Shared Trading Strategy of Multiple Microgrids Considering Joint Carbon and Green Certificate Mechanism

Abstract

With a background of carbon peak and neutrality, the economic and environmental requirements are increasing for microgrids. In view of the problem of energy wastage and conflicts of interest among multiple microgrid integrated energy systems, it is important to study the operation optimization of microgrid clusters while considering the sharing and trading of both carbon emissions and green certificates. In this study, a Stackelberg game mechanism is applied, in which the microgrid operator is the leader and its subscribers are the followers, forming a master–slave interaction model. Following this, breaking the trading barriers of energy and various policy markets, the joint carbon and green certificate trading mechanism is proposed. Moreover, a mutually beneficial shared trading model of multi-microgrids considering coupled energy and carbon and green certificate trading is proposed to avoid the problem of double counting of environmental attributes. In addition, a cooperative sharing center is assumed to propose a flexible multi-resource sharing price mechanism. It guides each microgrid operator to conduct internal multi-resource sharing trading, so as to reduce the daily operating costs of energy supplying entities in the cooperative system of multiple microgrids, effectively reduce carbon emissions, and improve the balance of network group mutual aid. According to the simulation results of an illustrative example, the proposed trading strategy can effectively unlock the potential of resource sharing and mutual aid within multi-microgrids and improve the economy and carbon reduction effects of the overall system.

1. Introduction

With the deterioration of the natural environment and the intensification of global warming, studies related to carbon emission reduction are paid more and more attention. Most of the current studies focused on analyzing the complex relationships between carbon emissions and economic development, energy structure, and social changes in different regions over longtime scales, so as to obtain the relationships between key influencing factors and carbon emissions, as well as provide a reliable basis for relevant policy formulation [1,2,3,4]. In addition, some studies have been reported on the carbon emissions reduction measures from the viewpoints of supply chain management [5,6,7].

The energy sector is the main source of carbon emissions, and thus is also the focus of carbon emissions reduction. With the development of science and technology, the use of energy has been refined and the energy supply has evolved from the centralized form to the distributed one. Under these backgrounds, as an autonomous unit that integrates energy production, transmission, storage, and consumption, the microgrid (MG) is paid more and more attention [8,9]. Currently, with the link between the macro-grid and consumers, the multi-energy coupling characteristics of microgrids are becoming more and more prominent, which is important for promoting renewable energy development and achieving the carbon peak and carbon neutrality [10,11]. An individual microgrid is susceptible to resource uncertainty, making it difficult to ensure the balance between supply and demand over time. With the gradual increase of penetration rate of microgrids in distribution networks, multiple microgrids in a region can form a multi-microgrid (MMG). In this way, the energy can be shared and traded among adjacent microgrids to enhance the stability and reliability of multi-microgrid operation, as well as promote the local consumption of renewable energy [12,13,14].

Many works have been reported about energy sharing and trading among microgrids. Hua et al. [15] and Rui et al. [16] designed a Stackelberg game model between microgrid cluster operators and microgrid operators, which could improve the economic efficiency of regional microgrids. Alizadeh et al. [17] proposed a cooperative game method to determine the MG’s daily dispatch and energy transaction. Wang et al. [18] constructed a bi-level optimization framework to coordinate the transactive energy of microgrids with the operation of distribution systems. Zhang et al. [19] established a cooperative operation model of multi-microgrids based on Nash negotiation theory, which could protect the privacy of the subject effectively, while the operation optimization of each microgrid still had to obey the overall interests and did not reflect the characteristics of each microgrid as an independent interest subject.

In recent years, the reduction targets of carbon emissions have become more and more urgent [20]. To speed up the green transformation of the energy and electricity system, corresponding policies are of vital importance. Therefore, the development of carbon and green certificate trading markets is essential to facilitate renewable energy consumption and emissions reduction of main majors. China has successively launched Green Certificate Trading (GCT) [21] and a nationwide unified Carbon Emission Trading (CET) market [22]. In the future, the electricity market, the carbon market and the green certificate market will effectively form a joint force and develop in a synergistic manner to promote the achievement of the “double-carbon” goal [23].

Currently, the synergistic development of the electricity, carbon, and green certificate markets is an urgent issue that has received much attention from all parties. Sun et al. [24] proposed a comprehensive energy system model that includes both carbon and green certificate trading mechanisms, as well as the demand response. Yan et al. [25] designed a dispatching optimization model for virtual power plants (VPPs) considering the carbon and green certificate trading. Wang et al. [26] and Lu et al. [27] proposed a multi-entity trading mechanism considering electricity–carbon coupling, which could enhance the economic and low-carbon benefits of multi-entities effectively. Suo et al. [28] established a day-ahead optimal dispatch model of integrated energy systems considering joint carbon–green certificate trading, but did not consider the problem of double calculation of environmental attributes of renewable energies. According to the above discussion, few existing studies considered the impact of the joint effect of carbon and green certificate trading on the optimal operation of the system. In addition, the solution of double counting of environmental attributes of renewable energy and its impact were paid little attention.

In this study, a dispatch strategy considering multi-resource sharing transactions among multiple microgrids based on a joint carbon-green certificate mechanism is proposed. Microgrid operators (MGOs) as participants instead of microgrids are involved in the market transactions. Firstly, a Stackelberg game-based pricing model for energy transactions between the MGO and users for a single MGO is proposed. Secondly, breaking down barriers to energy and various policy market transactions, a multi-resource sharing and trading mechanism among multiple microgrids is designed by considering CET and GCT coupling in an integrated manner. The cooperative sharing center establishes a flexible price trading mechanism to guide multi-resource sharing transactions among multi-MGOs. Finally, a numerical study is implemented to verify the feasibility of the proposed model and calculation method.

2. Materials and Methods

2.1. Shared Trading Framework of Multi-Microgrids

Figure 1 illustrates the framework for multi-resource sharing and trading among multi-microgrids under the joint carbon and green certificate mechanism. The main actors involved in the shared market include individual microgrid operators, customers, cooperative sharing centers and various types of markets (energy market/carbon market/green certificate market). The shared trading model between multiple resources across multi-microgrids breaks down the trading barriers in energy and various policy markets by integrating CET and GCT joint trading, so as to avoid the thorny issue of double counting of environmental attributes.

A fair and reasonable pricing mechanism is important for the cooperative sharing center to set trading prices and stimulate the trading among MGOs. In this study, a supply–demand ratio-based pricing model is employed by the cooperative sharing center to set an internal shared trading price. It is preferable to the external trading price based on the surplus and deficit of resources uploaded by each MGO.

The MGO is the leader in the transaction, supplying energy to the follower (the customer) and offering a price for the energy traded. Following this, the customer adjusts its energy consumption according to the supply price. In this way, both sides act in the direction of maximizing their respective interests and ultimately reach a mutually optimal equilibrium.

2.2. Multi-Microgrid Model

Due to the different features of various types of microgrids, it is possible to share energy and other assets between microgrids of the same level to achieve better economic benefits and emission reduction. In this study, three types of microgrids, namely industrial microgrids, commercial microgrids, and residential microgrids, are employed for analysis.

Each microgrid may be equipped with various equipment, including photovoltaic (PV), wind turbine (WT), combined heat and power (CHP), gas boiler (GB), electric boiler (EB), electric chiller (EC), absorption chiller (AC), electric vehicle (EV), P2G unit, natural gas/heat/electric storage equipment (NSE/HSE/ESE), etc.

2.2.1. CHP

A CHP unit can generate both electricity and heat by burning gas with a relatively high overall efficiency. It cannot generate more power than its installed capacity. In addition, the ramping ability of the unit needs to be constrained. Details are shown in Equation (1).

$$\left\{\begin{array}{l}{P}_{i,e,chp}^t=P_{i,g,chp}^t{\eta }_{e,chp}\\ P_{i,h,chp}^t=P_{i,g,chp}^t{\eta }_{h,chp}\\ 0\le P_{i,g,chp}^t\le P_{i,g,chp}^{\max }\\ \Delta P_{i,g,chp}^{\min }\le P_{i,g,chp}^{t+1}-P_{i,g,chp}^t\le \Delta P_{i,chp}^{\max }\end{array}\right.$$

(1)

where $I$ represents the collection of all MGO, $I=\left\{1,2,\dots ,n\right\}$, $i\in I$; $t$ represents the time interval; $P_{i,e,chp}^t$, $P_{i,h,chp}^t$, and $P_{i,g,chp}^t$ denote electricity production, heat production, and gas consumption of the CHP unit within MGO $i$, respectively; ${\eta }_{e,chp}$ and ${\eta }_{h,chp}$ represent the efficiencies of electricity and heat production, respectively; $P_{i,g,chp}^{\max }$ donates the maximum gas consumption of the CHP; $\Delta P_{i,g,chp}^{\min }$ and $\Delta P_{i,g,chp}^{\max }$ are the upper and lower bounds for ramp rate limits, respectively.

2.2.2. GB

As a heat generator, the energy conversion and consumption constraint of a GB is described in Equation (2).

$$\left\{\begin{array}{l}{P}_{i,h,GB}^t=P_{i,g,GB}^t\cdot {\eta }_{GB}\\ 0\le P_{i,g,GB}^t\le P_{i,g,GB}^{\max }\end{array}\right.$$

(2)

where $P_{i,h,GB}^t$ and $P_{i,g,GB}^t$ are heat generation and gas consumption of the GB, respectively; ${\eta }_{GB}$ represents the efficiency of the GB; $P_{i,g,GB}^{\max }$ denotes the maximum gas consumption.

2.2.3. EB and EC

Due to the similar mathematical model of an EB and an EC, $\chi =\left\{EB,EC\right\}$ is defined as the set of these two types of energy conversion. The models are shown as:

$$\left\{\begin{array}{l}{P}_{i,x}^t=P_{i,in,x}^t{\eta }_x\\ 0\le P_{i,in,x}^t\le P_{i,in,x}^{\max }\end{array}\right.$$

(3)

where $P_{i,in,x}^t$ and $P_{i,x}^t$ are electricity consumption and energy production of the equipment $\chi$; $P_{i,in,x}^{\max }$ denotes the maximum electricity consumption.

2.2.4. ESE, HSE, and NSE

Due to the similar mathematical model of the ESE, HSE, and NSE, $S=\left\{ESE,HSE,NSE\right\}$ is defined as the set of these three types of storage. The energy storage dynamics and constraints are shown as:

$$\left\{\begin{array}{l}{E}_{i,s}^{t+1}=E_{i,s}^t-\frac{P_{i,dis,s}^t\Delta t}{{\eta }_{i,dis,s}}+\frac{P_{i,ch,s}^t\Delta t}{{\eta }_{i,ch,s}}\\ D_{i,dis,s}^t+D_{i,ch,s}^t\le 1\end{array}\right.$$

(4)

where $E_{i,s}^t$, $P_{i,ch,s}^t$, and $P_{i,dis,s}^t$ denote the amount of stored energy, charging rate, and discharging rate of the storage unit $s\in S$, respectively; ${\eta }_{i,ch,s}$ and ${\eta }_{i,dis,s}$ denote the charging and discharging efficiencies of the storage unit.

Two binary variables $D_{i,ch,s}^t$ and $D_{i,dis,s}^t$ are introduced to avoid the charging and discharging happening simultaneously.

$$\left\{\begin{array}{l}0\le P_{i,dis,\lambda }^t\le D_{i,dis,s}^t{P}_{i,dis,\lambda }^{\max }\\ 0\le P_{i,ch,\lambda }^t\le D_{i,ch,s}^t{P}_{i,ch,\lambda }^{\max }\\ 0\le D_{i,dis,s}^t+D_{i,ch,s}^t\le 1\end{array}\right.$$

(5)

$$\left\{\begin{array}{l}{E}_{i,\lambda }^{\min }\le E_{i,\lambda }^t\le E_{i,\lambda }^{\max }\\ E_{i,\lambda }^1=E_{i,\lambda }^{24}\end{array}\right.$$

(6)

where $P_{i,ch,\lambda }^{\max }$ and $P_{i,dis,\lambda }^{\max }$ are the maximum charging and discharging rates of storage $s$, respectively; $E_{i,\lambda }^{\min }$ and $E_{i,\lambda }^{\max }$ are the minimum and maximum state-of-charge (SOC), respectively.

2.3. Joint Carbon and Green Certificates Trading Model

2.3.1. CET Model

The CET is a carbon emission allowance issued by the government or the relevant regulatory authority to the emission control entity. When the actual carbon emissions produced by the emission control entity are lower than the allowance, the balance can be sold in the carbon market; conversely, the excess allowance needs to be purchased. With a high proportion of renewable energy connected to the grid, the environmental attributes of the renewable energy in the purchased electricity cannot be ignored, so the renewable energy generation component of the purchased electricity needs to be taken into account when accounting for carbon emissions on the demand side.

There are various ways to allocate carbon allowances without compensation. In this study, the baseline method was employed to calculate carbon allowances based on the amount of electricity and heat supplied by the emission control entity. The amount of carbon emissions allocated is shown in Equation (7).

$$E{m}_i^t={\gamma }_i^E\left({\epsilon }_e{P}_{i,e}^t+{\epsilon }_h{P}_{i,h}^t\right)$$

(7)

where ${\gamma }_i^E$ represents the proportion of free carbon allowances, acting as a further tightening of the amount of carbon allowances issued; ${\epsilon }_e$ and ${\epsilon }_h$ are emission allowance factors per unit of electricity and heat, respectively; $P_{i,e}^t$ and $P_{i,h}^t$ are total electricity and heat demand, respectively.

The amount of actual carbon emissions is shown as:

$$E{s}_i^t=E_{i,g}^t+E_{i,b}^t$$

(8)

$$E_{i,g}^t=Q_{i,g}^t{H}_g{F}_{C{O}_2,g}$$

(9)

$$E_{i,b}^t=\left(1-\gamma re{s}_i^t\right)P_{i,e,buy}^t{F}_{C{O}_2,e}+P_{i,h,buy}^t{F}_{C{O}_2,h}$$

(10)

where $E{s}_i^t$ denotes the amount of actual carbon emission; $E_{i,g}^t$ and $E_{i,b}^t$ represent carbon emissions from natural gas consumption and emissions from energy purchased from the upper energy network of MGO $i$ in time interval $t$, respectively; $Q_{i,g}^t$ represents the total gas consumption; $H_g$ represents the average low calorific value of natural gas; $F_{C{O}_2,g}$ represents the carbon emission factors based on the lowest calorific value of natural gas; $\gamma re{s}_i^t$ represents the percentage of renewable energy in external electricity purchases; $P_{i,e,buy}^t$ and $P_{i,h,buy}^t$ represent electricity and heat purchased from the superior network, respectively; $F_{C{O}_2,e}^t$ and $F_{C{O}_2,h}^t$ are marginal carbon emission factors of the electricity grid and the heat network, respectively.

The amount of carbon trading involved is shown in Equation (11).

$$Ece{t}_i^t=E{s}_i^t-E{m}_i^t$$

(11)

where $Ece{t}_i^t$ denotes the amount of carbon trading involved.

Based on the amount of carbon trading involved, a carbon trading model is developed as follows:

$$Cce{t}_i^t=c_c^{t}Ece{t}_i^t$$

(12)

where $Cce{t}_i^t$ is the carbon trading cost; $c_c^t$ is the carbon trading price.

2.3.2. GCT Model

The Green Certificate is a certification of the non-energy attributes of green electricity, and its main purpose is to reflect the marginal social benefits of green electricity such as fossil energy substitution and environmental protection. Similar to the CET, the government assigns a weighting to the responsibility of renewable energy consumption to the subject. When the subject’s actual green certificate volume is in excess of the quota quantity, it can be taken to the market for sale; conversely, it is necessary to satisfy the assessment requirements by purchasing the corresponding number of green certificates.

In this study, two variables $Nrp{s}_i^t$ and $Nre{s}_i^t$ are introduced to describe the amounts of green certificates required to be consumed by the MGO and green certificates obtained for electricity generated by its internal renewable energy units, which are described in Equations (13) and (14), respectively.

$$Nrp{s}_i^t=\omega \left[1+\delta \left({\zeta }_{av}-\xi \right)\right]P_{i,e}^t/1000$$

(13)

$$Nre{s}_i^t=P_{i,res}^t/1000$$

(14)

where $\omega$ is the renewable energy quota factor; $\delta$ is the historical renewable energy quota completion weight; ${\zeta }_{av}$ is the average of historical renewable energy quota completions; $\xi$ is the historical renewable energy quota completions; $P_{i,res}^t$ is the renewable energy contribution from participating green certificates approved.

2.3.3. Joint Carbon–Green Certificate Model

The CET and GCT are designed to promote carbon emission reduction and energy restructuring from the perspectives of greenhouse gas emission control and energy restructuring, respectively. Currently, the two markets are running in a parallel relationship in China. However, both carbon allowances and green certificates can indicate the environmental rights and benefits. The same object should be certified according to the principle of “uniqueness of environmental rights and benefits certification” to avoid double counting of benefits. Therefore, in order to clarify the allocation of the environmental attributes of renewable energies, the link between green certificates and carbon allowances can be constructed in a reasonable way, linking the two markets of CET and GCT. Details are shown in Equation (15).

$$Eun{s}_i^t=E_{i,g}^t+E_{i,b}^t+Egc{t}_i^t$$

(15)

$$Egc{t}_i^t=P_{i,res}^t{F}_{C{O}_2,e}^t$$

(16)

where $Eun{s}_i^t$ indicates the carbon emission of MGO $i$ under the joint emission–green certificate trading mechanism; $Egc{t}_i^t$ is the carbon emission offset of renewable energy sources participating in the green certificate certification.

2.4. Hierarchical Optimal Trading Model of Multi-Microgrids

2.4.1. Cooperative Shared Decision-Making Model in Upper Level

The flexible price mechanism proposed in this study is a supply and demand rate (SDR) model to ensure the fairness in the pricing of energy, carbon allowances, and green certificates within the alliance.

The ratio of supply to demand for each traded resource within the cooperative alliance is the ratio of the sum of the surplus to the sum of the shortfall after autonomous optimal scheduling of each MGO, as shown in the following equations:

$$R_k^t=\frac{S{P}_{k,\sup }^t}{S{P}_{k,de}^t}$$

(17)

$$\left\{\begin{array}{l}S{P}_{k,\sup }^t={\displaystyle \sum _{i=1}^I{P}_{i,k,sell}^t}\\ S{P}_{k,de}^t={\displaystyle \sum _{i=1}^I{P}_{i,k,buy}^t}\end{array}\right.$$

(18)

where $K=\left\{e,h,CET,GCT\right\}$, $k\in K$; $R_k^t$ is the ratio of supply to demand for resource $k$ in time interval $t$, respectively; $P_{k,buy}^t$ and $P_{k,sell}^t$ are the volume of external purchases and sales of resource $k$ by MGO $i$ in time interval $t$, respectively.

The intra-cooperative alliance transaction price calculated by the elastic price mechanism for a multi-resource shared micro-network cluster proposed in this study is a segmented function of $R_k^t$, which is shown as:

$$p_{k,sell}^t=\left\{\begin{array}{l}\frac{c_{k,buy}^t\cdot \left(c_{k,buy}^t+c_{k,sell}^t\right)}{c_{k,buy}^t\left(1+R_k^t\right)+c_{k,sell}^t\left(1-R_k^t\right)}\\ ,0\le R_k^t\le 1\\ p_{k,buy}^t\cdot X_k^t+c_{k,sell}^t\left(1-X_k^t\right)\\ ,R_k^t>1\end{array}\right.$$

(19)

$$p_{k,buy}^t=\left\{\begin{array}{l}{p}_{k,sell}^t\cdot R_k^t+c_{k,buy}^t\left(1-R_k^t\right)\\ ,0\le R_k^t\le 1\\ \frac{c_{k,sell}^t\cdot \left(c_{k,buy}^t+c_{k,sell}^t\right)}{c_{k,sell}^t\left(1+X_k^t\right)+c_{k,buy}^t\left(1-X_k^t\right)}\\ ,R_k^t>1\end{array}\right.$$

(20)

where $c_{k,sell}^t$ and $c_{k,buy}^t$ are purchase and sale prices for resource $k$ in the superior market, respectively; $X_k^t$ is the reciprocal value of $R_k^t$.

2.4.2. Energy Trading Models in Middle and Lower Levels

Each MGO is the leader of the Stackelberg game and has the right to set energy prices within the network, which is influenced by factors such as customer load, its own equipment capacity and external energy prices. The customer is the follower in the game, with the objective of optimizing its own objective function, formulating energy consumption strategies and uploading them to the dispatching side of the MGO.

Objective Function:

The total system benefit of MGO $i$ is determined as the optimization objective to be maximized, including revenue from energy sales, the system operation cost, carbon emission trading cost, and green certificate cost, as shown in the following equations.

$$\max F_i={\displaystyle \sum _{t=1}^T\left(Isel{l}_i^t-f\mathrm{o}{p}_i^t-Cce{t}_i^t-Cgc{t}_i^t\right)}$$

(21)

$$Isel{l}_i^t=p_{i,el}^t{P}_{i,el}^t+p_{i,hl}^t{Q}_{i,hl}^t$$

(22)

$$f\mathrm{o}{p}_i^t={\displaystyle \sum _k^K\left(p_{k,buy}^t{P}_{i,k,buy}^t-p_{k,sell}^t{P}_{i,k,sell}^t\right)}+V_{i,g,buy}^t{p}_{}^{gas}+{\displaystyle \sum _{j\in {\Omega }_G}^{}\left(p_{i,j}^t{P}_{i,j}^t\right)}$$

(23)

where $p_{i,el}^t$ and $p_{i,hl}^t$ are prices for electricity and heat sales to customers of MGO $i$ in time interval $t$, respectively; $V_{i,g,buy}^t$ is the total gas purchase; $p_i^{gas}$ is the price of natural gas; ${\mathrm{\Omega }}_G$ is device collection, $j\in {\mathrm{\Omega }}_G$; $p_{i,j}^t$ and $P_{i,j}^t$ are the operating cost factor and the output of equipment $j$, respectively.

The benefit of customer $i$ is determined as the optimization objective to be maximized, which is shown in Equation (24).

$$\max U_i={\displaystyle \sum _{t=1}^T\left(a{e}_i{P}_{i,el}^t-\frac{b{e}_i}{2}{\left(P_{i,el}^t\right)}^2-p_{i,el}^t{P}_{i,el}^t\right)}+{\displaystyle \sum _{t=1}^T\left(a{h}_i{Q}_{i,hl}^t-\frac{b{h}_i}{2}{\left(Q_{i,el}^t\right)}^2-p_{i,hl}^t{Q}_{i,hl}^t\right)}$$

(24)

where $U_i$ is benefit of the customer $i$; $a{e}_i$ and $b{e}_i$ are preference constants related to the customer’s electricity consumption; $a{h}_i$ and $b{h}_i$ are preference constants related to the customer’s heat consumption.

Main Constraints:

1.

Constraints on energy balance

The equipment composition of different types of microgrids differs in part, resulting in diversified energy balance constraints. When the MGO is an industrial MGO, the energy balance is shown in Equation (25).

$$\left\{\begin{array}{l}{P}_{i,PV}^t+P_{i,wind}^t+P_{i,e,GT}^t+P_{i,dis,ESE}^t+P_{i,e,buy}^t\\ =P_{i,el}^t+P_{i,e,EC}^t+P_{i,ch,ESE}^t+P_{i,e,sell}^t\\ P_{i,h,GT}^t+P_{i,h,GB}^t+P_{i,dis,HSE}^t+P_{i,h,buy}^t\\ =P_{i,hl}^t+P_{i,ch,HSE}^t+P_{i,h,sell}^t\\ V_{i,g,GB}^t+V_{i,g,GT}^t+V_{i,g,load}^t+V_{i,ch,NSE}^t\\ =V_{i,g,buy}^t+V_{i,dis,NSE}^t\end{array}\right.$$

(25)

When the MGO is a commercial MGO, the energy balance is shown in Equation (26).

$$\left\{\begin{array}{l}{P}_{i,PV}^t+P_{i,e,GT}^t+P_{i,dis,ESE}^t+P_{i,e,buy}^t-P_{i,ch,ESE}^t\\ =P_{i,el}^t+P_{i,EV}^t+P_{i,e,EB}^t+P_{i,e,EC}^t+P_{i,e,sell}^t\\ P_{i,h,GT}^t+P_{i,h,EB}^t+P_{i,dis,HSE}^t+P_{i,h,buy}^t\\ =P_{i,hl}^t+P_{i,h,AC}^t+P_{i,ch,HSE}^t+P_{i,h,sell}^t\\ P_{i,cload}^t=P_{i,h,AC}^t+P_{i,h,EC}^t\end{array}\right.$$

(26)

When the MGO is a residential MGO, the energy balance is shown in Equation (27).

$$\left\{\begin{array}{l}{P}_{i,PV}^t+P_{i,dis,ESE}^t+P_{i,e,buy}^t\\ =P_{i,el}^t+P_{i,EV}^t+P_{i,e,EB}^t+P_{i,e,EC}^t+P_{i,ch,ESE}^t+P_{i,e,sell}^t\\ P_{i,h,EB}^t+P_{i,h,buy}^t=P_{i,hl}^t+P_{i,h,sell}^t\end{array}\right.$$

(27)

2.

Constraints on multi-resource sharing transaction

In order to avoid the simultaneous buying and selling of resources between the MGOs and the external source, the following constraints are set:

$$\left\{\begin{array}{l}0\le P_{i,k,sell}^t\le {\delta }_{i,k,sell}^t{P}_{i,k,sell}^{\max }\\ 0\le P_{i,k,buy}^t\le {\delta }_{i,k,buy}^t{P}_{i,k,buy}^{\max }\\ {\delta }_{i,k,buy}^t+{\delta }_{i,k,sell}^t\le 1\end{array}\right.$$

(28)

where ${\delta }_{i,k,buy}^t$ and ${\delta }_{i,k,sell}^t$ are both bull variables; $P_{i,k,sell}^{\max }$ and $P_{i,k,buy}^{\max }$ are the maximum limits on the sale and purchase of resource $k$ with external parties, respectively.

3.

Price constraints on energy transactions between MGO and customers

$$\left\{\begin{array}{l}{p}_{i,e\min }^t\le p_{i,el}^t\le p_{i,e\max }^t\\ p_{i,h\min }^t\le p_{i,hl}^t\le p_{i,h\max }^t\end{array}\right.$$

(29)

where $p_{i,e\min }^t$ and $p_{i,e\max }^t$ are the upper and lower limits for the price of electricity sold by MGO $i$ to customers, respectively; $p_{i,h\min }^t$ and $p_{i,h\max }^t$ are the upper and lower limits for the price of heat sold by MGO $i$ to customers, respectively.

2.5. Basic Data

In this study, three MGOs were employed for analysis. MGO1, MGO2, and MGO3 were assumed to be the industrial, residential, and commercial MGO, respectively. Due to the large differences of customer attributes of various MGOs, the energy demands also illustrate diversified features. This results in the different equipment configuration on the supply side of each MGO. The specific equipment configuration is shown in

Table 1. Equipment configuration of various MGOs.

Equipment	MGO1 Capacity (kW/kW·h)	MGO2 Capacity (kW/kW·h)	MGO3 Capacity (kW/kW·h)
PV	1000	880	320
WT	550	-	-
CHP	1000	-	400
GB	800	-	-
EB	-	200	60
EC	50	100	50
AC	-	-	50
NSE	150	-	-
ESE	500	-	250
HSE	800	300	250

, and related technical and economical parameters are shown in

Table 2. Technical and economical parameters of various equipment.

Equipment	Efficiency/COP	Unit Operation and Maintenance Costs /(¥/(kW·h))
PV	-	0.04
WT	-	0.1
CHP	0.3	0.04
GB	0.82	0.02
EB	0.96	0.018
EC	4	0.01
AC	1.2	0.025
ESE	0.95	0.018
HSE	0.89	0.016
NSE	0.93	0.02

[29,30]. The natural gas price is assumed to be 3.15 yuan/m³ [31]. As there is currently no uniformity in the purchase and sale price of thermal energy, here they are assumed to be 0.1 yuan/kWh and 0.4 yuan/kWh, respectively, by referencing [32,33]. Considering that CET and GCT are in the development stage at this stage, the prices at which each MGO sells carbon and green certificates to the market are assumed to be 60% of the prices at which it buys them from the market. The prices for carbon sold and purchased in the superior carbon market are assumed to be 100 yuan/t and 60 yuan/t, respectively. The prices of green certificates sold and purchased in the superior green certificate market are assumed to be 70 yuan/certificate and 42 yuan/certificate, respectively.

The renewable energy output is determined based on measured solar irradiation and wind speed, as shown in Figure 2. The load demand of customers in each MGO is obtained from eQUEST software simulation, as shown in Figure 3.

3. Results

Three scenarios were considered to analyze the performance of the proposed sharing trading strategy:

Scenario 1: Without consideration of carbon–green certificate mechanism, microgrids operate independently and only trade with the superior energy network;

Scenario 2: With consideration of carbon–green certificate mechanism, microgrids operate independently and only trade with the superior energy network;

Scenario 3: With consideration of joint carbon–green certificate mechanism, microgrids are traded according to the multi-resource shared trading strategy among microgrids proposed in this study.

3.1. Comparative Analysis of Results

The resource purchase prices of each MGO guided by the cooperative sharing center are smaller than the resource purchase prices from the external market, while the resource sale prices are larger than the resource sale prices to the external market. Therefore, the operating costs of MGO can be effectively reduced under the cooperative sharing transaction mechanism, and the overall benefits of multi-microgrids are better improved. The natural gas cost of MGO1 and MGO3 under three scenarios are shown in Figure 4. Since MGO2 is not equipped with gas equipment, it has no natural gas cost and is not shown in the figure. Moreover, the economic comparison of different scenarios is shown in

Table 3. Economic comparison of different scenarios.

MGO	Scenarios	Net Income (¥)	Operation Costs (¥)	CET Cost (¥)	GET Cost (¥)
MGO 1	Scenario 1	22,235.50	16,238.77	-	-
	Scenario 2	22,497.00	16,327.00	−87.45	−262.42
	Scenario 3	22,605.39	16,218.44	−82.10	−267.41
MGO 2	Scenario 1	2065.40	4435.38	-	-
	Scenario 2	2131.20	4435.40	42.56	−75.18
	Scenario 3	2244.93	4322.07	34.68	−67.7
MGO 3	Scenario 1	397.91	2223.03	-	-
	Scenario 2	510.74	2126.13	9.01	−24.99
	Scenario 3	545.72	2093.73	6.48	−24.99

. Under Scenario 2, the overall efficiency of the multi-microgrids is increased by 1.78% compared to Scenario 1. By introducing a shared trading strategy, the overall efficiency of Scenario 3 increases by 2.82% compared to Scenario 1, reflecting the economics of the proposed method.

In addition, as the low-carbon attributes are translated into low-carbon economic indicators in the objective function, the system is prompted to purchase less electricity from the external grid, preferring to use less carbon-emitting natural gas. In this way, the overall energy efficiency of the system increases, making the clean attributes more obvious. On the other hand, for MGO2, there is no room for operational optimization under different scenarios, as it has no gas-fired equipment and the main energy supplier is the PV module.

3.2. Trading Price Analysis

As shown in Figure 5, the peak PV output is at noon during the day, which is also the peak energy demand for MGO1 and MGO3, while the energy load of MGO2 is considered low during this period. Therefore, the total electricity demand and surplus are comparable during this period, and the MGO purchase price and sale price are similar according to the supply/demand ratio calculated by the cooperative sharing center. In addition, it can be found that the purchase and sale prices of MGOs are the same as those of the carbon market. This is due to the fact that the environmental attributes of renewable energy in MGOs are more beneficial for green certificate trading than for carbon trading, thus the MGO is not very motivated to participate in the carbon market.

4. Conclusions

In this study, a new trading strategy for energy sharing and low-carbon economic dispatch between multi-microgrid, which involves complex trading models and pricing methods, is proposed. The established multi-level optimal dispatch and multi-resource sharing method is solved in terms of the economic optimality of each MGO. In addition, the shared trading and pricing mechanisms of multiple resources among multiple trading agents are further investigated. Three typical MGOs have been assumed for numerical study. According to the simulation results, the following conclusions can be deduced.

• The proposed pricing mechanism consists of a master–slave game transaction between each MGO and its users, as well as a supply/demand ratio pricing mechanism where the cooperative sharing center publishes a resource transaction price to guide each MGO.
• The proposed joint carbon–green certificate trading mechanism avoids the double counting issues of the environmental value of renewable energies and facilitates the shift of the energy consumption structure of the multi-microgrid cluster to natural gas with lower carbon emissions. This proves that the joint carbon–green certificate mechanism can promote the development of the whole system towards clean and efficient energy use.
• The multi-resource sharing strategy further investigates the impact of carbon emissions and green energy consumption on the operational efficiency of MGOs. The net benefit of the proposed mechanism (Scenario 3) increases by 2.82%, compared to the scenario (Scenario 1) without joint carbon–green certificate trading and each micro-grid operating independently. This demonstrates the economics of the proposed shared trading strategy.
• The energy certificate–carbon multi-resource sharing and trading among multi-microgrids breaks trading barriers in energy and various policy markets, providing some reference for future related research.

In this study, while considering the current situation of carbon and green certificate market, the pricing mechanism was simplified. In the following studies, considering the growing maturity of various policy markets, ladder-type carbon trading, which divides the trading amount in different intervals with diversified prices, is expected to be introduced. In this way, the enthusiasm of trading can be stimulated and the efficiency of carbon emission reduction can be enhanced. In addition, the uncertainty of renewable energy may be also considered while determining the shared trading strategy.