost, is considered in the objective function. In the end, the optimizer
zation, the power generation requirements are communicated to the aims to minimize the LCOE, and the iterative optimization process will
former optimization to calculate the real net profits. be repeated until the optimal design solutions (i.e., PV areas and battery
As for the difference of the mentioned two optimizations, the Opti- storage capacity) are determined.
mization A is a Nash equilibrium of leads and the Optimization B is a
Nash equilibrium of followers. In other words, the system optimization
3.3. Objective functions and their constraints in the game theory-based
is begun in the Optimization A and the optimal design for the solar
optimizations
building system in Optimization B is conducted based on the optimiza-
tion results of the Optimization A. Then, the optimization results are
According to the above optimal design methods of the solar building
tested in Optimization A whether the optimal design in optimization A
system and microgrid power generation system in this Stackelberg
can be regarded as the global optimal results. If not, the optimal process
game, the objective functions aligned with their interests (as stated in
is repeated. This method can effectively avoid the local optimizations of
Section 2.2 and Section 2.3) are formulated respectively, which are
two players and a win-win situation for two players with different re-
shown in Eq. (10) and Eq. (11) respectively. Where, APV and Capbat are
quirements can be achieved.
the PV area and capacity of the battery storage. Constraints are set
within the lower and upper limits according to practical limits in the
optimization processes, as shown in Eqs. 12–14.

Fig. 2. Schematic diagram of proposed game-theoretic optimal design for the Fig. 3. Optimal design method of microgrid power generator capacity (Opti-
microgrid power generation and solar building systems. mization A).

4
J. Luo et al. Energy 313 (2024) 133715

electricity price of the utility grid is used to promote power generation

from the microgrid system and can also effectively decrease the utility
grid’s burden simultaneously.
As for the microgrid generation system, the unit price of the backup
power generator (UPbpg ) decreases with the capacity increase. It has
been calculated as shown in Eq. (15) based on the capacities (Capbap )
referring to Ref. [37].
( )
UPgas = 3711.78 − 280.47 × ln Capgas (15)

4.3. Development of the power generation models

Fig. 4. Optimal design method of PV areas and battery storage capacity for the 4.3.1. Rooftop PV generation model for the homestay hotel buildings
solar building system (Optimization B). PV model: The PV generation model is developed to quantify the
power generation from the PV panels (PPV ), according to Daud and
( ) Ismail [38]. It is calculated as below:
max ObjMG = FMG Capbpg (10) ( ( ))
PPV = Rad × APV × 1 + KPV TPV − Tref × ηpv (22)
min Objsbs = Fsbs (APV , Capbat ) (11)
TPV = Tamb + 0.0256 × Rad (23)
Capbpg,min ≤ Capbpg ≤ Capbpg,max (12)
0 ≤ PPV,t ≤ PMax
PV , ∀t ∈ [1, 8760] (24)
APV,min ≤ APV ≤ APV,max (13)
where, solar power generation considers the solar radiation (Rad), PV
Capbat,min ≤ Capbat ≤ Capbat,max (14) areas (APV ), the overall efficiency of PV panels (ηpv ), PV cell temperature
(TPV ), and ambient temperature (Tamb ). The PV cell temperature is
4. Basic information and model developments of proposed determined by solar radiation and ambient temperature, as shown in Eq.
systems (23). KPV is set as − 3.7 × 10− 3, and the reference temperature Tref is
25 ◦ C referring to Refs. [6,35]. Besides, the overall efficiency of PV
4.1. Basic information about the microgrid system panels is 11–25 % [39]. This work uses a constant value (20 %) to
simplify the calculation.
To verify the performance of the proposed game theory-based Battery storage model: Battery storage is widely used to overcome
optimal design, a microgrid comprising four hotel buildings equipped uncertainties and intermittent renewable power generation. The overall
with PV and battery storage is selected as the case study. These four charge and discharge efficiencies are set as 0.85 based on [40]. Besides,
buildings are homestay hotels with three stories, and the PV panels can the maximum charging rate (Rbatch,max ) and discharging rate (Rbatdch,max )
be equipped on the roofs of the four buildings. The ceiling height of each are set as 20 % of the battery capacity and 50 % of the battery capacity,
building is 2.5 m, and the area per floor is about 210 m2 with dimensions respectively, in the constraints shown in Eqs. (19) and (20) respectively.
of 15 m × 14 m. Their building envelopes are similar; the details are To ensure the battery storage works within a safety range, the minimum
shown in Table 1. (Capbat,min ) and maximum (Capbat,max ) limits of the battery capacity are
The microgrid is located in Sanya, Hainan Province, China, and the set as 20 % and 80 %, respectively (as shown in Eq. (21)).
exact location (a red star) is shown in Fig. 5. In this location, abundant
0 ≤ Rbatch ≤ Rbatch,max (19)
solar resources can be exploited for power generation to provide a sus-
tainable power supply, where the annual average solar radiation is 600 0 ≤ Rbatdch ≤ Rbatdch,max (20)
W/m2, and the maximum value is about 3600 W/m2.
Capbat,min ≤ Capbat ≤ Capbat,max (21)
4.2. Cost data of the solar homestay hotel buildings and microgrid power
generation system 4.3.2. Power generation model for the microgrid systems
The selected power generators of the microgrid power generation
Table 2 shows the detailed cost data of the solar homestay hotel system are driven by natural gas since natural gas is a clean-burning,
buildings. The renewable power generation system involves two key efficient fuel [41]. The operation power generation efficiency of the
parameters (i.e., PV panes and battery storage). Their initial cost and gas generator (εope,gas ) is quantified based on the standard power gen-
lifetime are considered since they significantly impact the overall cost of eration efficiency (εstd,gas ) and its operational partial load ratio (PLgas ), as
the solar homestay hotel buildings. As for the electricity prices, they are shown in Eq. (22) referring to Refs. [42,43]. The standard power gen-
different based on the different electricity retailers. The electricity price eration efficiency is sensitive to the change of capacity, calculated by Eq.
of the utility grid is determined according to the current price [35,36]. (23) referring to Refs. [44,45]. The partial load ratio is calculated in Eq.
Since there is no electricity price for the microgrid system, 90 % of the (24). Where, elegas is the requirement of the electricity generation for
power generation.
Table 1 ( ( )2 )
Main parameters of the considered four buildings [34]. εope,gas = εstd,gas × 0.715 + 0.478 × PLgas − 0.190 × PLgas (22a)

Name Parameter Value ( ( ) )

2
εstd,gas = 0.01 × 4.236 × ln Capgas + 70.3 (23a)
Dimmable lighting Lighting density 10 W/m
Other electrical equipment Equipment load density 15 W/m2 /
Fresh air Ventilation Ventilation rate 15 L/h × person− 1 PLgas = elegas Capgas (24a)
Cooling load Load density 153 W/m2
Infiltration rate – 0.2 air changes per hour

5
J. Luo et al. Energy 313 (2024) 133715

Fig. 5. Overview of the microgrid system location.

4.5. Electricity usage priority and control mechanism

Table 2
Cost data of the solar homestay hotel buildings.
To decrease the operation cost of the electricity consumers and in-
Name Items Value crease the utilization of renewable power generation, the priority of the
Renewable power generation system Initial cost of PV 288 $/m2 power supply is listed in Table 3. The renewable power generation is
panels used as the first option. It can significantly lower the rate of renewable
Lifetime of PV 20 years
energy generation that goes to waste. Battery storage can store the
panels
Initial cost of 255.6 $/kWh
redundant renewable generated power and used as the second opinion.
battery The mentioned first two power supplies are free and they can be
Lifetime cost of 15 years prioritized while the other two are used to ensure the reliability of the
battery power supply. Thus, the control mechanism can effectively ensure a
Name Electricity retailer Value reliable power supply to the electricity consumers and achieve the
Unit price of electricity from different Utility grid 0.2 $/kWh economic operation of the system. Two typical operational modes (i.e.,
electricity retailers Microgrid system 0.2 × 90 % islanded mode and grid-connected mode) of the microgrid system are
$/kWh considered and adopted in the control mechanism.
Fig. 6 illustrates the detailed control mechanism of operational
modes selection for the microgrid energy system. The islanded mode is
4.4. Energy simulation
adopted when the microgrid can adequately fulfil the demand. In other
words, if the power output from the first three power supply options is
The electrical loads of the solar homestay hotel buildings consist of
sufficient to meet the demand, the operational mode defaults to the
the cooling, lighting, and plug loads, considering their location (i.e.,
islanded mode. Conversely, when the demand is higher than the power
typical tropical region). They are simulated via TRNSYS based on the
output of the microgrid, the grid-connected mode is adopted. In this
pre-set values of the load densities. TRNSYS is a modular-based simu-
mode, the utility grid serves as an additional power supply to meet the
lation software specifically designed for modelling and optimizing en-
demand and ensure system’s reliability.
ergy systems. The load densities, including the lighting density,
equipment load density, and cooling load density, are fed into the “TYPE
56” of the TRNSYS. Type 56 stands out for its capability in multi-zone
building energy analysis. Type 56 allows for precise simulation of
solar radiation and heat transfer through windows of complex building
geometries and configurations. it also supports a wide range of heating,
cooling, and ventilation systems, providing detailed outputs on system Table 3
variables and energy consumption. Then, according to the weather Priority of the power supply.
conditions, the electrical load profiles of these buildings can be calcu-
Option Power supply type
lated. As for the coefficient of performance (COP) of the chillers, a
constant value is set as four in the cooling load calculation for these four Option 1 Renewable power generation
Option 2 Battery storage discharging
solar homestay hotel buildings.
Option 3 Microgrid backup power generation
Option 4 Supports of the utility grid

6
J. Luo et al. Energy 313 (2024) 133715

5.2. Results of optimal design solutions for microgrid and solar hotel
building systems

5.2.1. Game-theory-based optimal design solutions for microgrid and solar

hotel building systems
Fig. 7 shows the results of the game theory-based optimal design
under the different capacities of the microgrid power generators. The
green curve is according to the game theory-based optimization curve. It
presents the net profits of microgrid investors when the PV area and the
battery storage are optimized to obtain the minimum LCOE of electricity
consumers under designed microgrid power generator capacities. Be-
sides, the net profit under different microgrid power generator capac-
ities consists of two boundaries (i.e., upper and lower boundaries of the
microgrid net profits) since the different portfolios of the PV areas and
battery storages are involved. The largest difference between the upper
and lower boundaries is up to 90.9 % of the largest net profits. Thus, the
portfolios of the PV area and battery storage for the electricity con-
sumers have significant impacts on achieving large net profits. They
have to be considered in the microgrid power generation system design.
The green line is inconsistent with the upper boundary of the
microgrid net profits, which means the interests of the microgrid in-
Fig. 6. Control mechanism of operational modes selection for the microgrid vestors and electricity consumers are inconsistent. The green line is also
energy system. inconsistent with the lower boundary of the microgrid net profits, which
means the interests of the microgrid investors and electricity consumers
5. Results and discussions are not completely opposed. The maximum value appears at the red
point, representing the optimized design microgrid power generator
5.1. Descriptions of the reference for microgrid optimal design methods capacity (i.e., 70 kW). The largest net profits can be achieved when this
capacity of the microgrid power generator is selected, considering the
Microgrid optimal design methods to face the different requirements impacts of different portfolios of the PV area and battery storage.
and interests of microgrid investors and electricity consumers have been The net profits under determined different capacities of the power
investigated in academia and industry. According to the different orders generator are not always positive. It means that the economic risk due to
of the optimizations, three conventional microgrid optimal design increasing the capacity of the power generator exists, and the net profit
methods are presented, that is, Supply-to-Demand Optimal Design is negative. The reasons include the high initial costs and maintenance
method, Demand-to-Supply Optimal Design method, and Decentralized costs of the power generators and the relatively low efficiency of the
Optimal Design method. Detailed information on these three optimal power generation. The negative net profit may appear above the 110 kW
design methods is introduced below. capacity of the power generators or 180 kW capacity of the power
generators, considering different portfolios of the PV area and battery
• Supply-to-Demand Optimal Design method: In this approach, the storage.
capacity of the microgrid power generators is initially optimized for Then, according to the determined capacity of the microgrid power
the microgrid investors by using the electrical load profile without generators (i.e., 70 kW), the PV area and battery storage capacity are
consideration of the renewable power generation impacts. Then, optimized to decrease the overall cost of electricity consumers, as shown
according to the determined microgrid power generator capacity and in a 3-D graph (Fig. 8). The impacts of the combination between PV
different electricity prices of the different power generations, the PV areas and battery storage capacity on the economics are presented. The
panels’ area and the battery storage capacity are optimized for overall cost can increase sharply with the PV area increasing under
electricity consumers. determined battery storage capacity. Increasing the battery storage ca-
• Demand-to-Supply Optimal Design method: This method con- pacity solely impacts the overall cost when the PV area is constant. The
ducted the design optimization by using the opposite optimization minimum overall cost is obtained considering the portfolios of these two
direction compared with the Supply-to-Demand Optimal Design
method. The area of the PV panels and the battery capacity are
optimized firstly for electricity consumers. Then, the capacity of the
microgrid power generator is optimized for microgrid investors ac-
cording to the electrical load profile with consideration of the
renewable power generation impacts.
• Decentralized Optimal Design method: This optimal design
method conducted the design optimizations of the power generator
capacity for microgrid investors, the PV panels’ area, and the ca-
pacity of the battery storage for electricity consumers, respectively.
In the process of optimization, the impacts of the microgrid power
generator capacity on the optimization of the area of the PV panels
and the capacity of the battery storage are ignored. Besides, the
impacts of the determined area of the PV panels and the battery
storage capacity on optimizing the microgrid power generator ca-
pacity are also ignored.

Fig. 7. Results of the game-theory-based optimal design under the different

capacities of the microgrid power generators.

7
J. Luo et al. Energy 313 (2024) 133715

Fig. 9. Net profit of the microgrid power generation per unit among these
four cases.

Optimization Case and Decentralized optimization Case. The net profits

of the microgrid power generation per unit are both negative.

Fig. 8. Optimal design results of the PV area and battery storage capacity under
5.3.2. LCOE analysis from electricity consumers’ perspective
determined microgrid power generator capacity.
Decreasing LCOE is the major objective for electricity consumers.
Fig. 10 shows the LCOE of the electricity consumers and the comparison
renewable power generation facilities (i.e., PV area as 680 m2 and bat- results of the LCOE among these four cases. The PV installation of
tery storage capacity as 75 kWh). building roofs for electricity consumers effectively reduces the LCOE (at
least up to 12.7 %). The design solution in the proposed Game-Theory-
5.2.2. Design solutions among the different optimization cases Based Optimization Case is regarded as the best adoption considering
Table 4 shows the design solutions among the different optimization the interests of the microgrid investors and the interests of electricity
cases. The design solution of the Game-Theory-Based Optimization Case consumers since the reduction of the LCOE is about 14 %. Moreover,
is obtained, and the other three design solutions are obtained based on although the lower values of the LCOE appear in other cases, the net
the introduced optimization cases in Section 5.1. profit of the microgrid power (as shown in Fig. 9) is negative. It means
that the optimization case cannot be accepts for power suppliers in the
5.3. Economic analysis concerning the different optimal design methods microgrid power generation system. The different interests and re-
quirements for power suppliers and consumers should be considered
5.3.1. Net profit analysis from microgrid investors’ perspective together. The detailed reasons are further analyzed below.
Increasing the net profit is the major objective for microgrid in- As for the Demand-to-supply Optimization Case, the PV area and
vestors. Fig. 9 shows comparison results for the net profit of the battery storage are optimized before optimizing the microgrid power
microgrid power generation per unit. If the net profit value is positive, generator capacity. The optimal design solution of the microgrid power
power generation can obtain the cost revenue from power generation. generator capacity is smaller due to the consideration of the renewable
The proposed Game-Theory-Based Optimization Case has the best per- power generation impacts. The amount of electricity from the microgrid
formance among these four optimization cases. In this case, the net power generation system decreases due to the selected smaller capacity
profit of the microgrid power generation is up to 0.08 USD/kWh. As for of the microgrid power generator. Thus, the LCOE cannot decrease at the
the Demand-to-Supply Optimization Case, the optimal design on the lowest value in this optimization case. As for the Supply-to-Demand
demand side is considered before the power generation optimization of Optimization Case and Decentralized Optimization Case, the larger ca-
the supply side. However, without considering the interaction between pacity of the microgrid power generator is selected. However, the net
the supply and demand sides, and even under giving up the dominance profit of the microgrid power generation is negative, which cannot be
of the supply side for microgrid power generation, the net profit of the adopted by the microgrid investors in practice.
microgrid power generation in the Demand-to-Supply Optimization
Case only reaches a quarter of the power generation net profit per unit
5.4. Energy efficiency analysis concerning the different optimal design
compared to the Game-Theory-Based Optimization Case. Suppose the
methods
impacts of the PV panels and battery storage installations from the
electricity consumers on the optimal design of the microgrid power
From the microgrid investors’ perspective, high power generation
generation capacity are ignored. In that case, the optimization design
efficiency can effectively decrease the operation cost for power gener-
results cannot achieve the expected results and even cause high addi-
ation and increase the net profit of the microgrid power generation.
tional expenses in power generation such as Supply-to-Demand
Fig. 11 shows the annual power generation efficiency profile under
dynamic electricity consumption among these four cases. The selected
Table 4
larger capacity of the power generator in the Supply-to-Demand Opti-
Design solutions among the different optimization cases.
mization Case (Fig. 11b) and Decentralized Optimization case (Fig. 11d)
Case Natural gas generator PV area Battery can achieve higher efficiency during summer (from 2150 h to 6570 h).
capacity (kW) (m2) storage (kWh)
However, during the winter and some periods with relatively lower
Game-Theory-Based 70 680 75 electricity consumption, the power generation efficiencies (mostly
Optimization Case
below 75 %) in these two cases are lower than the other two cases with
Supply-to-Demand 100 667 67
Optimization Case selected relatively smaller capacity of the power generators. As for these
Demand-to-Supply 50 816 200 two cases, when selecting the smaller capacity of the power generators,
Optimization Case the determined capacity of the power generators in the Demand-to-
Decentralized 100 816 200 Supply Optimization Case is smaller compared to the design results in
Optimization Case
the Game-Theory-Based Optimization Case. The maximum efficiency in

8
J. Luo et al. Energy 313 (2024) 133715

Fig. 10. LCOE of the electricity consumers among these four cases.

Fig. 11. Annual power generation efficiency profile among four cases: Game-Theory-Based Optimization Case (a), Supply-to-Demand Optimization Case (b),
Demand-to-Supply Optimization Case (c), and Decentralized Optimization case (d).

the Game-Theory-Based Optimization Case is up to 88 %, while the effective measures to decrease the LCOE as much as possible. Thus,
maximum efficiency in the Demand-to-Supply Optimization Case is distributed renewable energy generations (especially PV installations on
about 86 %. Table 5 shows the annual average power generation effi- building roofs) have begun to promote and accelerate installation since
ciency among these four cases, where the best power generation effi- renewable power generation has a promising potential to decrease the
ciency is in the Game-Theory-Based Optimization Case, and the operation cost by decreasing the amount of purchased electricity [46].
maximum gap of the average power generation efficiency reaches 2.1 % Conventional optimizations of the microgrid design commonly
compared to Game-Theory-Based Optimization Case with the other consist of two categories (Fig. 12) concerning the requirements and in-
three cases. terests of microgrid investors and electricity consumers. The former
(Fig. 12a) commonly focused on providing optimal design solutions to
5.5. Discussions meet the interests of microgrid power generations solely for microgrid
investors. In contrast, the interests of the electricity consumers are
5.5.1. Advantages of the developed optimization methods ignored. In the optimal design, all different types of power generations
Microgrid systems that include multiple closely spatially distributed are considered for microgrid investors. Good optimization results can be
electricity consumers have been developed to decrease the utility grid’s obtained but are only sometimes consistent with practical conditions
burden further and enhance the power supply’s reliability. If the and may reduce the motivation of PV installations for electricity con-
mentioned power supply duties are achieved, and the net profit is pos- sumers. To face the different requirements and interests of microgrid
itive for microgrid investors, microgrid development will be accepted by investors and electricity consumers, the latter tries to provide some so-
the markets. On the other hand, electricity consumers seek to find lutions, where three typical optimal design approaches are shown in
Fig. 12b-①, Fig. 12b-②, and Fig. 12b-③. The details of these three
optimal design approaches are introduced in Section 5.1, which are
Table 5
The annual average power generation efficiency among these four cases. listed as the reference cases in this study. According to the tested results,
the proposed game theory-based optimal design approach performs
Case Name Average power generation efficiency
better than these three cases.
Game-Theory-Based Optimization Case 77.5 %
Supply-to-Demand Optimization Case 76.1 %
5.5.2. Limitations of the developed optimization methods
Demand-to-Supply Optimization Case 76.7 %
Decentralized Optimization case 75.4 % As demonstrated in this study, the two pivotal limitations cannot be

9
J. Luo et al. Energy 313 (2024) 133715

Fig. 12. Schematics of the conventional microgrid optimal design approaches.

overlooked when assessing the practicality and scalability of the pro- reaches up to 0.08 USD/kWh. Furthermore, under the premise of
posed optimization method. realizing the positive net profit, the generated electricity per unit is
four times higher than that of the second-best optimization case.
● Firstly, while the developed optimization technique has proven to be Considering the impacts of PV installations with battery storage
a successful case in simulations, it is crucial to conduct rigorous equipment in the building roof for electricity consumers, the highest
experimental validations to ensure its economics before it can be average power generation efficiency is achieved, reaching up to 77.5
widely applied in practical scenarios. This step is indispensable for %.
bridging the gap between theoretical models and real-world • From the viewpoint of electricity consumers, the proposed game
applications. theory-based microgrid optimal design approach leads to a sub-
● Secondly, when confronted with the complexities of practical con- stantial reduction in the LOCE. Under the practical premise of
ditions, the optimization method becomes significantly intricate. It feasibility (positive net profit of the microgrid power generation),
necessitates seamless collaboration among all stakeholders, the LCOE is reduced by approximately 14 %.
including building owners, energy providers, and regulatory bodies,
each with their unique requirements, interests, and constraints. This In the pursuit of advancing sustainable energy solutions, several
requirement poses a substantial engineering challenge, as collecting avenues for future research emerge from the findings of this study. One
comprehensive and accurate information from these diverse stake- crucial area of exploration lies in the scalability and adaptability of the
holders is often difficult and time-consuming. Moreover, coordi- proposed game theory-based microgrid optimization design approach.
nating their efforts to align with the optimization objectives can be a Examining its efficacy across a spectrum of microgrid scenarios and
daunting task, further complicating the implementation process. diverse geographical locations can offer valuable insights into its
broader applicability and potential modifications required for varying
6. Conclusions and future studies contexts. Moreover, delving into the long-term implications of wide-
spread adoption of this optimization approach is imperative. Future
This study proposes a novel game theory-based microgrid optimal studies should aim to unravel the economic and environmental impacts
design approach for planning microgrid power generators and PV in- over extended periods, considering factors such as technology evolution,
stallations with battery storage on the building roofs of electricity con- energy market dynamics, and evolving consumer behaviours.
sumers, considering the diverse requirements and interests of both
electricity suppliers and consumers. The analysis of microgrid investors’ CRediT authorship contribution statement
and electricity consumers’ requirements and interests is approached
from different perspectives. The main conclusions can be summarized as Jianing Luo: Writing – original draft, Software, Methodology,
follows. Investigation, Funding acquisition, Formal analysis, Data curation,
Conceptualization. Karthik Panchabikesan: Writing – original draft,
• The proposed game theory-based microgrid optimal design Methodology, Investigation. Kee-hung Lai: Writing – original draft,
approach, functioning as an interaction design strategy, effectively Methodology, Investigation. Timothy O. Olawumi: Writing – original
mitigates the negative effects of decentralized optimization in con- draft, Methodology, Investigation. Modupe Cecilia Mewomo: Writing
ventional optimal design methods. It comprehensively addresses the – original draft, Methodology, Investigation. Zhengxuan Liu: Writing –
varied requirements and interests of all stakeholders, leading to review & editing, Writing – original draft, Supervision, Software,
improved benefits for both microgrid investors and electricity con- Methodology, Investigation, Formal analysis, Conceptualization.
sumers compared to conventional design optimizations.
• From the standpoint of microgrid power generation, the approach
Declaration of competing interest
achieves high net profit coupled with enhanced high power gener-
ation efficiency. The net profit of the microgrid power generation
The authors declare that they have no known competing financial

10
J. Luo et al. Energy 313 (2024) 133715

interests or personal relationships that could have appeared to influence [20] Fudenberg D, Tirole J. Game theory. MIT press; 1991.
[21] Wang Y, Saad W, Han Z, Poor HV, Başar T. A game-theoretic approach to energy
the work reported in this paper.
trading in the smart grid. IEEE Trans Smart Grid 2014;5:1439–50.
[22] Dabush I, Cohen C, Pearlmutter D, Schwartz M, Halfon E. Economic and social
Acknowledgements utility of installing photovoltaic systems on affordable-housing rooftops: a model
based on the game-theory approach. Build Environ 2022:109835.
[23] Lin J, Dong J, Dou X, Liu Y, Yang P, Ma T. Psychological insights for incentive-
The research presented in this study is financially supported by based demand response incorporating battery energy storage systems: a two-loop
Natural Science Foundation of Sichuan Province (No: 2024NSFSC0915) Stackelberg game approach. Energy 2022;239:122192.
and is also supported by the Fundamental Research Funds for the Central [24] Bidgoli MA, Ahmadian A. Multi-stage optimal scheduling of multi-microgrids using
deep-learning artificial neural network and cooperative game approach. Energy
Universities (No.: A0920502052401-177). 2022;239:122036.
[25] Zhou Y. Incentivising multi-stakeholders’ proactivity and market vitality for
Data availability spatiotemporal microgrids in Guangzhou-Shenzhen-Hong Kong Bay Area. Appl
Energy 2022;328:120196.
[26] Liu Z, Sun Y, Xing C, Liu J, He Y, Zhou Y, et al. Artificial intelligence powered
Data will be made available on request. large-scale renewable integrations in multi-energy systems for carbon neutrality
transition: challenges and future perspectives. Energy and AI 2022;10:100195.
[27] Yu X, Pan D, Zhou Y. A Stackelberg game-based peer-to-peer energy trading market
References with energy management and pricing mechanism: a case study in Guangzhou. Sol
Energy 2024;270:112388.
[1] Jia Z, Lin B, Wen S. Electricity market Reform: the perspective of price regulation [28] Nwulu NI, Xia X. Multi-objective dynamic economic emission dispatch of electric
and carbon neutrality. Appl Energy 2022;328:120164. power generation integrated with game theory based demand response programs.
[2] Luo J, Yuan Y, Joybari MM, Cao X. Development of a prediction-based scheduling Energy Convers Manage 2015;89:963–74.
control strategy with V2B mode for PV-building-EV integrated systems. Renew [29] Luo J, Cao X, Yuan Y. Comprehensive techno-economic performance assessment of
Energy 2024:120237. PV-building-EV integrated energy system concerning V2B impacts on both building
[3] McCollum DL, Zhou W, Bertram C, De Boer H-S, Bosetti V, Busch S, et al. Energy energy consumers and EV owners. J Build Eng 2024:109075.
investment needs for fulfilling the Paris agreement and achieving the sustainable [30] Giraldez Miner JI, Flores-Espino F, MacAlpine S, Asmus P. Phase I microgrid cost
development goals. Nat Energy 2018;3:589–99. study: data collection and analysis of microgrid costs in the United States. Golden,
[4] Schmidt TS, Sewerin S. Technology as a driver of climate and energy politics. Nat CO (United States): National Renewable Energy Lab.(NREL); 2018.
Energy 2017;2:1–3. [31] Khodaei A, Shahidehpour M. Microgrid-based co-optimization of generation and
[5] Luo J, Li H, Wang S. A quantitative reliability assessment and risk quantification transmission planning in power systems. IEEE Trans Power Syst 2012;28:1582–90.
method for microgrids considering supply and demand uncertainties. Appl Energy [32] Kusakana K. Optimal energy management of a grid-connected dual-tracking
2022;328:120130. photovoltaic system with battery storage: case of a microbrewery under demand
[6] Luo J, Li H, Huang G, Wang S. A multi-dimensional performance assessment response. Energy 2020;212:118782.
framework for microgrids concerning renewable penetration, reliability, and [33] Owen G. Game theory. Emerald Group Publishing; 2013.
economics. J Build Eng 2023;63:105508. [34] Zhang J, Ji L. Optimization of daylighting, ventilation, and cooling load
[7] Alfergani A, Alaaesh S, Shamekh A, Khalil A, Asheibi A. Improved power sharing in performance of apartment in tropical ocean area based on parametric design. Adv
inverter based microgrid using multi-objective optimization. Comput Electr Eng Civ Eng 2021;2021.
2023;110:108902. [35] Luo J, Zhuang C, Liu J, Lai K-h. A comprehensive assessment approach to quantify
[8] Zhang G, Ge Y, Pan X, Zheng Y, Yang Y. Hybrid robust-stochastic multi-objective the energy and economic performance of small-scale solar homestay hotel systems.
optimization of combined cooling, heating, hydrogen and power-based microgrids. Energy Build 2023;279:112675.
Energy 2023;274:127266. [36] Qiu S, Wang K, Lin B, Lin P. Economic analysis of residential solar photovoltaic
[9] Lou J, Cao H, Meng X, Wang Y, Wang J, Chen L, et al. Power load analysis and systems in China. J Clean Prod 2021;282:125297.
configuration optimization of solar thermal-PV hybrid microgrid based on [37] Zheng CY, Wu JY, Zhai XQ, Wang RZ. Impacts of feed-in tariff policies on design
building. Energy 2024;289:129963. and performance of CCHP system in different climate zones. Appl Energy 2016;
[10] Parizad A, Hatziadoniu K. Security/stability-based Pareto optimal solution for 175:168–79.
distribution networks planning implementing NSGAII/FDMT. Energy 2020;192: [38] Daud AK, Ismail MS. Design of isolated hybrid systems minimizing costs and
116644. pollutant emissions. Renew Energy 2012;44:215–24.
[11] Elkadeem MR, Kotb KM, Abido MA, Hasanien HM, Atiya EG, Almakhles D, et al. [39] Venkateswari R, Sreejith S. Factors influencing the efficiency of photovoltaic
Techno-enviro-socio-economic design and finite set model predictive current system. Renewable Sustainable Energy Rev 2019;101:376–94.
control of a grid-connected large-scale hybrid solar/wind energy system: a case [40] Li H, Wang S. Coordinated optimal design of zero/low energy buildings and their
study of Sokhna Industrial Zone, Egypt. Energy 2024;289:129816. energy systems based on multi-stage design optimization. Energy 2019;189:
[12] Jung J, Villaran M. Optimal planning and design of hybrid renewable energy 116202.
systems for microgrids. Renewable Sustainable Energy Rev 2017;75:180–91. [41] Alhajeri NS, Dannoun M, Alrashed A, Aly AZ. Environmental and economic
[13] Yang Y, Bremner S, Menictas C, Kay M. Battery energy storage system size impacts of increased utilization of natural gas in the electric power generation
determination in renewable energy systems: a review. Renewable Sustainable sector: evaluating the benefits and trade-offs of fuel switching. J Nat Gas Sci Eng
Energy Rev 2018;91:109–25. 2019;71:102969.
[14] Tsao Y-C, Thanh V-V, Wu Q. Sustainable microgrid design considering blockchain [42] Brouwer AS, van den Broek M, Seebregts A, Faaij A. Operational flexibility and
technology for real-time price-based demand response programs. Int J Electr Power economics of power plants in future low-carbon power systems. Appl Energy 2015;
Energy Syst 2021;125:106418. 156:107–28.
[15] Gamil MM, Senjyu T, Takahashi H, Hemeida AM, Krishna N, Lotfy ME. Optimal [43] Kuramochi T, Ramírez A, Turkenburg W, Faaij A. Techno-economic prospects for
multi-objective sizing of a residential microgrid in Egypt with different ToU CO2 capture from distributed energy systems. Renewable Sustainable Energy Rev
demand response percentages. Sustainable Cities Soc 2021;75:103293. 2013;19:328–47.
[16] Swaminathan S, Pavlak GS, Freihaut J. Sizing and dispatch of an islanded [44] Van den Broek M, Veenendaal P, Koutstaal P, Turkenburg W, Faaij A. Impact of
microgrid with energy flexible buildings. Appl Energy 2020;276:115355. international climate policies on CO2 capture and storage deployment: illustrated
[17] Tomin N, Shakirov V, Kozlov A, Sidorov D, Kurbatsky V, Rehtanz C, et al. Design in the Dutch energy system. Energy Pol 2011;39:2000–19.
and optimal energy management of community microgrids with flexible renewable [45] Kuramochi T, Faaij A, Ramírez A, Turkenburg W. Prospects for cost-effective post-
energy sources. Renew Energy 2022;183:903–21. combustion CO2 capture from industrial CHPs. Int J Greenh Gas Control 2010;4:
[18] Pelka S, Chappin E, Klobasa M, de Vries L. Participation of active consumers in the 511–24.
electricity system: design choices for consumer governance. Energy Strategy Rev [46] Li B, Roche R, Miraoui A. Microgrid sizing with combined evolutionary algorithm
2022;44:100992. and MILP unit commitment. Applied energy 2017;188:547–62.
[19] Tang R, Wang S, Li H. Game theory based interactive demand side management
responding to dynamic pricing in price-based demand response of smart grids. Appl
Energy 2019;250:118–30.

11