Evolutionary game-based multi-agent joint investment cloud energy storage capacity planning method

Abstract

This invention discloses a multi-entity joint investment cloud energy storage capacity planning method based on evolutionary game theory, comprising the following steps: Step 1, establishing an intraday scheduling model for the cloud energy storage system by integrating historical load, irradiance, and temperature data; Step 2, taking cloud energy storage operators and the power grid as the main entities, and the planned energy storage capacity as the decision variable, establishing a multi-entity joint investment cloud energy storage capacity planning model based on the objective function and constraints, aiming to minimize the payment of both parties; Step 3, solving the example using a multi-strategy set evolutionary game model to obtain the system's evolutionary stable strategy; Step 4, conducting simulation analysis based on the multi-strategy set evolutionary game. This invention aims to lower the investment threshold for cloud energy storage through multi-entity joint investment in cloud energy storage construction, guide the orderly development of cloud energy storage models, thereby improving the renewable energy absorption capacity, mitigating wind and solar power curtailment, and enhancing power system stability.

Description

Technical Field

[0001] This invention relates to the field of power systems, and in particular to a multi-entity joint investment cloud energy storage capacity planning method based on evolutionary game theory. Background Technology

[0002] While the installed capacity of new energy power generation, mainly wind and solar power, continues to rise, the problem of wind and solar curtailment persists. At the same time, the variability and uncertainty of new energy sources such as wind and solar power increase the complexity of balancing load and power generation, and bring new challenges to maintaining system reliability. Their volatility and intermittency have an increasingly significant impact on the reliability and stability of the power system.

[0003] Cloud-based energy storage, as a novel business model, can rapidly absorb or release electrical energy, effectively mitigating fluctuations in wind and solar power output, improving the reliability of the power system and the grid connection capacity of new energy generation, and effectively compensating for the randomness of new energy sources. It provides a completely new solution for ensuring the safe and stable operation of the power system and addressing the centralized grid connection and consumption of large-scale new energy sources. However, the high cost of energy storage construction restricts its large-scale application.

[0004] The shift from single-entity investment in cloud energy storage by cloud energy storage operators to joint investment by multiple entities can lower the investment threshold for cloud energy storage and guide the orderly development of cloud energy storage models. However, in the new investment environment, how to balance the interests of different investors is a challenge in cloud energy storage capacity planning. Summary of the Invention

[0005] The technical problem to be solved by this invention is to provide a multi-entity joint investment cloud energy storage capacity planning method based on evolutionary game theory. The aim is to reduce the investment threshold of cloud energy storage, guide the orderly development of cloud energy storage models, and thereby improve the renewable energy consumption capacity, alleviate wind and solar curtailment, and improve the stability of the power system by having multiple entities jointly invest in cloud energy storage construction.

[0006] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a multi-agent joint investment cloud energy storage capacity planning method based on evolutionary game theory, comprising the following steps:

[0007] Step 1: Establish a daily scheduling model for the cloud energy storage system by integrating historical load, irradiance, and temperature data;

[0008] Step 2: Taking cloud energy storage operators and the power grid as the main entities, and the planned energy storage capacity as the decision variable, a multi-entity joint investment cloud energy storage capacity planning model is established based on the objective function and constraints, with the goal of minimizing the payment of both parties.

[0009] Step 3: Solve the example using a multi-strategy set evolutionary game model to obtain the system's evolutionary stable strategy;

[0010] Step 4: Conduct simulation analysis based on multi-strategy set evolutionary game.

[0011] A further improvement to the technical solution of the present invention is that the process of establishing the intraday scheduling model of the cloud energy storage system in step 1 is as follows:

[0012] The operation of the cloud energy storage system is uniformly scheduled by the cloud energy storage management system. The cloud energy storage management system monitors the user's electricity consumption curve through cloud batteries distributed on the user side and uploads the information to the management system. It rationally schedules the distributed power sources and energy storage batteries within the cloud energy storage system to ensure power quality and improve the user's electricity experience. With the goal of maximizing the operating profit of cloud energy storage, it optimizes the charging and discharging strategy of energy storage batteries and the output of distributed power sources by comprehensively considering historical load, irradiance, and temperature data, under the conditions of energy storage state of charge, wind and solar curtailment rate constraints, system output power and node voltage balance.

[0013] Let N be defined as a group of users in the system willing to install PVS, WTS, and ESS. The HLM module in the cloud battery calculates the energy load curve ln(h) for each user at each time period, defined as the energy that the customer needs to purchase within the time period h. Let the time scale be 1 hour. , Where H = 24 is the daily scheduling duration, and each HLM module pre-schedules user energy consumption at the beginning of each time scale. This represents the energy load of customer n over h hours on day d. When considering long-term multi-day scheduling for capacity planning, the label (d) represents a specific day.

[0014] Obtained from HLM Afterwards, the cloud-based battery sends daily load information to the cloud energy storage operator, setting... The cost of generating and distributing the electricity required by users at time h, based on the total load of the system. An assessment is conducted by the cloud energy storage operator, and a quadratic cost function is defined:

[0015]

[0016] in, (>0) is a constant cost coefficient, let... Based on the power plant's operating plan and the varying availability of intermittent energy over time, the cost function is the actual energy cost to generate the required load or the labor cost incurred by the operator in controlling and scheduling the equipment.

[0017] To facilitate advance scheduling, the service provider announces a daily pricing overview based on the obtained average cost through a communication network connecting all customers. ,therefore:

[0018]

[0019] Daily Price Overview The HLM module re-optimizes its energy consumption plan and sends the generated load configuration information back to the service provider. Then, the pricing information is changed, and the HLM module iterates the optimization process again.

[0020] Renewable solar and wind energy are complementary and should be prioritized to meet electricity demand. Energy storage batteries are used to address energy shortages. Finally, if available resources are insufficient to meet demand, micro gas turbines are used for power generation. Cloud energy storage can be categorized into three main scenarios based on different power generation and demand levels:

[0021] (1) Power generation meets demand

[0022] The electricity generated by photovoltaic panels and wind turbines equals the sum of user load demand, that is:

[0023]

[0024]

[0025] In the formula, This represents the output power of the solar panel at time t. This represents the output power of the fan at time t. This represents the user load at time t. This represents the energy stored in the battery pack at time t. Indicates that the battery pack is in Energy storage at any time;

[0026] (2) Power generation exceeds demand

[0027] When the electricity generated by renewable energy exceeds the user's load demand, the surplus electricity will be used to charge the energy storage batteries, that is:

[0028]

[0029]

[0030] In the formula, Indicates the battery pack charging power. This indicates the charging efficiency of the battery pack. This represents the energy stored in the battery pack at time t;

[0031] (3) Power generation is less than demand

[0032] When renewable energy generation is insufficient to meet user load demand, energy storage batteries and micro gas turbines are used to discharge electricity to compensate for the power generation shortfall.

[0033]

[0034]

[0035] In the formula, Indicates the battery pack discharge power. This indicates the discharge efficiency of the battery pack.

[0036] A further improvement to the technical solution of this invention lies in the following process for establishing the multi-entity joint investment cloud energy storage capacity planning model in step 2:

[0037] Step 2-1: Establish the objective function for cloud energy storage operators.

[0038] The goal of operator cloud energy storage is to minimize annual operating costs. Based on this, an objective function is established, in which the decision variables are the installation capacity of the solar panels, wind turbines, micro gas turbines and energy storage batteries to be planned.

[0039]

[0040] In the formula, ANC represents the annualized cost of the cloud energy storage system. Subsidies for photovoltaic power generation, For the electricity purchase and sale fees of cloud energy storage operators, This indicates a reduction in electricity revenue. The proportion of investment by cloud energy storage operators This refers to the revenue sharing coefficient for cloud energy storage operators.

[0041] (1) Annualized cost of cloud energy storage system

[0042] The annualized cost calculation for a cloud energy storage system is shown below:

[0043]

[0044] In the formula, , , and These represent the unit costs of wind turbines, solar photovoltaic panels, energy storage batteries, and micro gas turbines, respectively. , , and These represent the total number of wind turbines, solar photovoltaic panels, and batteries, respectively.

[0045] The cost of energy storage facilities includes construction costs, recycling costs, annual operating and maintenance costs, and the total ANC value for each component is expressed as follows:

[0046]

[0047]

[0048]

[0049]

[0050] In the formula, , , and These represent the construction costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. , , and These represent the annual operating and maintenance costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. , , and These represent the recycling costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively.

[0051] (2) Government subsidies for photovoltaic power generation

[0052] The calculation of photovoltaic power generation subsidy revenue is as follows:

[0053]

[0054] In the formula, Let t be the output power of the solar panel. The price per unit of photovoltaic output power at time t;

[0055] (3) Electricity purchase and sale costs

[0056] The load power calculation for the cloud energy storage system is as follows:

[0057]

[0058] In the formula, This represents the power exchange between the cloud energy storage operator and the power grid at time t. This represents the load power of the cloud energy storage system at time t; This represents the charging / discharging power of the stored energy at time t;

[0059] The calculation of electricity purchase and sale costs is as follows:

[0060]

[0061]

[0062] In the formula, This represents the electricity price that the main power grid purchases from the cloud energy storage system at time t. Let t be the grid-connected electricity price of the cloud energy storage system to the grid side;

[0063] Step 2-2: Establish the objective function on the power grid side.

[0064] Considering that the power grid only invests in the construction costs of cloud energy storage distributed power sources and energy storage batteries, and aims to maximize its profits, the power grid's objective function is constructed by taking the following factors as elements: the cost of power grid investment in cloud energy storage construction, the revenue from delaying distribution network losses, the revenue from delaying distribution network upgrades, the dividend income from investing in cloud energy storage, and the power grid's revenue from electricity sales and purchases. The specific objective function is as follows:

[0065]

[0066] The calculation formulas for each part are shown below:

[0067]

[0068]

[0069]

[0070] In the formula, This indicates the percentage of investment and construction costs for cloud energy storage operators. This indicates the cost of network losses in the distribution network. This indicates the electricity price that the distribution network operator purchases from the upstream power grid or power source. This represents the network access fees paid by the distribution network operator. Indicates the cost per unit power of network loss; For distribution network losses;

[0071] Benefits of delaying power grid upgrades through the construction of cloud energy storage The calculation is as follows:

[0072]

[0073] in,

[0074]

[0075]

[0076]

[0077] In the formula, The investment cost per unit capacity for expanding the power distribution network. This indicates the number of each type of energy storage battery installed. This indicates the installed capacity of each energy storage battery unit. This represents the discount rate. Indicates the inflation rate. This indicates the number of years that the upgrade and construction of the power distribution network will be delayed. This represents the annual growth rate of the load. To build a load reduction ratio after cloud energy storage, The load reduction amount at time t;

[0078] Steps 2-3, Constraints

[0079] (1) Component quantity constraint:

[0080]

[0081]

[0082]

[0083]

[0084] In the formula, , , and These represent the maximum number of wind turbines, solar photovoltaic panels, batteries, and gas turbines, respectively.

[0085] (2) Charge state constraints:

[0086]

[0087] In the formula, , These represent the lower and upper limits of charge, respectively.

[0088] (3) System operating power balance constraints:

[0089] ;

[0090] (4) Output power constraint:

[0091] ;

[0092] In the formula, This indicates the output power from cloud energy storage to the power distribution network. This indicates the upper limit of the output power from cloud energy storage to the power distribution network;

[0093] (5) Curtailment rate constraints:

[0094]

[0095]

[0096] In the formula, This indicates the planned output of the wind turbine. This indicates the actual grid-connected power of the wind turbine. This indicates the planned contribution of photovoltaic power. This indicates the actual grid-connected power of photovoltaic power. This represents the maximum allowable wind and solar curtailment rate for cloud energy storage systems.

[0097] A further improvement to the technical solution of the present invention is that the specific process of step 3 is as follows:

[0098] Step 3-1: Introduce a multi-strategy set evolutionary game model

[0099] Based on the three elements of game theory—participants, strategy set, payoff function, and the basic concepts of evolutionary game theory—an evolutionary game model between power grid and cloud energy storage operators is established.

[0100] (1) Participants: cloud energy storage operators, power grid

[0101] In evolutionary game theory analysis, the participants are biological groups. Cloud energy storage operators and the power grid are mapped as two separate populations, denoted as […]. and The population contains multiple individuals, each of which generates its own strategy and engages in randomized repeated games.

[0102] (2) Multi-strategy set

[0103] Cloud energy storage operator population With the power grid population Under constraints, n strategies are randomly generated, with the set of installed numbers for each power source and energy storage battery forming the strategy set. This represents the population of cloud energy storage operators. The strategy set is denoted as Power grid population The strategy set is denoted as The strategy set is represented as:

[0104]

[0105]

[0106] (3) Payment function

[0107] The payment function represents the economic benefits for cloud energy storage operators and the power grid under their respective strategies, and it also represents the population of cloud energy storage operators. The payment received is recorded as Power grid population The payment received is recorded as ,but:

[0108]

[0109]

[0110] (4) Replicator dynamic equation

[0111] Based on the analysis of evolutionary game modeling, a replicator dynamic equation for a multi-investor evolutionary game model is established, and the cloud energy storage operator population is considered. With the power grid population The fitness functions are expressed as follows:

[0112]

[0113]

[0114] Cloud energy storage operator population With the power grid population The average fitness is:

[0115]

[0116]

[0117] Using the proportion of individuals to the total population as a state variable, the cloud energy storage operator population... With the power grid population The replicator dynamic equations are expressed as follows:

[0118]

[0119]

[0120] Step 3-2: Solve the example using the multi-strategy set evolutionary game model. The solution steps are as follows:

[0121] (1) Randomly generate the initial participant population Randomly generated Group Policy Group;

[0122] (2) In the population One individual is randomly generated from each of the following. , And randomly select one set of strategies from the strategy set. , Calculate the payout value under this strategy. , ;

[0123] (3) Calculate individuals , In strategy , The fitness function value under the following conditions;

[0124] (4) Repeat steps (2) and (3) until... All strategy groups were selected;

[0125] (5) Calculate the population The total fitness function and the average fitness;

[0126] (6) Calculate the proportion of strategies adopted by each individual based on the replicator dynamic equation;

[0127] (7) Repeat steps (2)-(6), that is, re-perform the strategy selection process until the maximum evolution time is reached;

[0128] (8) Output the proportion of each strategy in the individual's choices. , ,as well as Evolutionary state , The strategy that yields the most stable evolutionary state under the maximum evolutionary time is the evolutionarily stable strategy.

[0129] The technological advancements achieved by this invention due to the adoption of the above technical solutions are as follows:

[0130] This invention proposes a multi-agent joint investment cloud energy storage capacity planning method based on evolutionary game theory. First, it fully analyzes the conflict of interest between cloud energy storage operators and the power grid in investing in cloud energy storage, establishing payment functions for different stakeholders and under different environments, taking into account the construction costs of cloud energy storage. Then, it solves the continuity problem of decision variables in the planning problem by using a multi-strategy set evolutionary game planning method with multiple planning strategies, and analyzes the evolutionary state of each strategy in a constantly changing game environment by establishing a replicator dynamic equation. Finally, it analyzes the evolutionary state of each strategy to obtain an evolutionarily stable strategy. This invention fully considers the investment environment, rationally plans the energy storage battery capacity, reveals the evolutionary law of the multi-strategy set in the game process, and obtains a highly stable evolutionarily stable strategy. This invention proves that the reasonable planning results not only improve the operating income of cloud energy storage operators and the power generation of distributed energy, but also increase the annual income of the distribution network, reduce the network loss of the distribution network, and achieve a win-win situation for cloud energy storage and the power grid. Attached Figure Description

[0131] Figure 1 This is a model diagram of the cloud energy storage system according to the method of the present invention;

[0132] Figure 2 This is the solution process for the intraday optimization scheduling strategy of cloud energy storage according to the method of the present invention;

[0133] Figure 3 This is a flowchart of the solution process for the multi-strategy set evolutionary game model of the present invention;

[0134] Figure 4 This is a graph showing the annual load curve of the cloud energy storage area according to the method of this invention;

[0135] Figure 5 This is an annual irradiance curve of the cloud energy storage area according to the method of the present invention;

[0136] Figure 6 This is a graph showing the annual temperature curve of the cloud energy storage area according to the method of this invention;

[0137] Figure 7 This is a graph showing the annual wind speed curve of the cloud energy storage area according to the method of this invention. Detailed Implementation

[0138] The present invention will be further described in detail below with reference to embodiments:

[0139] like Figures 1 to 2 As shown, a multi-agent joint investment cloud energy storage capacity planning method based on evolutionary game theory includes the following steps:

[0140] Step 1: Establish a daily scheduling model for the cloud energy storage system by integrating historical load, irradiance, and temperature data;

[0141] Specifically, the process of establishing the intraday scheduling model of the cloud energy storage system in step 1 is as follows:

[0142] The operation of the cloud energy storage system is uniformly scheduled by the cloud energy storage management system. The cloud energy storage management system monitors the user's electricity consumption curve through cloud batteries distributed on the user side and uploads the information to the management system. It rationally schedules the distributed power sources and energy storage batteries within the cloud energy storage system to ensure power quality and improve the user's electricity experience. With the goal of maximizing the operating profit of cloud energy storage, it optimizes the charging and discharging strategy of energy storage batteries and the output of distributed power sources by comprehensively considering historical load, irradiance, and temperature data, under the conditions of energy storage state of charge, wind and solar curtailment rate constraints, system output power and node voltage balance.

[0143] Let N be defined as a group of users in the system willing to install PVS, WTS, and ESS. The HLM module in the cloud battery calculates the energy load curve ln(h) for each user at each time period, defined as the energy that the customer needs to purchase within the time period h. Let the time scale be 1 hour. Where H = 24 is the one-day scheduling duration, and each HLM module pre-schedules the user's energy consumption at the beginning of each time scale. ln(d)(h) represents the energy load of customer n in h hours on day d. When considering long-term multi-day scheduling for capacity planning, the label (d) represents a specific day.

[0144] After HLM obtains ln(h=1), the cloud battery sends daily load information to the cloud energy storage operator. The cost of generating and distributing the electricity required by users at time h, based on the total load of the system. An assessment is conducted by the cloud energy storage operator, and a quadratic cost function is defined:

[0145]

[0146] in, It is a constant cost coefficient, let Based on the power plant's operating plan and the varying availability of intermittent energy over time, the cost function is the actual energy cost to generate the required load or the labor cost incurred by the operator in controlling and scheduling the equipment.

[0147] To facilitate advance scheduling, the service provider announces a daily pricing overview based on the obtained average cost through a communication network connecting all customers. ,therefore:

[0148]

[0149] Daily Price Overview The HLM module then re-optimizes its energy consumption plan and sends the generated load configuration information back to the service provider. Then, the pricing information is changed, and the HLM module iterates the optimization process again. It's worth noting that the issues investigated in cloud energy storage systems are not just day-ahead load management, but also long-term capacity planning, such as 10,000-day plans.

[0150] Renewable solar and wind energy are complementary and should be prioritized to meet electricity demand. Energy storage batteries are used to address energy shortages. Finally, if available resources are insufficient to meet demand, micro gas turbines are used for power generation. Cloud energy storage can be categorized into three main scenarios based on different power generation and demand levels:

[0151] (1) Power generation meets demand

[0152] The electricity generated by photovoltaic panels and wind turbines equals the sum of user load demand, that is:

[0153]

[0154]

[0155] In the formula, This represents the output power of the solar panel at time t. This represents the output power of the fan at time t. This represents the user load at time t. This represents the energy stored in the battery pack at time t. Indicates that the battery pack is in Energy storage at any time;

[0156] (2) Power generation exceeds demand

[0157] When the electricity generated by renewable energy exceeds the user's load demand, the surplus electricity will be used to charge the energy storage batteries, that is:

[0158]

[0159]

[0160] In the formula, Indicates the battery pack charging power. This indicates the charging efficiency of the battery pack. This represents the energy stored in the battery pack at time t;

[0161] (3) Power generation is less than demand

[0162] When renewable energy generation is insufficient to meet user load demand, energy storage batteries and micro gas turbines are used to discharge electricity to compensate for the power generation shortfall.

[0163]

[0164]

[0165] In the formula, Indicates the battery pack discharge power. This indicates the discharge efficiency of the battery pack.

[0166] Step 2: Taking cloud energy storage operators and the power grid as the main entities, and the planned energy storage capacity as the decision variable, a multi-entity joint investment cloud energy storage capacity planning model is established based on the objective function and constraints, with the goal of minimizing the payment of both parties.

[0167] From the perspective of cloud energy storage operators, rationally configuring distributed power sources can maximize internal operating revenue while minimizing construction costs. Considering the "self-generation and self-consumption" of local loads, cloud energy storage operators can gain more profit by building centralized energy storage facilities and aggregating distributed energy sources; however, high energy storage costs remain an unavoidable issue. Rationally configuring distributed energy sources and storage capacity within the cloud energy storage system to reduce investment costs and improve system operating revenue is a pressing problem to be solved in the planning of cloud energy storage systems.

[0168] From the perspective of the power grid, investing in cloud energy storage projects will, on the one hand, affect its revenue from electricity sales and purchases; on the other hand, investing in cloud energy storage can delay power grid upgrades and reduce grid losses. A key objective for the power grid is to rationally allocate distributed energy resources and storage capacity within the cloud energy storage system, balancing revenue and expenditure to maximize the grid's interests.

[0169] In summary, in scenarios where cloud energy storage operators and the power grid jointly invest, conflicts of interest inevitably arise between the two parties. Balancing the economic interests of multiple stakeholders and rationally planning energy storage capacity are typical areas of game theory research. Considering that all participants in the actual electricity market are boundedly rational and cannot fully grasp all the information of other participants, this chapter uses bounded rationality evolutionary game theory to study the conflicts of interest between the two parties and to rationally plan energy storage capacity.

[0170] The capacity planning process based on bounded rationality in cloud energy storage systems takes place within a system of limited information, and from a long-term development perspective, it is a process of spontaneous strategy evolution. During this process, there is no intervention in the decisions of the power grid and cloud energy storage operators; instead, the two different groups—the power grid and cloud energy storage operators—engage in dynamic interactive decision-making, ultimately spontaneously forming a stable combination of capacity planning strategies, which is termed the Long Term Evolution (ESS) under this model.

[0171] The process of establishing a multi-entity joint investment cloud energy storage capacity planning model is as follows:

[0172] Step 2-1: Establish the objective function for cloud energy storage operators.

[0173] Based on the above game-theoretic conflict analysis between cloud energy storage operators and the power grid, it can be seen that the goal of cloud energy storage operators is to minimize annual operating costs. Based on this, an objective function is established, in which the decision variables are the installation capacity of solar panels, wind turbines, micro gas turbines and energy storage batteries to be planned.

[0174]

[0175] In the formula, ANC represents the annualized cost of the cloud energy storage system. Subsidies for photovoltaic power generation, For the electricity purchase and sale fees of cloud energy storage operators, This indicates a reduction in electricity revenue. The proportion of investment by cloud energy storage operators This refers to the revenue sharing coefficient for cloud energy storage operators.

[0176] (1) Annualized cost of cloud energy storage system

[0177] The annualized cost calculation for a cloud energy storage system is shown below:

[0178]

[0179] In the formula, , , and These represent the unit costs of wind turbines, solar photovoltaic panels, energy storage batteries, and micro gas turbines, respectively. , , and These represent the total number of wind turbines, solar photovoltaic panels, and batteries, respectively.

[0180] The cost of energy storage facilities includes construction costs, recycling costs, annual operating and maintenance costs, and the total ANC value for each component is expressed as follows:

[0181]

[0182]

[0183]

[0184]

[0185] In the formula, construction cost is used express, This indicates the recovery cost, annual operating and maintenance costs. Indicates, specifically , , and These represent the construction costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. , , and These represent the annual operating and maintenance costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. , , and These represent the recycling costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively.

[0186] (2) Government subsidies for photovoltaic power generation

[0187] The calculation of photovoltaic power generation subsidy revenue is as follows:

[0188]

[0189] In the formula, Let t be the output power of the solar panel. The price per unit of photovoltaic output power at time t;

[0190] (3) Electricity purchase and sale costs

[0191] The load power calculation for the cloud energy storage system is as follows:

[0192]

[0193] In the formula, This represents the power exchange between the cloud energy storage operator and the power grid at time t. This represents the load power of the cloud energy storage system at time t; This represents the charging / discharging power of the stored energy at time t;

[0194] The calculation of electricity purchase and sale costs is as follows:

[0195]

[0196]

[0197] In the formula, This represents the electricity price that the main power grid purchases from the cloud energy storage system at time t. Let t be the grid-connected electricity price of the cloud energy storage system to the grid side;

[0198] Step 2-2: Establish the objective function on the power grid side.

[0199] Considering that the power grid only invests in the construction costs of cloud energy storage distributed power sources and energy storage batteries, and aims to maximize its profits, the power grid's objective function is constructed by taking the following factors as elements: the cost of power grid investment in cloud energy storage construction, the revenue from delaying distribution network losses, the revenue from delaying distribution network upgrades, the dividend income from investing in cloud energy storage, and the power grid's revenue from electricity sales and purchases. The specific objective function is as follows:

[0200]

[0201] The calculation formulas for each part are shown below:

[0202]

[0203]

[0204]

[0205] In the formula, This indicates the percentage of investment and construction costs for cloud energy storage operators. This indicates the cost of network losses in the distribution network. This indicates the electricity price that the distribution network operator purchases from the upstream power grid or power source. This represents the network access fees paid by the distribution network operator. Indicates the cost per unit power of network loss; For distribution network losses;

[0206] In the scenario where distribution network operators participate in the planning and investment of cloud energy storage, two mutually influential entities are formed, and their interaction impacts the construction of cloud energy storage. Game theory is applied to analyze this interaction in this scenario. The benefits of delaying distribution network upgrades due to the construction of cloud energy storage are discussed. The calculation is as follows:

[0207]

[0208] in,

[0209]

[0210]

[0211]

[0212] In the formula, The investment cost per unit capacity for expanding the power distribution network. This indicates the number of each type of energy storage battery installed. This indicates the installed capacity of each energy storage battery unit. This represents the discount rate. Indicates the inflation rate. This indicates the number of years that the upgrade and construction of the power distribution network will be delayed. This represents the annual growth rate of the load. To build a load reduction ratio after cloud energy storage, The load reduction amount at time t;

[0213] Steps 2-3, Constraints

[0214] (1) Component quantity constraint:

[0215]

[0216]

[0217]

[0218]

[0219] In the formula, , , and These represent the maximum number of wind turbines, solar photovoltaic panels, batteries, and gas turbines, respectively.

[0220] (2) Charge state constraints:

[0221]

[0222] In the formula, , These represent the lower and upper limits of charge, respectively.

[0223] (3) System operating power balance constraints:

[0224] ;

[0225] (4) Output power constraint:

[0226] ;

[0227] In the formula, This indicates the output power from cloud energy storage to the power distribution network. This indicates the upper limit of the output power from cloud energy storage to the power distribution network;

[0228] (5) Curtailment rate constraints:

[0229]

[0230]

[0231] In the formula, This indicates the planned output of the wind turbine. This indicates the actual grid-connected power of the wind turbine. This indicates the planned contribution of photovoltaic power. This indicates the actual grid-connected power of photovoltaic power. This represents the maximum allowable wind and solar curtailment rate for cloud energy storage systems.

[0232] Step 3: Solve the example using a multi-strategy set evolutionary game model to obtain the system's evolutionary stable strategy;

[0233] The specific process is as follows:

[0234] Step 3-1: Introduce a multi-strategy set evolutionary game model

[0235] Based on the three elements of game theory—participants, strategy set, payoff function, and the basic concepts of evolutionary game theory—an evolutionary game model between power grid and cloud energy storage operators is established.

[0236] (1) Participants: cloud energy storage operators, power grid

[0237] In evolutionary game theory analysis, the participants are biological groups. Cloud energy storage operators and the power grid are mapped as two separate populations, denoted as […]. and The population contains multiple individuals, each of which generates its own strategy and engages in randomized repeated games.

[0238] (2) Multi-strategy set

[0239] Cloud energy storage operator population With the power grid population Under constraints, n strategies are randomly generated, with the set of installed numbers for each power source and energy storage battery forming the strategy set. This represents the population of cloud energy storage operators. The strategy set is denoted as Power grid population The strategy set is denoted as The strategy set is represented as:

[0240]

[0241]

[0242] (3) Payment function

[0243] The payment function represents the economic benefits for cloud energy storage operators and the power grid under their respective strategies, and it also represents the population of cloud energy storage operators. The payment received is recorded as Power grid population The payment received is recorded as ,but:

[0244]

[0245]

[0246] (4) Replicator dynamic equation

[0247] Based on the analysis of evolutionary game modeling, a replicator dynamic equation for a multi-investor evolutionary game model is established, and the cloud energy storage operator population is considered. With the power grid population The fitness functions are expressed as follows:

[0248]

[0249]

[0250] Cloud energy storage operator population With the power grid population The average fitness is:

[0251]

[0252]

[0253] Using the proportion of individuals to the total population as a state variable, the cloud energy storage operator population... With the power grid population The replicator dynamic equations are expressed as follows:

[0254]

[0255]

[0256] Step 3-2: Solving the example using the multi-strategy set evolutionary game model. The method for finding the stable strategy in the multi-strategy set evolutionary game is based on evolutionary game theory. The specific solution process is shown in the appendix. Figure 3 .

[0257] (1) Randomly generate the initial participant population , Randomly generated Group Policy Group;

[0258] (2) In the population , One individual is randomly generated from each of the following. , And randomly select one set of strategies from the strategy set. , Calculate the payout value under this strategy. , ;

[0259] (3) Calculate individuals , In strategy , The fitness function value under the following conditions;

[0260] (4) Repeat steps (2) and (3) until... All strategy groups were selected;

[0261] (5) Calculate the population , The total fitness function and the average fitness;

[0262] (6) Calculate the proportion of strategies adopted by each individual based on the replicator dynamic equation;

[0263] (7) Repeat steps (2)-(6), that is, re-perform the strategy selection process until the maximum evolution time is reached;

[0264] (8) Output the proportion of each strategy in the individual's choices. , ,as well as Evolutionary state , The strategy that yields the most stable evolutionary state under the maximum evolutionary time is the evolutionarily stable strategy.

[0265] First, the annual load, irradiance, temperature, and wind speed curves for the area where the cloud energy storage system is installed are provided in the attached figures. Figure 4 Appendix Figure 5 Appendix Figure 6 Appendix Figure 7 As shown in Table 1, the relevant parameters of the internal energy storage battery in the cloud energy storage system of the present invention are presented in Table 1.

[0266] Table 1. Parameters of the internal energy storage battery in the cloud energy storage system

[0267]

[0268] Table 1 (continued)

[0269]

[0270] Step 4: Conduct simulation analysis based on multi-strategy set evolutionary game.

[0271] To illustrate the economic efficiency and effectiveness of this invention, simulations were performed using the following four scenarios:

[0272] Scenario 1: Cloud energy storage operators invest in cloud energy storage independently;

[0273] Scenario 2: Cloud energy storage operators and power grids jointly invest in and operate energy storage, employing simple multi-objective optimization;

[0274] Scenario 3: Cloud energy storage operators and power grids jointly invest in and operate energy storage, employing traditional cooperative game theory;

[0275] Scenario 4: Cloud energy storage operators and power grids jointly invest in and operate energy storage, using evolutionary game theory.

[0276] The optimized calculation results are shown in Tables 2 and 3.

[0277] Table 2 Comparison of planning results in different scenarios

[0278]

[0279] Table 3 Distributed power generation

[0280]

[0281] It can be seen that the proposed multi-stakeholder joint investment cloud energy storage capacity planning method based on evolutionary game theory can maximize the interests of both the grid side and cloud energy storage operators while reducing wind and solar curtailment and improving the absorption capacity of new energy sources. This verifies the economic efficiency and effectiveness of the proposed method.

[0282] The simulation analysis described above is merely a description of preferred embodiments of the present invention and is not intended to limit the scope of the invention. Without departing from the spirit of the invention, those skilled in the art can make their own interpretations.

[0283] All modifications and improvements to the technical solution of this invention shall fall within the protection scope of this invention.

Claims

1. An evolutionary game-based multi-agent joint investment cloud energy storage capacity planning method, characterized in that: Includes the following steps: Step 1: Establish a daily scheduling model for the cloud energy storage system by integrating historical load, irradiance, and temperature data; The process of establishing the intraday scheduling model of the cloud energy storage system in step 1 is as follows: The operation of the cloud energy storage system is uniformly scheduled by the cloud energy storage management system. The cloud energy storage management system monitors the user's electricity consumption curve through cloud batteries distributed on the user side and uploads the information to the management system. It makes reasonable scheduling of distributed power sources and energy storage batteries within the cloud energy storage system to ensure power quality and improve the user's power experience. With the goal of maximizing the operating profit of cloud energy storage, it optimizes the charging and discharging strategy of energy storage batteries and the output of distributed power sources by comprehensively considering historical load, irradiance, and temperature data, under the conditions of energy storage state of charge, wind and solar curtailment rate constraints, system output power and node voltage balance, using Cplex. Let N be the set of users in the system who are willing to install PVS, WTS and ESS, the HLM module in the cloud battery calculates the energy load curve of each user in each time period , defined as the energy that the customer needs to purchase in time period h, assuming the time scale is 1 hour, that is , where H = 24 is the length of the daily schedule, and each HLM module schedules the user's energy consumption in advance at the beginning of each time scale, , represents the energy load of customer n at h hours on day d, when considering long-term multi-day scheduling of capacity planning, the label (d) indicates a specific day; After obtaining ln(h=1) in HLM, the cloud battery sends daily load information to the cloud energy storage operator, set Total load at h moment, the cost of generating and distributing the power demand required by users Assessed by the cloud energy storage operator, set a quadratic cost function: wherein, is a constant cost coefficient, set The cost function is the actual energy cost to produce the required load or the labor cost used by the operator to control, schedule the devices depending on the power plant operation plan and the availability of intermittent energy sources over time; To facilitate advance scheduling, the service provider announces a daily pricing overview based on the obtained average cost through a communication network connecting all customers. ,therefore: Daily Price Overview The HLM module re-optimizes its energy consumption plan and sends the generated load configuration information back to the service provider. Then, the pricing information is changed, and the HLM module iterates the optimization process again. Renewable solar and wind energy are complementary and should be prioritized to meet electricity demand. Energy storage batteries are used to address energy shortages. Finally, if available resources are insufficient to meet demand, micro gas turbines are used for power generation. Cloud energy storage can be categorized into three scenarios based on different power generation and demand levels: (1) Power generation meets demand The electricity generated by photovoltaic panels and wind turbines equals the sum of user load demand, that is: In the formula, This represents the output power of the solar panel at time t. This represents the output power of the fan at time t. This represents the user load at time t. This represents the energy stored in the battery pack at time t. Indicates that the battery pack is in Energy storage at any time; (2) Power generation exceeds demand When the electricity generated by renewable energy exceeds the user's load demand, the surplus electricity will be used to charge the energy storage batteries, that is: In the formula, Indicates the battery pack charging power. This indicates the charging efficiency of the battery pack. This represents the energy stored in the battery pack at time t; (3) Power generation is less than demand When renewable energy generation is insufficient to meet user load demand, energy storage batteries and micro gas turbines are used to discharge electricity to compensate for the power generation shortfall. In the formula, Indicates the battery pack discharge power. The discharge efficiency of the battery pack is represented; Step 2, taking cloud energy storage operators and the power grid as the main entities, and the planned energy storage capacity as the decision variable, starting from the objective function and constraints, and aiming to minimize the payment of both parties, a multi-entity joint investment cloud energy storage capacity planning model is established. Step 3: Solve the example using a multi-strategy set evolutionary game model to obtain the system's evolutionary stable strategy; Step 4: Conduct simulation analysis based on multi-strategy set evolutionary game.

2. The method for multi-agent joint investment in cloud energy storage capacity planning based on evolutionary game theory as described in claim 1, characterized in that: The process of establishing the multi-entity joint investment cloud energy storage capacity planning model in step 2 is as follows: Step 2-1: Establish the objective function for cloud energy storage operators. The goal of operator cloud energy storage is to minimize annual operating costs. Based on this, an objective function is established, in which the decision variables are the installation capacity of the solar panels, wind turbines, micro gas turbines and energy storage batteries to be planned. In the formula, ANC represents the annualized cost of the cloud energy storage system. Subsidies for photovoltaic power generation, For the electricity purchase and sale fees of cloud energy storage operators, This indicates a reduction in electricity revenue. The proportion of investment by cloud energy storage operators This refers to the revenue sharing coefficient for cloud energy storage operators. (1) Annualized cost of cloud energy storage system The annualized cost calculation for a cloud energy storage system is shown below: In the formula, , , and These represent the unit costs of wind turbines, solar photovoltaic panels, energy storage batteries, and micro gas turbines, respectively. , , and These represent the total number of wind turbines, solar photovoltaic panels, and batteries, respectively. The cost of energy storage facilities includes construction costs, recycling costs, annual operating and maintenance costs, and the total ANC value for each component is expressed as follows: In the formula, , , and These represent the construction costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. , , and These represent the annual operating and maintenance costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. , , and These represent the recycling costs of wind turbines, photovoltaic systems, energy storage batteries, and micro gas turbines, respectively. (2) Government subsidies for photovoltaic power generation The calculation of photovoltaic power generation subsidy revenue is as follows: In the formula, Let t be the output power of the solar panel. The price per unit of photovoltaic output power at time t; (3) Electricity purchase and sale costs The load power calculation for the cloud energy storage system is as follows: In the formula, This represents the power exchange between the cloud energy storage operator and the power grid at time t. This represents the load power of the cloud energy storage system at time t; This represents the charging / discharging power of the stored energy at time t; The calculation of electricity purchase and sale costs is as follows: In the formula, This represents the electricity price that the main power grid purchases from the cloud energy storage system at time t. Let t be the grid-connected electricity price of the cloud energy storage system to the grid side; Step 2-2: Establish the objective function on the power grid side. Considering that the power grid only invests in the construction costs of cloud energy storage distributed power sources and energy storage batteries, and aims to maximize its profits, the power grid's objective function is constructed by taking the following factors as elements: the cost of power grid investment in cloud energy storage construction, the revenue from delaying distribution network losses, the revenue from delaying distribution network upgrades, the dividend income from investing in cloud energy storage, and the power grid's revenue from electricity sales and purchases. The specific objective function is as follows: The calculation formulas for each part are shown below: In the formula, This indicates the percentage of investment and construction costs for cloud energy storage operators. This indicates the cost of network losses in the distribution network. This indicates the electricity price that the distribution network operator purchases from the upstream power grid or power source. This represents the network access fees paid by the distribution network operator. Indicates the cost per unit power of network loss; For distribution network losses; Benefits of delaying power grid upgrades through the construction of cloud energy storage The calculation is as follows: in, In the formula, The investment cost per unit capacity for expanding the power distribution network. This indicates the number of each type of energy storage battery installed. This indicates the installed capacity of each energy storage battery unit. This represents the discount rate. Indicates the inflation rate. This indicates the number of years that the upgrade and construction of the power distribution network will be delayed. This represents the annual growth rate of the load. To build a load reduction ratio after cloud energy storage, The load reduction amount at time t; Steps 2-3, Constraints (1) Component quantity constraint: In the formula, , , and These represent the maximum number of wind turbines, solar photovoltaic panels, batteries, and gas turbines, respectively. (2) Charge state constraints: In the formula, , These represent the lower and upper limits of charge, respectively. (3) System operating power balance constraints: ； (4) Output power constraint: ； In the formula, This indicates the output power from cloud energy storage to the power distribution network. This indicates the upper limit of the output power from cloud energy storage to the power distribution network; (5) Curtailment rate constraints: In the formula, This indicates the planned output of the wind turbine. This indicates the actual grid-connected power of the wind turbine. This indicates the planned contribution of photovoltaic power. This indicates the actual grid-connected power of photovoltaic power. This represents the maximum allowable wind and solar curtailment rate for cloud energy storage systems.

3. The multi-agent joint investment cloud energy storage capacity planning method based on evolutionary game theory as described in claim 2, characterized in that: The specific process of step 3 is as follows: Step 3-1: Introduce a multi-strategy set evolutionary game model Based on the three elements of game theory—participants, strategy set, payoff function, and the basic concepts of evolutionary game theory—an evolutionary game model between power grid and cloud energy storage operators is established. (1) Participants: cloud energy storage operators, power grid In evolutionary game theory analysis, the participants are biological groups. Cloud energy storage operators and the power grid are mapped as two separate populations, denoted as […]. and The population contains multiple individuals, each of which generates its own strategy and engages in randomized repeated games. (2) Multi-strategy set Cloud energy storage operator population With the power grid population Under constraints, n strategies are randomly generated, with the set of installed numbers for each power source and energy storage battery forming the strategy set. This represents the population of cloud energy storage operators. The strategy set is denoted as Power grid population The strategy set is denoted as The strategy set is represented as: (3) Payment function The payment function represents the economic benefits for cloud energy storage operators and the power grid under their respective strategies, and it also represents the population of cloud energy storage operators. The payment received is recorded as Power grid population The payment received is recorded as ,but: (4) Replicator dynamic equation Based on the analysis of evolutionary game modeling, a replicator dynamic equation for a multi-investor evolutionary game model is established, and the cloud energy storage operator population is considered. With the power grid population The fitness functions are expressed as follows: Cloud energy storage operator population With the power grid population The average fitness is: Using the proportion of individuals to the total population as a state variable, the cloud energy storage operator population... With the power grid population The replicator dynamic equations are expressed as follows: Step 3-2: Solve the example using the multi-strategy set evolutionary game model. The solution steps are as follows: (1) Randomly generate the initial participant population , Randomly generated Group Policy Group; (2) In the population , One individual is randomly generated from each of the following. , And randomly select one set of strategies from the strategy set. , Calculate the payout value under this strategy. , ; (3) Calculate individuals , In strategy , The fitness function value under the following conditions; (4) Repeat steps (2) and (3) until... All strategy groups were selected; (5) Calculate the population , The total fitness function and the average fitness; (6) Calculate the proportion of strategies adopted by each individual based on the replicator dynamic equation; (7) Repeat steps (2)-(6), that is, re-perform the strategy selection process until the maximum evolution time is reached; (8) Output the proportion of each strategy in the individual's choices. , ,as well as Evolutionary state , The strategy that yields the most stable evolutionary state under the maximum evolutionary time is the evolutionarily stable strategy.

Images