Long-term scheduling method for integrated energy system of power-gas interconnected community based on game

Abstract

A game-theoretic long-term scheduling method for an electricity-gas interconnected community integrated energy system includes the following steps: Step 1: Establishing a mathematical model based on the community integrated energy system architecture; Step 2: Constructing a game-theoretic optimal scheduling model for all parties involved in the integrated energy system; Step 3: Solving the game-theoretic optimal scheduling model. In Step 1, the community integrated energy system includes an energy supply end, a community area end, and a user end. The energy supply end includes a natural gas company and a power supply company. The community area end includes gas turbines, gas storage devices, and electrical energy storage devices. The user end includes natural gas access points and electricity access points. The natural gas company connects to the community area end and the user end through the natural gas network. The power supply company connects to the community area end and the user end through the power supply network. At the community area end, gas-to-electricity conversion is achieved through gas turbines.

Description

Technical Field

[0001] This invention relates to the field of integrated energy system optimization and scheduling technology, specifically to a long-term scheduling method for an electricity-gas interconnected community integrated energy system based on game theory. Background Technology

[0002] Energy is the material foundation for the development of human society, and the environment is a condition for sustainable social development. With economic and social development, humanity not only has higher demands for energy supply but also faces problems such as traditional energy shortages and severe pollution. Traditional energy and independent energy supply models are gradually failing to meet the needs of the times. Against this backdrop, the Energy Internet will become the mainstream model for future energy supply. In the Energy Internet, multiple energy sources operate in tandem. The first level involves the integration of a large amount of renewable energy, with wind and solar power becoming the main energy sources. The second level involves a large-scale cycle of non-electric energy sources such as gas and heat with electricity. Based on the characteristics of various energy sources, they complement each other, achieving both energy structure optimization, energy conservation and emission reduction, and ensuring the stability and economy of energy use. Inevitably, with the joint supply of multiple types of energy, conflicting interests and environmental requirements exist among the various energy supply entities, while load demanders also have more options for response based on their own interests. How to ensure the economic and social benefits for all parties involved in supply and demand under this environment has become an urgent problem to be studied and solved in the Energy Internet.

[0003] At present, the research on integrated energy systems in the energy Internet is divided into several main aspects. On the one hand, it is the power flow calculation and energy flow calculation under the integrated energy system, as well as the energy flow analysis considering the characteristics of various energy transmission pipe networks at the same time, such as the literature (Liu Hong, Zhao Chenxiao, Ge Shaoyun, etc. Time-series power flow calculation of electric-thermal integrated energy system based on refined heat network model [J]. Automation of Electric Power Systems, 2020. DOI: 10.7500 / AEPS20200312009). The research in this aspect is based on the traditional power system power flow calculation and adds elements of integrated energy, such as natural gas transmission pipelines and heat energy transmission pipelines. The research in this aspect uses traditional power flow calculation methods to study and analyze multi-energy factors. On the other hand, integrated energy systems are set under different environments, different regional scopes, and different research objects, and the optimal new energy consumption and economy are achieved through optimization. For example, integrated energy systems are established and optimized for different intelligent science and technology parks, industrial industrial parks, and intelligent living communities. There is also another aspect, which is to combine with the demand side to conduct joint optimization of integrated energy demand response, such as the literature (Wang Dan, Huang Deyu, Hu Qing'e, etc. Modeling and strategy of integrated demand response based on clearing of electric-thermal joint market [J]. Automation of Electric Power Systems, 2020, 44(12): 13-21.). In previous related research on power demand response, only in the single energy of electric energy, the peak shaving and valley filling of power load are carried out through the response of the power demand side, that is, through the time translation of the power load, the time curve of the power load tends to be flat. In the integrated energy system, users no longer simply increase or decrease the use of electric energy within a period, but can cooperate with the response by converting the energy type, and the purpose of peak shaving and valley filling of the power load can also be achieved, but the convenience and comfort of users are not greatly affected. The research object of integrated demand response is still large-scale energy hubs and small-scale smart parks, communities, etc., but the demand response model or demand response constraint is added. Because the interests of both supply and demand sides are considered, it is more meaningful. However, among the above-mentioned research objects, smart parks with wind, light, thermal power cogeneration, and energy storage are relatively idealized. In the current actual situation, only a small number of regional-level and user-level integrated energy system demonstration projects have been built. In an environment with zero foundation, not only the construction cost is high, but the cycle time is also long. Therefore, the universality and scalability of most of the research conducted under set conditions need to be examined. And with the increase in the types of energy, its complexity has led to the advancement of optimization methods, such as the two-layer optimization method and the strategy optimization of game theory. Here, considering the realistic basic conditions of the household load aggregation cluster, an electric-gas combined operation environment is set up based on the existing conditions, and game optimization analysis is carried out on this basis. In this way, the construction cycle is short, the cost of成果转化 is low, and it is more realistic.

[0004] It should be noted that the term "成果转化" in the original text seems to be a misnomer or an unclear expression. I translated it as "成果转化" as it is, but it might need to be further clarified or corrected in the original context.In the energy internet, natural gas, as a major energy source besides electricity, is widely supplied to businesses and households, with a mature transportation pipeline network. Most end-users nationwide have independent access points for electricity and gas. The interconnection of electricity and natural gas achieves complementarity, aligning with current energy conditions, reducing construction costs, and positively impacting the stability of the traditional power grid. On the demand side, individual household loads, due to their low flexibility and dispersed nature, are difficult to manage effectively. Therefore, household load aggregation communities are introduced to participate in energy demand-side management. In traditional power systems, the game is typically between the power supplier and the consumer. Under energy interconnection conditions, with an increased number of energy suppliers, the two-way game is insufficient for research needs; a three-way game study must be introduced.

[0005] Based on the research of the aforementioned scholars and the analysis of the current situation of integrated energy supply, this paper takes power supply companies, natural gas companies, and community residential load clusters as research objects, and establishes an integrated energy system, in which each party has its own interests and needs. On this basis, a dynamic continuous game model with two principals and one subordinate is established. The principal uses price as its strategy set, and the subordinate uses demand response as its strategy set. Evolutionary game theory is used to reach game equilibrium. Summary of the Invention

[0006] The purpose of this invention is to provide a long-term scheduling method for a comprehensive energy system based on a three-way game between the power supplier, the natural gas supplier, and the power consumer.

[0007] A game-theoretic long-term scheduling method for an integrated energy system in a community with interconnected electricity and gas systems includes the following steps:

[0008] Step 1: Establish a mathematical model based on the community's integrated energy system architecture;

[0009] Step 2: Construct a game-theoretic optimization scheduling model for all stakeholders in the integrated energy system;

[0010] Step 3: Solve the game-theoretic optimization scheduling model;

[0011] In step one, the community integrated energy system includes the energy supply side, the community area side, and the user side;

[0012] Energy supply includes natural gas companies and power supply companies;

[0013] The community-level components include gas turbines, gas storage devices, and electrical energy storage devices.

[0014] The user end includes natural gas access points and electricity access points;

[0015] Natural gas companies connect to community areas and users through natural gas networks; power companies connect to community areas and users through power supply networks.

[0016] At the community level, gas turbines are used to convert gas into electricity;

[0017] Considering the community's small gas turbine, rooftop photovoltaic system, gas storage tanks, and battery energy storage, the mathematical model of the community's integrated energy system can be expressed as:

[0018]

[0019] Where α is the conversion coefficient for electrical energy used for electrical utility, β is the conversion coefficient for gas used for power generation, and γ is... p γ is the electro-thermal conversion efficiency coefficient. h γ is the gas-to-heat conversion efficiency coefficient. s P is the gas-to-electricity conversion efficiency coefficient. out For electrical efficiency energy, H out For thermal energy, P in For electrical input energy, V in V is the gas input volume. sto To store the gas volume, P sto To store electrical energy, P sol This refers to the photovoltaic power generation capacity.

[0020] In step two,

[0021] The uncertain expression for obtaining the actual user demand response:

[0022]

[0023] ΔP out ΔP ′out These are fuzzy expressions for the upper and lower limits of the uncertainty response quantity of the comprehensive demand response.

[0024] The uncertainty of the electricity demand response is expressed in the expected form of a triangular fuzzy function, as shown in equation (3):

[0025]

[0026] As shown in equation (4), the expected value of the uncertainty in the thermal energy demand response is:

[0027]

[0028] λ is the decision trend value, which is generally taken as 1 / 2.

[0029] The objective function f of the power supply company (EPC) game model p It consists of two parts: the cost of electricity production and the negative cost of electricity sales revenue, which can be expressed as shown in equation (5):

[0030] fp =c p P in -j p P in (5)

[0031] In the formula, c p This is the comprehensive cost coefficient for electricity; electricity operation and maintenance costs are included in electricity production costs. p Let G be the electricity price. HLC (Household Load Cluster) meets the overall daily energy needs of the community residents. Since living habits generally do not change much, the consumption of electricity and heat can be considered constant over a certain period. Electricity demand can only be supplied by electricity, while heat demand can be supplied by either electricity or natural gas. Let G be... p_p For electrical energy used for electrical efficiency, G q_p Natural gas for conversion into electricity, G p_h To convert electrical energy into thermal efficiency, G q_h The thermal efficiency of natural gas conversion can be obtained as follows:

[0032]

[0033] ε is the comprehensive energy demand response value, which indicates that the heat utility load consumed is the coefficient obtained by choosing natural gas. Then 1-ε is the part of the heat utility load consumed that is converted from electrical energy.

[0034]

[0035]

[0036] but The demand of natural gas companies is transformed into a game theory mathematical model, where the objective function is the negative value of the revenue. This objective function is related to the volume of gas output, production costs, and transportation costs. a This is the cost coefficient per cubic meter of natural gas, which includes production and transportation costs. a For natural gas prices. a Let be the objective function of the NAC (Natural Gas Company) game model. We obtain...

[0037]

[0038]

[0039] The game-theoretic mathematical model of community resident load clustering is expressed mathematically as follows:

[0040]

[0041] In the formula, f L The objective function is HLC (User Load Cluster).

[0042] The game model described above includes three participants: a power supply company (EPC), a natural gas company (NAC), and a community electricity cluster (HLC). The strategy sets are the electricity price, the gas price, and the overall energy demand response value, respectively.

[0043] The game model can then be expressed as:

[0044] G = {{EPC, NAC, HLC}; j p ,j a ,ε;f p ,f a ,f L} (13)

[0045] In step three,

[0046] A non-dominated sorting genetic algorithm is used to find the Pareto front, i.e., the set of all non-dominated solutions. Based on the actual interests of each player, nonnegativity and the constraint range of the strategy set are obtained, and practically feasible Pareto solutions are selected. Based on the characteristics of a two-master-one-follower game, the optimal Pareto solution under different conditions is analyzed. Furthermore, the NASH equilibrium solution is solved based on the optimal reaction equation. p ,j a ε is a non-empty convex set, f p ,f a ,f L It is continuous and quasi-concave, therefore an equilibrium solution exists. ε * Let represent the equilibrium strategy values ​​for electricity price, gas price, and demand response, respectively. Then, mathematically, this can be described as...

[0047]

[0048] Compared with the prior art, the present invention has the following technical effects:

[0049] 1) This invention establishes a comprehensive energy system operation mathematical model involving power supply companies, natural gas companies, and users, analyzes the game relationship among the three, realizes long-term energy dispatch optimization operation, and guides the optimized operation of the energy internet;

[0050] 2) Compared with the existing research results on the operation strategy of the envisioned integrated energy system, the solution of the present invention is based on the joint operation environment of multiple energy types such as electricity and gas. Given that most communities already have electricity and gas energy access, the present invention has a basis for technology transfer and a short cycle.

[0051] 3) This invention models community users as a cluster of household loads, which facilitates unified optimization and management; it uses a non-dominated sorting genetic algorithm combined with the optimal reaction function to solve the equilibrium solution of the master-slave game model, which reduces computational complexity. Attached Figure Description

[0052] The present invention will be further described below with reference to the accompanying drawings and embodiments:

[0053] Figure 1 This is a schematic diagram of the community integrated energy system in this invention;

[0054] Figure 2 This is a schematic diagram of the Pareto front solution set in an embodiment of the present invention. Detailed Implementation

[0055] A long-term scheduling optimization method for an integrated energy system in an interconnected electricity and gas community based on master-slave game theory includes the following steps:

[0056] Step 1: Establish a mathematical model based on the community's integrated energy system architecture:

[0057] In the context of the energy internet, the natural gas pipelines and household electrical circuits in future integrated smart energy communities will no longer be completely separate and independent, but will be coupled through physical equipment and user response. On the energy supply side, the two main energy sources will remain the existing ones: electricity and natural gas. Within the community, small gas turbines will be used to convert gas into electricity, and gas and electricity storage devices will be constructed according to the community's actual conditions. Residential load aggregation groups, as the load aggregation of the entire community, will provide comprehensive demand response to energy supply. Natural gas companies and power companies will supply energy to the community separately, each with its own interests. By combining its own small-scale power generation, energy storage, and energy conversion capabilities, the community will achieve optimal long-term dispatch and operation while meeting the energy needs of community users through comprehensive demand response.

[0058] For household loads, there are relatively few instances of power and gas rationing, as both electricity and gas are considered constant sources. Household heat load has two energy sources: one is electricity converted into heat load from the grid, and the other is obtained through natural gas combustion. Based on the current characteristics of household loads, a portion of the electricity is converted into heat load, and the remainder is used as electricity. Natural gas is used solely as a heat load source. out H out For electrical load and thermal load, P in V in The electrical-gas energy path inlet is characterized by work and volume, respectively, for... Figure 1 The community integrated energy system, considering small gas turbines, rooftop photovoltaics, gas storage tanks, and battery energy storage, can be mathematically modeled according to the law of conservation of energy as follows:

[0059]

[0060] Where α is the conversion coefficient for electrical energy used for electrical utility, β is the conversion coefficient for gas used for power generation, and γ is... p γ is the electro-thermal conversion efficiency coefficient. h γ is the gas-to-heat conversion efficiency coefficient. s P is the gas-to-electricity conversion efficiency coefficient. out For electrical efficiency energy, H out For thermal energy, P in For electrical input energy, V in V is the gas input volume. sto To store the gas volume, P sto To store electrical energy, P sol This refers to the photovoltaic power generation capacity.

[0061] Step 2: Construct a game-theoretic optimization scheduling model for all stakeholders in the integrated energy system.

[0062] In actual integrated energy demand response, users do not respond accurately and are affected by various factors, thus introducing uncertainty. ΔP out ΔP ′out These are the fuzzy expressions for the upper and lower limits of the uncertainty response quantity of the comprehensive demand response curve. The uncertainty of this actual demand response is expressed as equation (2):

[0063]

[0064] The uncertainty of the electricity demand response is expressed in the expected form of a triangular fuzzy function, as shown in equation (3):

[0065]

[0066] λ is the decision trend value, which is generally taken as 1 / 2, that is, the standard expectation value of the triangular fuzzy function.

[0067] Similarly, the expected value of the uncertain expression for heat demand response is:

[0068]

[0069] Electricity demand-side load includes base load and flexible load. Base load is uninterruptible, non-transferable, and cannot participate in demand response. Flexible load is divided into interruptible load and transferable load. Interruptible load can be directly reduced, while transferable load can adjust its operating mode according to electricity demand, using electricity at different times, i.e., the load shifts along the time axis. When integrated energy systems are operating in conjunction, the load type can be converted to other load types, i.e., vertical load conversion is achieved; this is called transferable load. A load aggregation cluster based on existing household loads is mathematically modeled to closely reflect the basic conditions of real-world household users. Gas-to-electricity coupling involves selecting the functional mode of heat load between power and gas sources. The objective function f of the power company's game theory model is... p It consists of two parts: the cost of electricity production and the negative cost of electricity sales revenue, which can be expressed as shown in equation (5):

[0070] f p =c p P in -j p P in (5)

[0071] In the formula, c p This is the comprehensive cost coefficient for electricity; electricity operation and maintenance costs are included in electricity production costs. p Let G be the electricity price. HLC (Household Load Cluster) meets the overall daily energy needs of the community residents. Since living habits generally do not change much, the consumption of electricity and heat can be considered constant over a certain period. Electricity demand can only be supplied by electricity, while heat demand can be supplied by either electricity or natural gas. Let G be... p_p For electrical energy used for electrical efficiency, G q_p Natural gas for conversion into electricity, G p_h To convert electrical energy into thermal efficiency, G q_h The thermal efficiency of natural gas conversion can be obtained as follows:

[0072]

[0073] ε is the comprehensive energy demand response value, which indicates that the heat utility load consumed is the coefficient obtained by choosing natural gas. Then 1-ε is the part of the heat utility load consumed that is converted from electrical energy.

[0074]

[0075]

[0076] but

[0077] The demand of natural gas companies is transformed into a game theory mathematical model, where the objective function is the negative value of the revenue. This objective function is related to the volume of gas output, production costs, and transportation costs. a This is the cost coefficient per cubic meter of natural gas, which includes production and transportation costs. a For natural gas prices. a Let be the objective function of the NAC (Natural Gas Company) game model. We obtain...

[0078]

[0079]

[0080] The game-theoretic mathematical model of community resident load clustering is expressed mathematically as follows:

[0081]

[0082] In the formula, f L For the payment function of HLC, j a Indicates the price of natural gas.

[0083] The game model described above includes three participants: a power supply company (EPC), a natural gas company (NAC), and a community electricity cluster (HLC). The strategy sets are the electricity price, the gas price, and the overall energy demand response value, respectively.

[0084] The game model can then be expressed as:

[0085] G = {{EPC, NAC, HLC}; j p ,j a ,ε;f p ,f a ,f L} (13)

[0086] Step 3: Solving the game theory model:

[0087] A non-dominated sorting genetic algorithm (NSGA-II) is used to derive the Pareto front, i.e., the set of all non-dominated solutions. Based on the actual interests of each player, nonnegativity and the constraint range of the strategy set are obtained, and practically feasible Pareto solutions are selected. Based on the characteristics of a two-leader-one-follower game, the optimal Pareto solution under different conditions is analyzed. Furthermore, the NASH equilibrium solution is solved based on the optimal reaction equation. p ,j a ε is a non-empty convex set, f p ,f a ,f LIt is continuous and quasi-concave, therefore an equilibrium solution exists. ε * Let represent the equilibrium strategy values ​​for electricity price, gas price, and demand response, respectively. Then, mathematically, this can be described as...

[0088]

[0089] Example:

[0090] The strategy set data references current energy data from some regions, and the energy conversion data references approximate values ​​of the physical conversion efficiency of various energy sources. The gas price is approximately 5 yuan / m³. 3 The calorific value of natural gas is approximately 30*10. 6 Joules / m 3 The calorific value of the electrical energy is approximately 3*10. 6 Joules / kW·h, the calorific value of natural gas converted to electrical calorific value is approximately 10 kW·h / m³. 3 Assuming a single household's daily electricity consumption ranges from a minimum of 2 kWh to a maximum of 8 kWh, with electricity costs at 0.3 yuan / kWh and gas costs at 2 yuan / m³, the following calculations apply. 3 The minimum heat demand is 10*10 6 Joules, with a maximum value of 20*10 6 The Pareto front is calculated using the NSGA-II algorithm on the MATLAB platform to solve for 1000 households. Once the electricity price exceeds 2 yuan / kW·h, users maintain a minimum electricity consumption. The optimization cycle is one day. The Pareto front is solved using the NSGA-II algorithm on the MATLAB platform.

[0091] Figure 2 The table shows the Pareto front obtained using the NSGA-II algorithm. The three axes represent the objective functions of the three players. The specific values ​​of individual solutions on the Pareto front are shown in Table 1, with each point representing an individual solution. Based on the rationality of the established game model, feasible solutions are selected from the Pareto front, meaning solutions where all players have negative objective functions and generate payoffs. The asterisked points represent the selected feasible solutions, corresponding to the shaded solutions in Table 1. Figure 2 The feasible solutions are relatively concentrated, and are located in the middle of the graph. This may be because the equilibrium of the game model dictates that no single player can unilaterally gain a large benefit.

[0092] Table 1 Pareto Frontier

[0093] Tab.1 Pareto frontier

[0094]

[0095] The set of three-party strategies corresponding to the selected Pareto feasible solutions is shown in Table 2:

[0096] Table 2 Pareto Feasible Strategy Set

[0097] Tab.2Pareto Feasible strategy set

[0098]

[0099] Because the power supply company and the natural gas company are the dominant players in the game, the slave community load cluster passively chooses its strategy. Therefore, in the feasible solution, both dominant players want to operate at the point with the highest profit, namely points 29 and 20. However, at point 29, the profit difference between the two dominant players is the largest, and the slave's objective function value is the largest. Therefore, the game is most likely to operate at point 20, at which point the slave's objective function value is also close to its minimum.

Claims

1. A long-term scheduling method for a community integrated energy system in an electricity-gas interconnected community based on game theory, characterized in that, It includes the following steps: Step 1: Establish a mathematical model based on the community's integrated energy system architecture; Step 2: Construct a game-theoretic optimization scheduling model for all stakeholders in the integrated energy system; Step 3: Solve the game-theoretic optimization scheduling model; In step one, the community integrated energy system includes the energy supply side, the community area side, and the user side; Energy suppliers include natural gas companies and power supply companies; The community-level components include gas turbines, gas storage devices, and electrical energy storage devices. The user end includes natural gas access points and electricity access points; Natural gas companies connect to community areas and users through natural gas networks; power companies connect to community areas and users through power supply networks. At the community level, gas turbines are used to convert gas into electricity; Considering the community's small gas turbine, rooftop photovoltaic system, gas storage tanks, and battery energy storage, the mathematical model of the community's integrated energy system can be expressed as: （1）； in, This is a proportionality coefficient for converting electrical energy into electrical utility. This is the ratio of the amount of gas used for power generation. The electro-thermal conversion efficiency coefficient. The gas-to-heat conversion efficiency coefficient. The gas-to-electricity conversion efficiency coefficient. For electrical energy use, For thermal energy, To input energy into electricity, For the gas input volume, To store gas volume, To store electrical energy capacity, For photovoltaic power generation capacity; Objective function of the power supply company EPC game model It consists of two parts: the cost of electricity production and the negative cost of electricity sales revenue, which can be expressed as shown in equation (5): （5）； In the formula, This is the comprehensive cost coefficient for electricity; electricity operation and maintenance costs are included in electricity production costs. For electricity pricing, the user load cluster HLC meets the overall daily energy needs of the community residents. Since living habits generally do not change significantly, it can be assumed that electricity and heat demand are constant over a certain period. Electricity demand can only be supplied by electricity, while heat demand can be supplied by either electricity or natural gas. Electrical energy for electrical efficiency Natural gas for conversion into electricity To convert electrical energy into thermal energy, The thermal efficiency of natural gas conversion can be obtained as follows: （6）； The comprehensive energy demand response value represents the coefficient obtained by selecting natural gas, indicating that the heat utility load consumed is equal to the coefficient obtained by selecting natural gas. That is, the portion of its heat load that is consumed is converted from electrical energy; （7）； （8）； but (9).

2. The method according to claim 1, characterized in that, In step two, The uncertain expression for obtaining the actual user demand response: （2）； , These are fuzzy expressions for the upper and lower limits of the uncertainty response quantity in the overall demand response; The uncertainty of the electricity demand response is expressed in the expected form of a triangular fuzzy function, as shown in equation (3): （3）； As shown in equation (4), the expected value of the uncertainty in the thermal energy demand response is: （4）； This represents the decision trend value. The demand from natural gas companies can be transformed into a game theory mathematical model, where the objective function is the negative value of the revenue. This objective function is related to the volume of gas output, production costs, and transportation costs. This is the cost coefficient per cubic meter of natural gas, which includes production and transportation costs. For natural gas prices, For the objective function of the NAC game model of the natural gas company, we get: （10）； （11）； The game-theoretic mathematical model of community resident load clustering is expressed mathematically as follows: （12）； In the formula, The objective function for the user load cluster HLC; The above game model includes three participants: the power supply company EPC, the natural gas company NAC, and the community residential electricity cluster HLC. The strategy sets are electricity price, gas price, and comprehensive energy demand response value, respectively. The game model can then be expressed as: （13）。 3. The method according to claim 1, characterized in that, In step three, A non-dominated sorting genetic algorithm is used to find the Pareto front, i.e., the set of all non-dominated solutions. Based on the actual interests of each player, nonnegativity and the constraint range of the strategy set are obtained, and practically feasible Pareto solutions are selected. Based on the characteristics of a two-master-one-follower game, the optimal Pareto solution under different conditions is analyzed. Furthermore, the NASH equilibrium solution is solved based on the optimal reaction equation. It is a non-empty convex set. It is continuous and quasi-concave, therefore an equilibrium solution exists. Let represent the equilibrium strategy values ​​for electricity price, gas price, and demand response, respectively. Mathematically, this can be described as: （14）。

Images