Engineering Letters, 31:1, EL_31_1_09

______________________________________________________________________________________

Economic and Optimal Dispatch Model of

Electricity, Heat and Gas for Virtual Power
Plants in Parks Considering Low Carbon
Targets
Wei-Guo Zhang, Qing Zhu, Hong-Juan Zheng, Lin-Lin Gu and Hui-Jie Lin

approximately 10 billion tonnes, China has a great

Abstract—To promote the consumption and development of responsibility to reduce carbon emissions as it is world's
new and green energy sources, we propose an economically largest carbon emitter[1]. To slow the depletion of fossil fuel
optimal dispatch model to simulate virtual power plants (VPPs) sources and reduce pollution in the environment, the
in parks considering low-carbon objectives based on a dual-
carbon background. First, individual models involved in the
integrated energy supply of electricity, heat and gas has
optimal dispatch of VPPs are introduced. A two-stage operation flourished by achieving a complementary combined supply
process of electricity-to-hydrogen transformation and hydrogen between multiple energy sources and improving the
methanation of power-to-gas equipment is refined and analysed. efficiency of primary energy use [2-5]. Among these sources,
Unstable wind power is converted into hydrogen and supplied virtual power plants (VPPs) have focused on the coupling of
to the hydrogen load during the electrolysis of the water-to- multiple energy flows, which allows the mutual
hydrogen stage in conjunction with the development of
hydrogen energy demands. Second, to minimise the sum of
transformation and transmission of electricity, heating,
operating and carbon emission costs, an optimal dispatch model cooling and gaseous forms of energy. VPPs also allow for the
utilised by the virtual power plant considering a low-carbon overall planning and operational coordination and
objective is established. A simulation of the calculation case optimisation of the distribution, transformation, storage and
shows that the proposed model can achieve efficient energy consumption of multiple energy sources to realise diversified
utilisation and improved economic and carbon costs for a VPP energy supplies, energy services and energy use methods.
in the park. Finally, a sensitivity analysis of the system cost is
conducted by adding weighting coefficients; the optimal range
VPPs achieve internal collaborative optimisation by
of weighting coefficients for the VPP in the park proposed in this aggregating multiregional, large-scale distributed energy
paper are derived for reference by considering economic sources and participating in overall grid dispatch [5-6]. A
benefits and environmental protection. multiobjective optimal VPP scheduling model, including
distributed power supply, electric vehicles, and loads, was
Index Terms—carbon emissions, virtual power plant, power- constructed in [7], and an interruptible load (IL) in the form
to-gas, demand response, optimal dispatch of a demand response was considered in the electric load.
Some researchers [8] have studied the optimal coordinated
VPP scheduling in the main electricity market and rotating
I. INTRODUCTION
standby market. This VPP aggregated distributed power

I n recent years, as energy, climate and environmental

challenges have worsened, countries have been making
changes to their energy structures, and energy consumption
sources and electrical loads. Other researchers [9] have
proposed a coordinated optimal dispatch model of electric
and thermal VPPs with microgas turbines (MTs), gas boilers
has gradually shifted to various renewable energy sources. It and energy storage devices. Additionally, some scholars [10]
is estimated that China's annual carbon emissions are have studied VPPs from the perspective of demand response
and have established an incentive- and price-based demand
Manuscript received June 25, 2022; revised November 25, 2022.
This work is supported by Science and Technology project of State Grid
response VPP model. Other scholars [11] have aggregated a
Corporation（Research and application of zero-carbon evolution technology multiregion VPP integrated energy optimisation dispatch
of micro energy grids in national carbon neutral parks, NO. 5100- model for cogeneration units, gas boilers and wind turbines.
202118566A-0-5-SF）. Researchers [12] have also established a VPP optimal
Wei-guo Zhang is a PhD candidate at Southeast University, Nanjing dispatch model including MTs, gas boilers, distributed power
210096, China (Corresponding author: e-mail: [email protected] )
Qing Zhu is a senior engineer of NARI-TECH Nanjing Control Systems sources, and electric vehicles. Existing studies have mainly
Co., Ltd., Nanjing 211106, China (e-mail: [email protected]) focused on external participation in the main electricity
Hong-juan Zheng is a mid-level engineer of NARI-TECH Nanjing market; they rarely involve VPP participation in the heat
Control Systems Co., Ltd., Nanjing 211106, China (e-mail:
[email protected] ) market or the gas market and only realise the optimal dispatch
Lin-lin Gu is a mid-level engineer of NARI-TECH Nanjing Control of electricity [7-10] or the coordinated optimal dispatch
Systems Co., Ltd., Nanjing 211106, China (e-mail: between electricity and heat [11-12] for VPPs, failing to
[email protected] )
Hui-jie Lin is a mid-level engineer of NARI-TECH Nanjing Control
realise the coordinated optimal dispatch among multiple
Systems Co., Ltd., Nanjing 211106, China (e-mail: energy sources.
[email protected]) More and more people pay attention to the optimal dispatch

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

problem of VPPs. The objectives adopted in the economic of energy in the production, transmission and distribution
dispatch model include the minimum total cost and the horizontally, while the user-side interaction is fully tapped,
maximum total profit [13]-[15]. In reference [16], a VPP– which effectively improves the flexibility of the the source
load game model was developed with the highest VPP return network, and is connected to the information interaction
as the objective function. In reference [17], researchers systems of the electricity, heat and gas networks to take part
developed an economically optimised dispatch model for a in the large grid dispatch model, reduce carbon costs and
VPP by adding electric vehicles to enhance its flexibility. In optimise system profits [24]. In this paper, a VPP system in
[18], VPPs in separate regions were optimised with the the park which contains P2G is constructed, as shown below.
objective of maximising their overall revenue, and bidding The supply side is supplied by distributed energy, the upper
strategies for VPP participation in the different markets were grid and the gas grid, which provide the energy source for the
obtained. In reference [19], researchers have proposed a VPP system; The conversion side contains the electrolysis tank,
scheduling approach with two stages in order to study impact hydrogen storage tank and methane reactor unit for P2G
of energy storage equipment and customer revenue on VPP system, the waste heat boiler and gas turbine unit for CHP
revenue. Some scholars [20] considered the uncertainties of units and the gas boiler unit; The combined electrical, gas,
electricity prices over time and developed an improved VPP hydrogen and thermal loads form the load side of the system.
multistage stochastic optimisation model. In reference [21], The electrolysis tank converts potentially discarded wind-
in order to obtain the best operational strategy to promote new generated or low-cost electricity into hydrogen, part of which
energy consumption, the researchers established a two-level is injected into the natural gas (NG) system via a methane
electricity-thermal energy trading strategy with an objective reactor to produce methane for direct supply to the gas load
function to reduce the cost of energy acquired by the VPP. or gas turbine cogeneration, reducing the amount of gas
The above studies have mainly enhanced the operational purchased from the upper gas grid and reducing the system's
performance of the energy market by adding energy storage energy purchase costs. The remaining portion, which can
devices, improving the dispatching strategy and tapping into achieve the arbitrage of "high generation and low storage," is
the demand response of consumers; this strategy lacks the stored in the hydrogen storage tanks and supplied towards the
complexity of energy migration, and it assumes the hydrogen load during the times when electricity prices are at
minimisation of operational costs as the objective their highest. This reduces the energy loss brought on by the
optimisation function without sufficiently considering carbon methanation step's use and increases the system's overall
emission costs. efficiency in terms of energy utilization.
According to data published by the International Energy China's first power-to-gas conversion unit using wind and
Agency, carbon emissions from industrial parks in China solar energy has been completed in Dalian and has achieved
account for approximately 31% of the country's total good results. In the future, as the relevant technology
emissions and exhibit continuous growth. As the core unit of develops and matures, the use of the above-mentioned multi-
industrial clustering and development, industrial parks have energy coupled units to optimise VPP scheduling and
become a driving force in China's economic growth and an operation will achieve even better economic benefits.
important factor for achieving scientific and precise carbon Therefore, this paper studies the optimal scheduling of VPP
emission reduction and eventual carbon neutrality [22-24]. with power to gas conversion, which is forward-looking and
Based on the above analysis, we take a park as the context for of theoretical guidance.
our study, consider the operating and carbon emission costs
of a VPP in the park, and assemble a two-stage power-to-gas Power flow Heat flow Fuel flow Hydrogen flow

conversion model with the minimal costs for all systems in Supply side Conversion side Load side
the park as the scheduling target. Through this study, we
refine power-to-gas (P2G) conversion into a two-stage Electrical
Upper grid
operation of electricity-to-hydrogen transformation and Electric
load
P2G
hydrogen methanation, prioritising the high-grade utilisation WT、PV energy storage
Electrolysis tank
of hydrogen energy in the hydrogen production link, which H2 Hydrogen
can reduce the energy loss caused by step conversion in the storage tanks
Hydrogen
CHP
whole operation process. The energy loss caused by Methane reactors load
GT
conversion can be reduced throughout the operation process. CH4
A VPP that aggregates multiple energy sources is constructed, Upper Gas
Gas
load
and an optimal control model under multiple scenarios is Network
HRSG Gas storage
determined by using CPLEX with YALMIP in MATLAB to
Thermal storage GB
evaluate the recommended model's performance.
Heat network Heat load
II. SYSTEM STRUCTURE OF THE VPP
The VPP system covers a wide range of energy supply and Fig. 1. VPP structure
conversion devices, including wind power plants (WPPs), PV
generators, combined heat and power (CHP), P2G, gas
boilers and energy storage devices. The system can prioritise
the advantages of interactively coupling electricity, heat and
gas to achieve synergistic optimisation among different
energy sources vertically, coordinate and optimise each type

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

III. SINGLE UNIT MODELS INVOLVED IN OPTIMAL H GB (t ) = PGB

heat
(t )t (7)
SCHEDULING OF THE VPP heat
where the GB’s output thermal power is denoted as PGB (t);
A. Model of power supply equipment the NG flow consumed by the GB is denoted as Q GB (t);
HGB (t) refers to the heat generated by the GB; and ηGB refers
Distributed wind and photovoltaic resources are clean
to the gas heat conversion efficiency of the GB.
energy sources, and the amount of carbon emissions they
produce can be successfully reduced, which is more in line C. Model of energy storage equipment
with the low carbon goals of this article, the power supply (1) Electricity storage (ES)
equipment in this paper mainly includes wind turbines and Electricity storage (ES) can effectively overcome the
PV units. shortcomings of clean energy output volatility. The specific
(1) Wind turbine generation mathematical model is as follows:
The output formula is as follows:
EES (t + 1) = (1 − δ ) EES (t ) + [ ηES PES (t + 1)
ES store store

 0 v  vin (8)
 3 − PES (t + 1) / ηES ]t
release release

av + bv + cv + d vin  v  ve
2

PWT (t ) =  (1) Where the charging and discharging powers of the ES

 Pe ve  v  vout store release
system are denoted as PES (t) and PES (t), respectively;

 0 v  vout ηES refers to the charging efficiency of the ES ; ηrelease
store
ES refers
Where the output power of the wind turbine is denoted as to discharging efficiency of the ES system; δES is the self-loss
𝑃WT (𝑡); 𝑃𝑒 refers to the rated power of the wind turbine; 𝑣𝑒 rate of the ES system; the storage capacities of the ES system
refers to the rated wind speed; 𝑣𝑖𝑛 refers to cut-in wind speed; are denoted as EES (t).
𝑣𝑜𝑢𝑡 stands for the cut-off wind speed; and 𝑎, 𝑏, c and 𝑑 are (2) Heat storage (HS)
coefficients. The specific mathematical model is as follows:
H HS (t + 1) = (1 − δ ) H HS (t ) + [ ηHS PHS (t + 1)
HS store store
(2) Photovoltaic (PV) power generation
The power output is modelled as follows: (9)
− PHS (t + 1) / ηHS ]t
release release

PPV (t ) = λ  ρPV  φ(t ) (2)

store release
where the PV unit's output power is denoted as 𝑃PV (𝑡); The where PHS (t) and PHS (t) are the heat storage and heat
PV panels' conversion efficiency is denoted by the λ, while release power of the HS system during period 𝑡, respectively;
their total area is denoted by the 𝜌𝑃𝑉 ; and φ(t) is the ηstore
HS refers to heat storage efficiency of the HS system;

illumination intensity during period 𝑡. ηrelease

HS refers to heat release efficiency of the HS system; δHS
refers to self-loss rate of the HS system; and HHS (t) and
B. Model of heating equipment HHS (t-1) stand for the heat storage of the HS system,
(1) Combined heat and power (CHP) respectively.
There are two components of a CHP unit: a gas turbine (GT) (3) Gas storage (GS)
and a heat recovery steam generator (HRSG), which is The specific mathematical model is as follows:
mathematically modelled as follows: QGS (t + 1) = (1 − δ )QGS (t ) + [ ηGS PGS (t + 1) / Lgas
GS store store

QGT (t ) Lgas (10)

ele
PGT (t ) = ηGT (3) − PGS
release
(t + 1) / ( ηGS Lgas )]t
release

t
store release
ele
PGT (t )(1 − ηGT − ηq ) where 𝑃GS (𝑡) and 𝑃GS (𝑡) refer to gas storage and
P heat
HRSG (t ) = K he
HRSG M sr
HRSG
(4) bleeding power of the GS system; 𝜂GS store release
and 𝜂GS refer to
ηGT
gas storage efficiency and gas discharge efficiency of GS
H HRSG (t ) = PHRSG
heat
(t )t (5) system, respectively; 𝛿 GS is the self-loss rate of the GS
system; and 𝑄GS (𝑡) refers to NG volumes stored by the GS
𝑒𝑙𝑒
where 𝑃𝐺𝑇 (𝑡) and 𝑄𝐺𝑇 (𝑡) refer to the output electric system during periods 𝑡, respectively.
power and the GT's volumetric consumption of NG; 𝐿𝑔𝑎𝑠
D. Model of power-to-gas equipment
refers to the low calorific value of NG; the thermal power
output and the heat output value of HRSG are denoted as (1) Power to gas (P2G)
heat
𝑃HRSG (𝑡) and 𝐻HRSG (𝑡); 𝜂𝐺𝑇 refers to GT’s power generation There are two types of P2G equipment: one converts
efficiency ; 𝜂𝑞 refers to the heat dissipation loss rate of the electric energy into hydrogen with an efficiency of 75%–85%,
he sr and the other converts hydrogen into electric energy into
HRSG; 𝐾HRSG and 𝑀HRSG are the heating coefficient and flue
natural gas with an efficiency of 45%–60% through
gas recovery rate of the bromine cooler, respectively; and Δt
methanation. A two-step chemical equation is shown in
is the unit scheduling period. Equation (11).
(2) Gas boiler (GB)
A GB is a type of energy conversion equipment that 2H 2 O ⎯⎯⎯⎯
electrolysis
→ 2H 2 + O 2
produces heat by consuming natural gas. A commonly used  (11)
CO 2 +4H 2 ⎯⎯⎯→ CH 4 +2H 2 O
catalyst
mathematical model expression is
QGB (t ) Lgas A traditional P2G system produces natural gas by
heat
PGB (t ) = ηGB (6)
t methanating all the hydrogen generated by water electrolysis,

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

and if there is a demand for hydrogen load in the system, it is Equation (15).
purchased from the higher gas network, which not only PMR = ηMR PMRi (15)
reduces safety but also prevents efficient P2G use. This paper
considers the high efficiency of an initial electricity-to- where 𝑃MR and 𝑃MRi are the output power and input power
hydrogen process and the developing trend of a hydrogen of the methane reactor during period 𝑡, respectively, and 𝜂MR
energy demand in the future; the two factors are combined to refers to the conversion efficiency of the methane reactor.
draw a two-stage P2G operation process, as shown in Figure (4) Hydrogen storage tank
2. The first stage is the hydrogen energy generation stage, The model is shown in Equation (16).
where the wind power that may originally be discarded PHSTo,t
through rectification flows into the electrolyser through the EHST,t = EHST,t-1 + ηHSTi,t PHSTi,t − (16)
electrolysis of water to convert electrical energy. The ηHSTo
electrolyser converts electricity into hydrogen energy by
water electrolysis, and at the same time, to improve where 𝐸HST,t refers to capacity of the hydrogen storage
operational flexibility, hydrogen energy is stored in tank; the input and output power of the hydrogen storage tank
additional storage tanks to supply hydrogen. The second stage are denoted as 𝑃HSTi and 𝑃HSTo during period 𝑡, respectively;
is the consumption of hydrogen energy, and if there is any 𝜂HSTi and 𝜂HSTo refer to hydrogen storage efficiency and
surplus hydrogen, it is fed into the methane reactor and hydrogen release efficiency.
converted into CH4. This is used by the CHP and the GB, thus E. Integrated demand response load
not only achieving an efficient use of energy but also
With the increase in flexible loads such as electric vehicles
reducing risk when purchasing and transporting hydrogen,
and inverter air conditioners on the energy consumption side,
thus achieving a win‒win situation.
it is increasingly important to study demand response loads
to improve the flexible operation of integrated energy
systems. It is characterised by the fact that the total amount
O2 H2O H2O CO2 of load and the amount of load in each time period is not
Wind power, Methane Natural Gas
constant during the dispatch time. Within the defined time
Electrolyser period and the amount of load, the demand response load can
upper grid reactors CH4 Network
be shifted, reduced and substituted to achieve the desired
electricity
requirements by processing the different transgressive loads.
H2
In general, demand response loads can be shifted, curtailable
Hydrogen or substitutable loads. In this paper, only thermal loads are
storage tanks replaced by electrical loads. The specific model is as follows:
Fig. 2. Two-stage P2G operation process Pload (t ) = Pload
'
(t ) + Pload
cut
(t ) + Pload
tran
(t ) + Pload
rep
(t ) (17)
A conventional P2G conversion process produces natural H load (t ) = H load
'
(t ) + H load
cut
(t ) + H load
tran
(t ) − Pload
rep
(t ) (18)
gas by consuming electrical energy to meet demand of the
system. The power output is modelled as Equation (12-13): − Pload,max
cut
(t )  − Pload
cut
(t )  0 (19)
P (t ) = P (t )ηP2G
gas
P2G
ele
P2G
(12)
T

QP2G (t ) =
P (t ) t
gas
P2G
(13)
P
t =1
tran
load
(t ) t = 0, tran
Pload (t )  Pload,max
tran
(t ) (20)
Lgas
′ ′
where 𝑃load (𝑡), 𝑃load (𝑡) , 𝐻load (𝑡) and 𝐻load (𝑡) refer to the
where the NG power output and the electric power electrical load demand value, predicted value of electrical
gas
consumed by the P2G system are denoted as 𝑃P2G (𝑡) and load demand, thermal load demand value and predicted value
ele cut
𝑃P2G (𝑡); 𝑄P2G (𝑡) refers to the NG volume generated by the of thermal load demand of the VPP, respectively; 𝑃load (𝑡),
tran rep
P2G system during period 𝑡; and 𝜂P2G refers to electrical 𝑃load (𝑡) and 𝑃load (𝑡) are the reducible, transferable and
conversion efficiency of the P2G process. replaceable electrical loads of the VPP system, respectively;
cut tran
(2) Electrolyser 𝐻load (𝑡) and 𝐻load (𝑡) are the reducible and transferable heat
The electrolyser can convert electrical energy into loads; and the maximum values of the reducible and
cut
hydrogen by the electrolysis of water, the mathematical transferable electrical loads are denoted as 𝑃load,max (𝑡) and
model of which is shown in Equation (14). tran
𝑃load,max (𝑡).
PEC = ηEC PECi (14)
IV. ECONOMIC OPTIMAL DISPATCH MODEL FOR VPPS
where 𝑃EC and 𝑃ECi are the output power and input power
of the electrolyser, respectively; and 𝜂EC refers to conversion A. Target function
efficiency of the electrolyser. To achieve economic optimisation and minimise carbon
(3) Methane reactor emissions, we transform a multiobjective problem into a
Hydrogen reacts with CO2 in a methane reactor via the single-objective optimisation problem by introducing a
Sabatier process to form CH4 and water, where the carbon tax 𝑐𝑡𝑎𝑥 , which converts carbon emissions into carbon
mathematical model of the methane reactor is shown in costs. With the objective function of minimising the sum of

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

the operating cost 𝐶1 and the carbon emission cost 𝐶2 of the CMAIN = [ f PV (t ) PPV (t ) + f WT (t ) PWT (t )
VPP in the park, the optimisation objective is established as
follows: + f CHP ( PGTele (t ) + PHRSG
heat
(t ))
 + f GB PGB
heat
(t ) + f P2G PP2G
gas
(t )
min C = C1 + C2 (24)
 + f ES ( PESstore (t ) − PESrelease (t ))
 24

C1 =  [(CG + CIDR + CMAIN + CGRID )] (21) + f HS ( PHSstore (t ) − PHSrelease (t ))

 t =1

 24 + f GS ( PGSstore (t ) − PGSrelease (t ))]t

 2 tax  (eCHP + eGB )
C = c
 t =1
where 𝑓PV , 𝑓WT , 𝑓CHP , 𝑓GB , 𝑓P2G , 𝑓ES , 𝑓HS and 𝑓GS are the
where 𝐶G , 𝐶IDR , 𝐶MAIN and 𝐶GRID are the system energy corresponding unit maintenance costs of the PV unit, wind
purchase costs, compensation costs for demand-side response turbine unit, CHP system, GB, P2G system, and energy
load, the expense of each unit's equipment upkeep and the storage device, respectively.
interaction cost of purchasing and selling electricity with the 4) Interaction costs between the system and the higher-
superior power grid, respectively, and 𝑒CHP and 𝑒GB are the level grid
carbon emission cost of CHP and the gas-fired boiler, The demand for electricity in industrial parks is generally
respectively. large, while scenic power units generally have limited and
(1) Operating cost 𝐶1 uncertain installed capacities. Therefore, in the processes of
1) System energy purchase costs system optimisation and control, if the power supply through
There are a number of devices that require electricity and the wind power output is insufficient, it needs to buy power
NG to start them up, so the system of VPP requires the from the superior grid. Combining actual peak and valley
purchase of external energy to meet the operating needs. The tariffs and the operating characteristics of the CHP, P2G, EB
cost can be expressed as Equation (22): and ES systems, the levels of interaction between the systems
CG =  f BG (t ) PBG (t ) + f gas (t )QBG (t )  t (22)
and the upper grid can be further improved. In addition, if the
system produces excess power, it can also be sold to the
where 𝑓BG (𝑡) and 𝑓gas (𝑡) refer to the unit prices when the higher power grid, thus reducing the system operating costs.
system purchases electricity from the upper grid and NG from The cost of the system's interaction with the parent grid is as
the superior gas network in period 𝑡 , 𝑄BG (𝑡) is the external follows:
natural gas volume purchased by the system during period 𝑡, CGRID = [ fBG (t ) PBG (t ) − fSG (t ) PSG (t )]t (25)
and 𝛥𝑡 is the unit scheduling period.
2) Compensation costs for demand-side response load where the power purchase price and power sale price are
volumes denoted as 𝑓BG (𝑡) and 𝑓SG (𝑡) , respectively; 𝑃BG (𝑡) and
𝑃SG (𝑡) refer to the power purchased and sold to the superior
CIDR = ( f cutele Pload
cut
(t ) + f tran
ele tran
Pload (t ) power grid during period t.
(2) Cost of carbon emissions 𝐶2
+ f repele Pload
rep
(t ) + f cutheat H load
cut
(t ) (23) The cost of carbon emissions consists of two parts: the cost
of carbon emissions from the CHP system (𝑒𝐶𝐻𝑃 ) and the GB
+ f tran
heat tran
H load (t ) ) t
system (𝑒𝐺𝐵 ), where 𝑒𝐶𝐻𝑃 is the difference between the CHP
ele ele ele unit carbon emissions 𝑐𝐶𝐻𝑃 and the carbon emission
where 𝑓cut , 𝑓tran and 𝑓rep refer to reducible, transferable
allowances 𝑐𝐶𝐻𝑃𝑏 .
and replaceable electric loads' unit compensation cost
coefficients, respectively; 𝑓cut heat heat
and 𝑓tran refer to unit eCHP = cCHP − cCHPb ,
 (26)
compensation cost coefficients of the heat load that can be
cut tran eGB = μGB PGBh
reduced and transferred, respectively; 𝑃load (𝑡), 𝑃load (𝑡) and
rep
𝑃load (𝑡) are the reducible, transferable and replaceable 
electrical loads of the system during period 𝑡, respectively;  PCHP = PCHPe + γPCHPh ,
the reducible, transferable heat loads are denoted as 𝐻load cut
(𝑡) 
 24

 CHP  (α1 PCHP + α2 PCHP + α3 ) (27)

tran
and 𝐻load (𝑡), respectively. c = 2

3) Maintenance costs for each unit in the system

 t =1
Each unit incurs corresponding maintenance costs during  24
the operation of the system. Maintenance costs vary from unit cCHPb = β  PCHP
to unit. The maintenance cost is as follows:  t =1

where the unit 𝐶𝑂2 emission cost of the GB is denoted as

𝜇GB ; The electric power of the CHP unit transformed to the
pure condensing condition at period t is indicated by the
symbol 𝑃CHP ; 𝛾 refers to reduction value of the electric power
when the CHP unit increases the unit thermal power under
fixed air intake; 𝛼1 , 𝛼2 , and 𝛼3 refers to the carbon emission

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

coefficients of CHP; the carbon trading quota per unit of − RGT

down
t  PGTele (t ) − PGTele (t -1)  RGT
up
t
electricity is denoted as 𝛽.  down
− RGB t  PGB (t ) − PGB (t -1)  RGB t (31)
heat heat up
B. Constraint condition
 down
− RP2G t  PP2G (t ) − PP2G (t -1)  RP2G t
gas gas up
(1) Power balance constraint
Power balance, heat balance and gas balance constraints
need to be met in the scheduling and operation of the park’s where the minimum and maximum output electric powers
VPP. of GT are denoted as 𝑃GTmin (𝑡) and 𝑃GTmax (𝑡), respectively;
the minimum and maximum values of the output power of
 Pload (t ) = PPV (t ) + PWT (t ) + PGTele (t )
 GB are denoted as 𝑃GBmin (𝑡) and 𝑃GBmax (𝑡); the minimum
− PP2G (t ) + PBG (t ) − PSG (t )
ele
 and maximum values of the output NG power of P2G are
up down
 + PES (t ) − PES (t )
release store denoted as 𝑃P2Gmin (𝑡) and 𝑃P2Gmax (𝑡) ; 𝑅GT and 𝑅GT are
 upper and lower climbing speed limits of the GT system,
up
 H load (t ) = H HRSG (t ) + H GB (t ) (28) respectively; 𝑅GB and 𝑅GB down
are the upper and lower
 + H HS (t − 1) − H HS (t ) climbing speed limits of the GB system, respectively; and the
 upper and lower climbing speed limit of P2G are denoted as
Qload (t ) = QBG (t ) + QP2G (t ) − QGB (t ) up down
𝑅P2G and 𝑅P2G .

 − QGT (t ) + QGS (t − 1) − QGS (t ) (4) Energy storage equipment constraints
 EESmin  EES (t )  EESmax
where the electrical load, thermal load and gas load are  min
 H HS  H HS (t )  H HS
max
denoted as 𝑃load (𝑡) , 𝐻load (𝑡) and 𝑄load (𝑡) , respectively; (32)
𝑃BG (𝑡) and 𝑃SG (𝑡) refer to the power purchased and sold  min
QGS  QGS (t )  QGS
max
release
from the superior power grid, respectively; 𝑃ES (𝑡) and
store
𝑃ES (𝑡) are the discharging power and the charging power of
0  Pstore (t )  Ss (t ) Ps max
ES during period 𝑡; 𝐻HS (𝑡) refers to the heat storage of the  (33)
HS system during periods 𝑡 ; 𝑄BG (𝑡) and 𝑄P2G (𝑡) are the 0  Prelease (t )  S r (t ) Pr max
amounts of external NG purchased by the system and NG
generated by the P2G system, respectively; and the gas  Ss (t ) = {0,1}
 (34)
storage capacity of the GS system is denoted as 𝑄GS (𝑡).
 S r (t ) = {0,1}
(2) System interaction constraints with the higher-level
master network 0  Ss (t ) + Sr (t )  1 (35)
When the power within the VPP is higher or lower than the
min max min max min max
load, the power balance can be ensured through power where 𝐸ES ,𝐸ES 𝐻HS
, , 𝐻HS , 𝑄GS and 𝑄GS
refer to
interactions with the public grid for power purchase and sale minimum and maximum capacities of ES, HS, and GS,
and between VPPs, and when considering the power respectively; 𝑃store (𝑡) and 𝑃release (𝑡) refer to charging and
interactions and demand response between VPPs, the power discharging powers of ES; the upper limits of the charging
exchange with the public grid can be reduced, thus reducing and discharging power of ES are denoted as 𝑃smax and 𝑃rmax ,
the overall operating costs of the VPP. In order to reduce the respectively; the charging and discharging states of ES are
regulation pressure on the upper mains network, this paper denoted as 𝑆s (𝑡) and 𝑆r (𝑡) during period 𝑡, respectively; and
only takes into account the system's energy purchase from the 1 and 0 indicate that ES is in the charging and discharging
upper mains network without considering energy sales. states or the noncharging and discharging states, respectively.
 PB min (t )  PBG (t )  PB max (t )
 V. EXAMPLE ANALYSIS
 PS min (t )  PSG (t )  PS max (t ) (29)
Q ( t )  Q ( t )  Q ( t ) A. Basic data
 B min BG B max The VPP is designed to consist of wind turbines, PV units,
where the minimum and maximum power limits for the cogeneration units, GBs, and ES devices aggregated together,
system to purchase power from the superior grid are denoted and an industrial park is selected for the research. A dispatch
as 𝑃Bmin (𝑡) and 𝑃Bmax (𝑡) ; 𝑃Smin (𝑡) and 𝑃Smax (𝑡) refer to cycle is set as 24 h, and a unit dispatch period is set as 1 h. In
the minimum and maximum power limits for the system to this example, the low calorific value of natural gas is 9.7
sell electricity to the upper grid, respectively; and 𝑄Bmin (𝑡) kWh/m3, and GHV is 35.54 MJ/m3. The wind-power
and 𝑄Bmax (𝑡) refer to the upper and lower limit of interaction prediction curves for a typical day is shown in Figure 3 , and
between the superior gas network and the system during the electrical, thermal and gas load forecasting curves is
period 𝑡, respectively. shown in Figure 4.
(3) Unit equipment constraints
 PGTmin (t )  PGTele (t )  PGT max (t )

 PGBmin (t )  PGB (t )  PGB max (t )
heat
(30)

 PP2Gmin (t )  PP2G (t )  PP2G max (t )
gas

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

1400
Wind power forecast
600 Electrical load
Photovoltaic predicted power
Heat load
1200 Gas load
500 Wind power forecast Electrical load

400 1000
Heat load
Power/kW

Power/kW
300 800

200
600

100 Photovoltaic predicted

power 400 Gas load
0
200
0 4 8 12 16 20 24 0 4 8 12 16 20 24
Time/h Time/h
Fig. 3. Typical daily wind and light power prediction curve Fig. 4. Typical electric, heat and gas load forecasting curve

TABLE I
EQUIPMENT PARAMETERS
Type of Minimum output Maximum output power
Efficiency Maintenance costs (yuan/kWh)
equipment power (kW) (kW)
CHP 0 1000 0.45 0.15

GB 300 600 0.85 0.18

P2G 300 600 0.6 0.15

TABLE II
PARAMETERS OF ENERGY STORAGE EQUIPMENT
Type of Maximum power for Maximum power of Self-consumption Maintenance costs
equipment charging (kW) discharging (kW) rate of equipment (yuan/kWh)

ES 300 400 0.01 0.018

HS 600 700 0.05 0.016

GS 500 600 0.02 0.017

TABLE Ⅲ
DEMAND RESPONSE LOAD COMPENSATION PRICE
Type of load Cost (yuan/kWh)

Transferable electrical/heat load 0.16

Reducible electrical/heat load 0.14

Substitutable electrical load 0.12

In the example, the unit maintenance costs of wind power

B. Scenario setting and analysis of result
and PV units are both taken as 0.23 yuan/kWh. In the cost of
To verify that VPP aggregated in this paper is more
carbon emissions, 𝑎1 , 𝑎2 and 𝑎3 are taken as 0.0532,
economical and clean, the following three scenarios are
190.12 and 13175.4, respectively. 𝛾 is taken as 0.15, and 𝛽 established to compare and analyse the system operation
is taken as 0.798. The technical parameters of each unit of results under different scenarios.
equipment are shown in Tables Ⅰ and Ⅱ. Scenario 1: The VPP is operated in a conventional manner
The compensation prices for the demand response loads in for a subsequent visual comparison with the VPP operation
the example are shown in Table Ⅲ. The system uses the local in Scenarios 2 and 3.
time-of-use tariff for the purchase and sale of electricity with Scenario 2: Consideration of the electrical and thermal load
the superior grid, with a peak hour purchase price of 0.95 demand responses based on Scenario 1
yuan/kWh and a sale price of 0.54 yuan/kWh from 11:00– Scenario 3: Based on Scenario 2, the hydrogen obtained in
21:00, an ordinary hour purchase price of 0.65 yuan/kWh and the first stage of the P2G conversion process is stored in a
a sale price of 0.47 yuan/kWh from 8:00–10:00 and 22:00– storage tank to supply the load, and the remaining hydrogen
24:00, and a peak hour purchase price of 0.38 yuan/kWh and is then subjected to a methanation reaction (i.e., the model
a sale price of 0.33 yuan/kWh from 1:00–7:00.

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

developed in this paper). under the selected typical day under different scenarios
(1) Impacts of the VPP build method on costs allows the system operating costs for each scenario to be
The simulation study of the VPP in this industrial park obtained, as shown in Table Ⅳ.

TABLE Ⅳ
EACH SCENARIO COST COMPARISON
Cost category/yuan Scenario 1 Scenario 2 Scenario 3

Cost of system interaction with the parent grid/yuan 1608.1 458.6 634.7

System purchase of external gas cost/yuan 20536 20399 10711.4

Equipment maintenance cost/yuan 9352.8 9130.9 10562.5

Demand response load compensation cost/yuan 0 415 409.8

Cost of carbon emissions/yuan 5685.6 5632.8 5013.5

Total running costs/yuan 37182.5 36036.3 27331.9

According to the analysis in Table Ⅳ, since Scenario 1 is the use of wind power to produce hydrogen, which provides
operated in a traditional manner without considering the a certain capacity of hydrogen for the system, and this method
integrated demand responses of electricity and heat and the also reduces the danger in the process of transporting
two-stage P2G conversion process, the system purchases hydrogen and improves the efficiency of the multienergy
more energy and has the highest electricity purchase costs and coupling within the VPP. In addition, the total cost of
sale interactions with the superior grid, but there is no demand Scenario 3 is reduced by 26.5% and 24.2% compared to
response load compensation cost. On the basis of Scenario 1, Scenario 1 and Scenario 2, respectively, so the optimisation
Scenario 2 addresses the combined demand response for heat model proposed in this paper, i.e., Scenario 3 has better
and electricity, so Scenario 2 has an additional demand economics in the operation of the VPP.
response load compensation cost of $415 compared to (2) Analysis of the P2G operation
Scenario 1. However, due to the addition of demand response As seen in Figure 5, during the periods of 1:00–7:00 and
loads, the baseline load decreases during peak periods, 22:00–24:00, when wind power is high and the price of
reducing the pressure on the generating units compared to electricity is low, the P2G output is high, while during the
scenarios where demand response is not considered, other periods from 7:00 to 22:00, the wind power output
effectively relieving the internal load of the virtual power decreases and the P2G output is basically nonexistent,
plant and making the system in Scenario 2 reduce the cost of converting electricity from the low-cost periods into other
electricity purchase and sale interaction with the superior grid forms of energy to supply the load during the high-cost
by 70.5% compared to Scenario 1. Scenario 3 has the highest periods. The hydrogen produced in the electrolyser is fed into
operation and maintenance cost because its two-stage the methane reactor to produce natural gas to supply the gas
electricity to gas equipment is considered compared to load, and the hydrogen produced in most of the remaining
Scenario 2. Due to the operation of the electrolyser and other hours is stored directly in the hydrogen storage tank to supply
equipment, its purchased electricity cost is slightly higher, but the hydrogen load, reducing the energy loss caused by
the purchased gas cost is substantially lower, mainly due to multistage conversion compared to methanation again
.
2000
1800
Methane reactor output
Electrolyser output
1600
Hydrogen storage tanks
1400
Power/kW

1200
1000
800
600
400
200
0
0 4 8 12 16 20 24
Time/h
Fig. 5. Output of the electrolyser and the methane reactor

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

3000
Grid interactive power ES charging and discharging capacity
Photovoltaic power output Electrolyser power consumption
2500 CHP electrical output Electrical load
Wind turbine output
2000

1500
Power/kW

1000

500

−500

−1000

−1500
0 4 8 12 16 20 24
Time/h
Fig. 6. Scenario 3 power balance operation result

2000
HS charge and discharge heat
GB heat output
1500 CHP heat output
Heat load
1000
Power/kW

500

−500

−1000
0 4 8 12 16 20 24
Time/h
Fig. 7. Scenario 3 thermal balance operation result

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

Gas purchase GB gas consumption

3000 GS filling and discharging gas volumes GT gas consumption
Methane reactor output Gas load

2000
Power/kW

1000

−1000

−2000
0 4 8 12 16 20 24
Time/h
Fig. 8. Scenario 3 natural gas balance operation result

(3) Scheduling optimisation results analysis met by the thermal power output of the CHP units and the GB
Figures 6–8 show the results of the electricity balance, heat output, while the HS system exerts heat, reducing the
balance and gas balances under Scenario 3. operating costs of the system.
As seen in Figure 6, since the incorporated demand The natural gas balance optimisation results in Scenario 3
response of electricity and heat, as well as the two-stage P2G are shown in Figure 8. The gas load demand is mainly met by
conversion process, are taken into account in Scenario3, the the system purchasing external gas throughout the 24-hour
system load power demand is met at this time by the system's day, while the CHP unit consumes more gas to output
purchase of energy out from upper power system, the electrical and thermal energy to meet the electrical and
electrical power output of the CHP units, and the outputs of thermal loads due to its low power output cost. When the
the wind and PV units. During the 1:00–7:00 and 22:00– system reaches its maximum purchase of gas from the upper
24:00 periods, when wind power is abundant and electricity grid, the output is supplemented by the GS system and
prices are low, the P2G conversion output is higher; however, methane reactors. During the period 1:00–7:00, when
during the other periods, the P2G conversion output is electricity prices are low, the methane reactor starts working
basically nonexistent, converting the electricity during the and produces natural gas by consuming electricity, and the
low-cost hours into other forms of energy to supply the load excess gas is stored by the GS system. During the hours of
during the high-cost hours. This phenomenon reduces the 11:00–21:00, when electricity prices are high, the methane
system operating costs. As the wind turbines produce more reactor, which more expensively produces gas, stops working,
power at night, the excess power is still stored by the ES and the gas load demand is mainly met by the system
system at this time. Since the PV unit output is higher during purchasing gas from external sources and by the GS bleeding
the day and the electrical load after the demand response is gas. The various pieces of equipment work synergistically to
not much different from that at night, the system prioritises reduce system operating costs.
using the wind turbine generation during the daytime period (4) Sensitivity analysis
to fully use the clean energy output. During the 11:00–21:00 Different weights 𝑟1 and 𝑟2 are assigned to the system
period at the peak of the electricity tariff, the ES system is operating cost 𝐶1 and the carbon emission cost 𝐶2 ,
used to meet the electrical load demand by discharging. respectively, as shown in Equation (40). Their impacts on the
The results of the system heat balance optimisation in system are analysed. The weight coefficients are used to
Scenario 3 are shown in Figure 7. During the 1:00–7:00 indicate the amount of attention being given, and 𝑟1 + 𝑟2 = 1.
period, the heat load demand is mainly met by the CHP output When 𝑟1 > 𝑟2 , more attention is given to the system's
because it is at a low electricity price at this time; when the operating cost, i.e., economy; when 𝑟1 < 𝑟2 , more attention is
CHP system reaches its output limit, the GB output is used, given to the system's carbon emission cost, i.e.,
and the excess heat generated is stored by the HS system. environmental friendliness. Figure 9 shows the variation
During the 8:00–10:00 period, the EB system has sufficient curves of economic indicators, environmental indicators and
power input because the scenic units are out of power; thus, the total costs for different weighting factors.
the EB system remains in working condition during this
min C ' = λ1C1 + λ2C2 (36)
period. From 11:00–21:00, due to the higher electricity prices
and the lower gas unit prices, the heat load demand is mainly where 𝐶 ′ is the total cost after adding the weighting factor.

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

60000 35000
Running costs
55000
Cost of carbon emissions
50000 Total cost Total cost 34500
45000
40000 34000

Total cost/yuan
Costs/yuan

35000
30000 33500
25000
20000 33000
15000
10000 32500
5000
0 32000
1.0,0.0 0.9,0.1 0.8,0.2 0.7,0.3 0.6,0.4 0.5,0.5 0.4,0.6 0.3,0.7
Weighting factor
Fig. 9 Comparison of indicators under different weight coefficients

Different coefficient settings affect the final optimisation plant system in the period of high cost is met, the economic
results: When 𝑟2 is large, decision-makers pay more attention goal is achieved, and the energy utilization rate and low-
to environmental protection, the carbon emission cost carbon cleanliness are improved.
decreases, and the operating cost significantly increases, and (2) In the coordinated electricity-heat-gas optimal control
vice versa. Figure 9 shows that when the weight coefficient of virtual power plants, consideration of electricity and heat
changes from extreme attention to economic benefits (𝑟1 =1, load demand response and low-carbon objectives can
𝑟2 =0) to slight attention to economic benefits (𝑟1 =0.7, 𝑟2 =0.3), promote multienergy synergistic optimisation, ensuring a
with the reduction in attention to the economy, the operating symmetrical power balance between system electricity, heat
cost increases, and the carbon emission cost decreases. and gas, improving the stability and flexibility of system
However, paying excessive attention to carbon emissions operation, and facilitating energy flow and interaction
leads to a significant increase in operating costs; at this time, between system sources, storage and load.
the reduction in carbon emissions is very limited. Therefore, (3) By introducing weighting coefficients for the
the coefficient range on the right side of the coordinate axis sensitivity analysis of the system to form a comparison of
has no actual reference value, and the cost performance is optimal allocation options under different scenarios, the
very low. In summary, the ranges of reasonable coefficients results show that different optimisation preference settings
are as follows: (𝑟1 =0.7, 𝑟2 =0.3) ~ (𝑟1 =0.4, 𝑟2 =0.6). can significantly affect the system allocation results. The
reasonable dispatch coefficients for the economic and
VI. CONCLUSION environmental aspects of the VPP studied in the paper range
In this paper, by introducing the multienergy coupled from (𝑟1 =0.7, 𝑟2 =0.3) ~ (𝑟1 =0.4, 𝑟2 =0.6). In the future, by
operation of power-to-gas and virtual power plants with considering the degree of concern for different aspects in the
internal CHP units and gas turbines and constructing a model optimisation process, suitable weighting coefficients can be
for the optimal dispatch operation of participating VPPs, with selected to achieve flexible dispatch control of economic and
the objective of minimising the sum of operating costs and environmental indicators.
carbon emission costs, combined with arithmetic simulations,
the following conclusions are drawn. REFERENCES
(1) The electrolyser, methane reactor and hydrogen storage [1] LAI S X, LU J P, LUO X Y, et al, "Carbon emission evaluation model
and carbon reduction strategies for newly urbanized areas, "
tank constitute the mutual coupling between electricity, Sustainable Production and Consumption, vol.31, pp. 13-25, 2022.
hydrogen and natural gas within the virtual power plant, fully [2] Zhang C , Xu Y , Dong Z Y . "Robustly Coordinated Operation of A
considering the difference in conversion efficiency between Multi-Energy Micro-Grid in Grid-Connected and Islanded Modes
under Uncertainties, " IEEE Transactions on Sustainable Energy,
the two stages of P2G and taking into account the future pp.1-1, 2019.
development trend of integrated energy systems, converting [3] Light Zaglago. Frank-K. Dzokoto, and Lynda Ankrah, "Challenges to
wind energy into hydrogen required by the system, reducing Smart Grid Technology," Lecture Notes in Engineering and Computer
Science: Proceedings of The World Congress on Engineering 2021, 7-
the energy loss caused by step conversion; To a certain extent,
9 July, 2021, London, U.K., pp175-180
through energy transfer, the load demand of the virtual power

Volume 31, Issue 1: March 2023

 Engineering Letters, 31:1, EL_31_1_09
______________________________________________________________________________________

[4] Miaomiao Wang, Qingwen Luo, Lulu Kuang, and Xiaoxi Zhu,
"Optimized Rolling Grey Model for Electricity Consumption and
Power Generation Prediction of China," IAENG International Journal
of Applied Mathematics, vol. 49, no.4, pp.577-587, 2019.
[5] Rahimiyan M , Baringo L , “Strategic Bidding for a Virtual Power
Plant in the Day-Ahead and Real-Time Markets: A Price-Taker
Robust Optimization Approach,” IEEE Transactions on Power
Systems, vol.31, no.4, pp. 2676-2687, 2016.
[6] R. Liu, Y. Liu and Z. Jing, "Impact of Industrial Virtual Power Plant
on Renewable Energy Integration," 2020 IEEE/IAS Industrial and
Commercial Power System Asia (I&CPS Asia), pp. 1198-1202, 2020.
[7] João Soares, Mohammad Ali Fotouhi Ghazvini, Zita Vale, P.B. de
Moura Oliveira, “A multi-objective model for the day-ahead energy
resource scheduling of a smart grid with high penetration of sensitive
loads,” Applied Energy, vol 162, pp 1074-1088, 2016.
[8] Y. Wang, X. Ai, Z. Tan, L. Yan and S. Liu, “Interactive Dispatch
Modes and Bidding Strategy of Multiple Virtual Power Plants Based
on Demand Response and Game Theory,” IEEE Transactions on
Smart Grid, vol. 7, no. 1, pp. 510-519, 2016.
[9] X. Zhang, M. Shahidehpour, A. Alabdulwahab and A. Abusorrah,
“Hourly Electricity Demand Response in the Stochastic Day-Ahead
Scheduling of Coordinated Electricity and Natural Gas Networks,”
IEEE Transactions on Power Systems, vol. 31, no. 1, pp. 592-601,
2016.
[10] W J Niu, Y Li, B B Wang, “Virtual power plant modelling for demand
response considering uncertainty,” Proceedings of the CSEE, vol. 34,
no. 22, pp. 3630-3637, 2014.
[11] T. F. Ma, J. Y. Wu, L. L. Hao, Y. J. Li, H. G. Yan, D. Z. Li and S. S.
Chen, “Analysis of energy flow modeling and optimal operation of
micro-energy networks based on energy hubs”. Power System
Technology, vol. 42, no. 01, pp. 179-186, 2018.
[12] Y. Y. Liu, C. W. Jiang, S. M. Tan, J. Z. Hu and Q. S. Li, “Optimal
scheduling strategy for virtual power plants considering risk-adjusted
capital return threshold constraint,” Proceedings of the CSEE, vol. 36,
no. 17, pp. 4617-4627, 2016.
[13] T. S. Li, J. L. Liu and M. Y. Wu, “Optimal scheduling of virtual power
plants accounting for environmental costs,” Heilongjiang Electric
Power, vol.27, no.1, pp.5-12, 2015.
[14] Y Zhang, Q Xu and L Yang, “A study on the cooperation space
between virtual power plants and conventional units considering
carbon trading,” Energy Engineering, vol. 5, pp. 1-7, 2016,
[15] G. Q. Sun, Y. Z. Zhou, Z. N. Wei, et al, “Robust optimization bidding
model for virtual power plants under carbon emission constraints,”
Proceedings of the CSEE, vol. 37, no.11, pp. 3118-3128, 2017.
[16] Albadi M H , El-Saa Da Ny E F , “A summary of demand response in
electricity markets,” Electric Power Systems Research, vol. 78, no. 11,
pp. 1989-1996, 2008.
[17] Gorostiza, Francisco Sanchez ; Gonzalez-Longatt, Francisco,
“Optimised TSO–DSO interaction in unbalanced networks through
frequency-responsive EV clusters in virtual power plants,” IET
Generation Transmission & Distribution, vol. 14, no. 21, pp. 4908-
4917, 2020.
[18] D Kaczorowska, Rezmer J , Jasinski M , et al, “A Case Study on
Battery Energy Storage System in a Virtual Power Plant: Defining
Charging and Discharging Characteristics”. Energies, vol.13, no.24,
pp. 121-129, 2020.
[19] R. Gao, H. X. Guo, R. H. Zhang, T. Mao, Q. Y. Xu, B. R. Zhou and P.
Yang. “A Two-Stage Dispatch Mechanism for Virtual Power Plant
Utilizing the CVaR Theory in the Electricity Spot Market,” Energies,
vol. 12, no.17, pp. 1020-1029, 2019.
[20] C Y Ma, C F Dong, et al, “Short-term trading and optimal operating
strategies for commercial virtual power plants accounting for
stochastic factors,” Power System Technology, vol. 40, no. 05, pp.
1543-1549, 2016.
[21] Fan S , Liu J , Wu Q , et al. Optimal Coordination of Virtual Power
Plant with Photovoltaics and Electric Vehicles: A Temporally
Coupled Distributed Online Algorithm[J]. Applied Energy, 2020, 277..
[22] Chen X Y, Kang C Q, O'Malley M , et al. “Increasing the Flexibility
of Combined Heat and Power for Wind Power Integration in China:
Modeling and Implications,” IEEE Transactions on Power Systems,
vol.30, no.4, pp. 1848-1857, 2015.
[23] Alhamed K, Dincer I, Lund H, et al. “Exergoeconomic analysis and
optimization of a solar energy-based integrated system with oxy-
combustion for combined power cycle and carbon capturing,” Energy,
vol.250, pp.123814, 2022.
[24] Z L Wei, S Yu, et al, “Concept and development of the micro-energy
network,” Automation Of Electric Power Systems, vol.37, no.13, pp.
1-9, 2013.

Volume 31, Issue 1: March 2023