A shared energy storage selection method based on multi-attribute decision making and multi-agent fusion

Abstract

The present invention relates to a shared energy storage selection method based on multi-attribute decision-making and multi-agent fusion, the method comprising the following steps: establishing a shared energy storage and user operation control system model, and solving it through linear programming; on the basis of the system model, setting shared energy storage selection indicators from the two aspects of shared energy storage and users, in order to evaluate the pros and cons of different shared energy storage types; evaluating the shared energy storage selection indicators from the two aspects of shared energy storage and users through a multi-criteria compromise solution ranking method, and obtaining the evaluated scores; fusing the indicator values of each shared energy storage and user for each shared energy storage selection scheme, and selecting the optimal shared energy storage type after comprehensive consideration of each subject. The patent of the present invention comprehensively considers that shared energy storage serves multiple subjects, combines the multi-criteria compromise solution ranking method with evidence theory, and performs shared energy storage selection, which has important significance and application value for the research and promotion of shared energy storage.

Description

Technical field:

[0001] The present invention relates to the field of distributed energy technology, and in particular to a shared energy storage selection method based on multi-attribute decision-making and multi-agent fusion. Background technology:

[0002] Currently, the proportion of renewable energy, represented by wind and photovoltaic power, continues to increase, leading to curtailment of wind and solar power. Energy storage technology is an effective means of addressing this issue. By storing and releasing electricity, it can recover the energy from curtailed wind and solar power, and also achieve peak-valley arbitrage profits through low-charge and high-discharge periods. Under the independent investment model, users typically require long-term loans, which hinders the rapid deployment and application of energy storage systems. The shared model allows equipment to be leased to users. Drawing on the concept of the sharing economy, this model allows owners of distributed or centralized energy storage to not only use the energy storage equipment for their own needs but also to transfer it to third parties. This allows the energy storage equipment to be centrally dispatched, provide ancillary services for the system, and receive capacity fees. Given the current high energy storage costs and long payback periods, this business model is worthy of exploration. Due to the wide variety of battery types, different users of shared energy storage services may have different requirements for battery types. Therefore, selecting the appropriate battery type to fully leverage the advantages of shared energy storage plays a crucial role in the promotion and application of shared energy storage systems.

[0003] Multi-criteria compromise ranking methods are used when decision makers cannot or do not know how to accurately express their preferences, when there are conflicts and incommensurability (different measurement units) between evaluation criteria, and when the decision makers dealing with the conflicting issues are willing to accept a compromise solution. They simultaneously consider maximizing group utility and minimizing individual regret, as well as incorporating the decision makers' subjective preferences. They offer high ranking stability and reliability. Evidential theory is also widely used in expert systems and information fusion. Summary of the invention:

[0004] The present invention aims to provide a shared energy storage selection method based on multi-attribute decision making and multi-agent fusion. The technical solution adopted by the present invention is:

[0005] A shared energy storage selection method based on multi-attribute decision-making and multi-agent fusion includes the following steps:

[0006] Step 1: Establish a shared energy storage and user operation control system model, including the objective function and constraints, and solve it through linear programming. Specifically, it includes:

[0007] Step 1.1: Establish a shared energy storage operation control model. The objective function is to minimize the investment and operation cost of shared energy storage. The constraints are the shared energy storage charge and discharge power constraints and the shared energy storage capacity constraints.

[0008] Step 1.2: Establish a user operation control model. The objective function is to minimize the cost of purchasing electricity from the grid. The constraints are the charging and discharging power constraints of the shared energy storage leased by the user and the user power balance constraints.

[0009] Step 2: Different shared energy storage types have different operational control results. Based on the system model, set shared energy storage selection indicators from two aspects: shared energy storage and users, to evaluate the advantages and disadvantages of different shared energy storage types. Specifically, the indicators include:

[0010] Step 2.1: Based on the operational control results obtained in the first step, establish a corresponding indicator system for shared energy storage, including the annual investment and operating costs of each shared energy storage, the cost of purchasing electricity and charging, the income from leasing shared energy storage to users, and the residual value income;

[0011] Step 2.2: Based on the operational control results obtained in the first step, establish a corresponding indicator system for users, including: the user's income from reducing electricity purchase costs from the grid, the income from selling abandoned wind or solar energy, and the cost of leasing shared energy storage;

[0012] Step 3: Evaluate the shared energy storage selection indicators from the perspectives of shared energy storage and users using a multi-criteria compromise solution ranking method to obtain the evaluation scores; specifically, the following are included:

[0013] Step 3.1: Build evaluation models for each shared energy storage and user, establish a set of shared energy storage type alternatives, a set of evaluation attributes, and a weight vector of the attributes;

[0014] Take a shared energy storage or user as an example, let P = {1, 2, ..., p}, Q = {1, 2, ..., q};

[0015] A={A1,A2,…,A p} represents a set of p shared energy storage type schemes, each of which includes a battery type, and there is no intersection between the schemes, where A s represents the s-th shared energy storage type scheme, s∈P;

[0016] C={C1,C2,…,C q} represents a set of q attributes, i.e., the corresponding indicators, and the indicators obtained in step 2 are sorted, where C j represents the jth attribute, i.e. the jth indicator, j∈Q;

[0017] ω={ω1,ω2,…,ω q} represents the weight vector of the attribute, where ω j For attribute C j The weights satisfy ω j ≥0 and ω j The sum of is equal to 1;

[0018] x s,j For Plan A s For attribute C j As a result, each shared energy storage and user builds their own evaluation model according to this step;

[0019] Step 3.2: Standardize the shared energy storage and user indicators obtained in step 2 to eliminate the impact of dimension on the final results;

[0020] Step 3.3: Determine the group utility and individual regret of each shared energy storage selection scheme for each shared energy storage and user entity, and calculate the value of the decision-making index for each shared energy storage selection scheme. The smaller the index value, the better the scheme. The optimal scheme for different entities may be different.

[0021] Step 4: Combine the shared energy storage obtained in step 3 with the user's index values for each shared energy storage selection scheme, and select the optimal shared energy storage type after comprehensively considering all subjects; specifically, including:

[0022] Step 4.1: Based on the index values of each shared energy storage and user for each shared energy storage selection scheme obtained in step 3, solve the basic probability distribution of each one respectively;

[0023] Step 4.2: The total number of shared energy storage and users is n. The basic probability distribution of n shared energy storage and user entities is integrated according to the synthesis rule of evidence theory to obtain the final result. The shared energy storage type corresponding to the smaller value in the result is selected. This shared energy storage type is the optimal solution.

[0024] Compared with the closest prior art, the excellent effects of the present invention are:

[0025] In the technical solution of the present invention, the multi-criteria compromise solution ranking method is used in situations where decision makers cannot or do not know how to accurately express their preferences, there are conflicts and incommensurability between evaluation criteria, and the decision makers dealing with conflicting issues can accept compromise solutions. It can simultaneously consider group utility maximization and individual regret minimization, as well as incorporate the decision maker's subjective preferences, and has high ranking stability and credibility. In many situations, it is necessary to comprehensively consider uncertain information from multiple sources to complete the solution of the problem. The combination rules of evidence theory can play an important role in solving this problem. Evidence theory is also widely used in expert systems, information fusion and other fields. This patent considers that the battery selection process involves a large number of fuzzy factors, which are greatly influenced by personal subjective factors, and there are also conflicting indicators in the decision-making process. By combining the multi-criteria compromise solution ranking method with evidence theory, through the analysis of the operating characteristics of user-rented shared energy storage, a battery selection decision-making index system is constructed. The indicators of different aspects of each subject are comprehensively considered to obtain a corresponding comprehensive score, and the optimal shared energy storage battery selection solution is obtained, which is conducive to the research and promotion of shared energy storage. Description of the drawings:

[0026] Figure 1 It is a flow chart of a shared energy storage selection method based on multi-attribute decision-making and multi-agent fusion.

[0027] Figure 2 It is a flow chart of the shared energy storage and user multi-criteria compromise solution sorting method. Specific implementation method:

[0028] Example:

[0029] A shared energy storage selection method based on multi-attribute decision-making and multi-agent fusion includes the following steps:

[0030] Step 1: Establish a shared energy storage and user operation control system model, including the objective function and constraints, and solve it through linear programming. The specific process is as follows:

[0031] Step 1.1: Establish a shared energy storage operation control model. The objective function is to minimize the investment and operation cost of shared energy storage. The constraints are the shared energy storage charge and discharge power constraints and the shared energy storage capacity constraints.

[0032] Objective function:

[0033]

[0034] Constraints:

[0035] U k，min ≤U k，t ≤U k，max ,

[0036]

[0037]

[0038]

[0039] U k，0 =U k，T

[0040] Where C ess，k is the investment and operating cost of shared energy storage k, is the charging power of shared energy storage k in period t, is the discharge power of shared energy storage k in period t, α k k is the shared energy storage investment and operation and maintenance cost per unit power, U k，t is the capacity of shared energy storage k in period t, U min，k is the minimum capacity of shared energy storage k, U max，k is the maximum capacity of shared energy storage k, is the maximum charging power of shared energy storage k, is the maximum discharge power of shared energy storage k, x k，t is a 0-1 variable to ensure that charging and discharging are not performed at the same time, β k is the self-loss coefficient of the shared energy storage device k, η k is the charging and discharging efficiency of shared energy storage k, U k，0 is the initial capacity of shared energy storage k, U k，T is the ending capacity of shared energy storage k;

[0041] Step 1.2: Establish a user operation control model. The objective function is to minimize the cost of purchasing electricity from the grid. The constraints are the charging and discharging power constraints of the shared energy storage leased by the user and the user power balance constraints.

[0042] Objective function:

[0043]

[0044] Constraints:

[0045]

[0046]

[0047]

[0048]

[0049]

[0050] Where C user，i is the cost of purchasing electricity from the grid for user i, I is the total number of users, is the electricity purchase price during period t, is the power purchased by user i from the upper power grid during period t, is the total load of user i in period t, is the photovoltaic output used by user i in period t, is the wind power output used by user i in period t, Leasing shared energy storage for users to charge their own power, Lease shared energy storage for users to discharge their own power, The photovoltaic output of user i in period t, The wind power output allocated to user i in period t, The maximum power of the shared energy storage leased to user i to charge itself, The maximum power that user i can discharge from the shared energy storage he rents, x i，t It is a 0-1 variable to ensure that charging and discharging are not performed at the same time;

[0051]

[0052] Where K e is the total amount of shared energy storage;

[0053] Step 2: Different shared energy storage types have different operational control results. Based on the system model, set shared energy storage selection indicators from two aspects: shared energy storage and users, to evaluate the advantages and disadvantages of different shared energy storage types;

[0054] Step 2.1: Based on the operational control results obtained in the first step, establish a corresponding indicator system for shared energy storage, including the annual investment and operating costs of each shared energy storage, the cost of purchasing electricity and charging, the income from leasing shared energy storage to users, and the residual value income. Some indicators are as follows:

[0055]

[0056]

[0057]

[0058]

[0059] Where, D is the k-year investment and operation cost of shared energy storage, ay The number of days per year that shared energy storage operates, is the electricity purchase cost for shared energy storage in k years, T is the total number of time periods in a day, is the abandoned wind or solar power output purchased by the shared energy storage k from the user during period t, μ is the unit price of the abandoned wind or solar power purchased by the shared energy storage from the user, is the power purchased by shared energy storage k from the grid during period t, σ t is the price of electricity purchased from the grid during period t by the shared energy storage, is the income from leasing shared energy storage k to users, M ess is the unit price for leasing shared energy storage, is the residual income of shared energy storage k, is the initial investment cost of shared energy storage k, R v，k is the residual value rate of shared energy storage k;

[0060] Step 2.2: Based on the operational control results obtained in the first step, establish a corresponding indicator system for users, including the user's income from reducing electricity purchase costs from the grid, the income from selling abandoned wind or solar energy, and the cost of leasing shared energy storage. Some indicators are as follows:

[0061]

[0062]

[0063]

[0064] Where, Reduce the cost of purchasing electricity from the power grid for user i, The power purchased by user i from the upper grid during period t when he does not rent the shared energy storage. The income from selling abandoned wind or solar energy for user i, The cost of leasing shared energy storage for user i;

[0065] Step 3: Shared energy storage and user multi-criteria compromise solution ranking method, evaluate the shared energy storage selection indicators from two aspects, shared energy storage and user, using the multi-criteria compromise solution ranking method to obtain the evaluation scores;

[0066] Step 3.1: Build evaluation models for each shared energy storage and user, establish a set of shared energy storage type alternatives, a set of evaluation attributes, and a weight vector of the attributes;

[0067] Taking a shared energy storage or user as an example, let P = {1, 2, ..., p}, Q = {1, 2, ..., q};

[0068] A={A1,A2,...,A p} represents a set of p shared energy storage type schemes, each of which includes a battery type, and there is no intersection between the schemes, where A s represents the s-th shared energy storage type scheme, s∈P;

[0069] C={C1,C2,...,Cq} represents a set of q attributes, i.e., the corresponding indicators, and the indicators obtained in step 2 are sorted, where C j represents the jth attribute, i.e. the jth indicator, j∈Q;

[0070] ω={ω1,ω2,...,ω q} represents the weight vector of the attribute, where ω j For attribute C j The weights satisfy ω j ≥0 and ω j The sum of is equal to 1;

[0071] x s，j For Plan A s For attribute C j As a result, the decision evaluation matrix is shown in the following table. Each shared energy storage and user builds their own evaluation model according to this step;

[0072] Decision Evaluation Matrix

[0073] <![CDATA[C1]]> <![CDATA[C2]]> <![CDATA[C j ]]> <![CDATA[C q ]]> ω <![CDATA[ω1]]> <![CDATA[ω2]]> <![CDATA[ω j ]]> <![CDATA[ω q ]]> <![CDATA[A1]]> <![CDATA[X 1，1 ]]> <![CDATA[X 1，2 ]]> <![CDATA[X 1，j ]]> <![CDATA[X 1，q ]]> <![CDATA[A2]]> <![CDATA[X 2，1 ]]> <![CDATA[X 2，2 ]]> <![CDATA[X 2，j ]]> <![CDATA[X 2，q ]]> <![CDATA[A s ]]> <![CDATA[X s，1 ]]> <![CDATA[X s，2 ]]> <![CDATA[X s，j ]]> <![CDATA[X s，q ]]> <![CDATA[A p ]]> <![CDATA[X p， 1]]> <![CDATA[X p，2 ]]> <![CDATA[X p，j ]]> <![CDATA[X p，q ]]>

[0074] Step 3.2: Standardize the shared energy storage and user indicators obtained in step 2 to eliminate the impact of dimension on the final results;

[0075] Take a shared energy storage or user as an example, let x s，j As its decision indicator, r s，j As the result of the normalization of its indicators, the indicators are normalized by the following formula:

[0076] If the shared energy storage and the user's decision indicator x s，j For benefit indicators:

[0077]

[0078] If the shared energy storage and the user's decision indicator x s，j For cost indicators:

[0079]

[0080] in

[0081]

[0082]

[0083] Each shared energy storage and user standardizes their respective indicators according to this step;

[0084] Step 3.3: Determine the group utility and individual regret of each shared energy storage selection scheme for each shared energy storage and user entity, and calculate the value of the decision-making index for each shared energy storage selection scheme. The smaller the index value, the better the scheme. The optimal scheme for different entities may be different.

[0085] Take a shared energy storage or user as an example, let and The maximum and minimum values of each column in its standardized decision evaluation matrix are:

[0086]

[0087]

[0088] Calculate the group utility value S of its shared energy storage selection scheme s and individual regret value R s :

[0089]

[0090]

[0091] According to the results of the group utility value and individual regret value of the shared energy storage selection scheme, the value of the decision index of the shared energy storage selection scheme is calculated. The smaller the index value, the better the scheme:

[0092]

[0093] in,

[0094]

[0095]

[0096]

[0097]

[0098] In the formula, v represents the coefficient of the shared energy storage selection decision mechanism. If v>0.5, it means that the decision is made according to the decision mechanism of maximizing the group effect of shared energy storage selection. If v<0.5, it means that the decision is made according to the decision mechanism of minimizing the individual regret value of shared energy storage selection. If v=0.5, it means that the decision is made according to the decision mechanism of maximizing the group effect and minimizing the individual regret value of shared energy storage selection, which are equally important. Finally, the index value of each shared energy storage selection scheme can be obtained, and the smaller the index value, the better.

[0099] Each shared energy storage and user follows this step to obtain their own index value for each shared energy storage selection scheme;

[0100] Step 4: Combine the shared energy storage obtained in step 3 with the user's index values for each shared energy storage selection scheme, and select the optimal shared energy storage type after comprehensive consideration of all subjects;

[0101] Step 4.1: Based on the index values of each shared energy storage and user for each shared energy storage selection scheme obtained in step 3, solve the basic probability distribution of each one respectively;

[0102] A is a shared energy storage selection scheme. For each shared energy storage and user, the basic probability distribution on A is an m-function that satisfies:

[0103] m(φ)=0.

[0104]

[0105] Among them, taking a shared energy storage or user as an example, let a1, a2, ..., a p Represents the index value for each shared energy storage selection scheme, where a s represents the index value of the s-th shared energy storage type scheme, s∈P,b1,b2,...,b p represents the basic probability distribution for each shared energy storage selection scheme, where b s Denotes the basic probability distribution of the s-th shared energy storage type scheme, s∈P, then:

[0106]

[0107] Each shared energy storage and user obtains their own basic probability distribution for each shared energy storage selection scheme according to this step;

[0108] Step 4.2: According to the synthesis rule of evidence theory, the basic probability distribution of each shared energy storage and user is integrated to obtain the final result, thereby obtaining the selection of shared energy storage;

[0109] Let the total number of shared energy storage and users be n;

[0110]

[0111]

[0112] According to the results of evidence theory fusion, the shared energy storage type corresponding to the smaller value in the result is selected. This shared energy storage type is the optimal solution.

Claims

1. A shared energy storage selection method based on multi-attribute decision making and multi-agent fusion, characterized in that: The steps include: Step 1: Establish a shared energy storage and user operation control system model, including the objective function and constraints, and solve it through linear programming. Specifically, it includes: Step 1.1: Establish a shared energy storage operation control model. The objective function is to minimize the investment and operation cost of shared energy storage. The constraints are the shared energy storage charge and discharge power constraints and the shared energy storage capacity constraints. Step 1.2: Establish a user operation control model. The objective function is to minimize the cost of purchasing electricity from the grid. The constraints are the charging and discharging power constraints of the shared energy storage leased by the user and the user power balance constraints. Step 2: Different shared energy storage types have different operational control results. Based on the system model, set shared energy storage selection indicators from two aspects: shared energy storage and users, to evaluate the advantages and disadvantages of different shared energy storage types. Specifically, the indicators include: Step 2.1: Based on the operational control results obtained in the first step, establish a corresponding indicator system for shared energy storage, including the annual investment and operating costs of each shared energy storage, the cost of purchasing electricity and charging, the income from leasing shared energy storage to users, and the residual value income; Step 2.2: Based on the operational control results obtained in the first step, establish a corresponding indicator system for users, including: the user's income from reducing electricity purchase costs from the grid, the income from selling abandoned wind or solar energy, and the cost of leasing shared energy storage; Step 3: Evaluate the shared energy storage selection indicators from the perspectives of shared energy storage and users using a multi-criteria compromise solution ranking method to obtain the evaluation scores; specifically, the following are included: Step 3.1: Build evaluation models for each shared energy storage and user, establish a set of shared energy storage type alternatives, a set of evaluation attributes, and a weight vector of the attributes; Take a shared energy storage or user as an example, let P = {1, 2, ..., p}, Q = {1, 2, ..., q}; A={A1,A2,…,A p } represents a set of p shared energy storage type schemes, each of which includes a battery type, and there is no intersection between the schemes, where A s represents the s-th shared energy storage type scheme, s∈P; C={C1,C2,…,C q } represents a set of q attributes, i.e., the corresponding indicators, and the indicators obtained in step 2 are sorted, where C j represents the jth attribute, i.e. the jth indicator, j∈Q; ω={ω1,ω2,…,ω q } represents the weight vector of the attribute, where ω j For attribute C j The weights satisfy ω j ≥0 and ω j The sum of is equal to 1; x s,j For Plan A s For attribute C j As a result, each shared energy storage and user builds their own evaluation model according to this step; Step 3.2: Standardize the shared energy storage and user indicators obtained in step 2 to eliminate the impact of dimension on the final results; Step 3.3: Determine the group utility and individual regret of each shared energy storage selection scheme for each shared energy storage and user entity, and calculate the value of the decision-making index for each shared energy storage selection scheme. The smaller the index value, the better the scheme. The optimal scheme for different entities may be different. Step 4: Combine the shared energy storage obtained in step 3 with the user's index values for each shared energy storage selection scheme, and select the optimal shared energy storage type after comprehensively considering all subjects; specifically, including: Step 4.1: Based on the index values of each shared energy storage and user for each shared energy storage selection scheme obtained in step 3, solve the basic probability distribution of each one respectively; Step 4.2: The total number of shared energy storage and users is n. The basic probability distribution of n shared energy storage and user entities is integrated according to the synthesis rule of evidence theory to obtain the final result. The shared energy storage type corresponding to the smaller value in the result is selected. This shared energy storage type is the optimal solution.

2. The shared energy storage selection method based on multi-attribute decision-making and multi-agent fusion according to claim 1 is characterized in that: The steps include: Step 1: Establish a shared energy storage and user operation control system model, including the objective function and constraints, and solve it through linear programming. The specific process is as follows: Step 1.1: Establish a shared energy storage operation control model. The objective function is to minimize the investment and operation cost of shared energy storage. The constraints are the shared energy storage charge and discharge power constraints and the shared energy storage capacity constraints. Objective function: Constraints: IN k,min ≤U k,t ≤U k,max , IN k,0 =U k,T , Where C ess,k is the investment and operating cost of shared energy storage k, is the charging power of shared energy storage k in period t, is the discharge power of shared energy storage k in period t, α k k is the shared energy storage investment and operation and maintenance cost per unit power, U k,t is the capacity of shared energy storage k in period t, U min,k is the minimum capacity of shared energy storage k, U max,k is the maximum capacity of shared energy storage k, is the maximum charging power of shared energy storage k, is the maximum discharge power of shared energy storage k, x k,t is a 0-1 variable to ensure that charging and discharging are not performed at the same time, β k is the self-loss coefficient of the shared energy storage device k, η k is the charging and discharging efficiency of shared energy storage k, U k,0 is the initial capacity of shared energy storage k, U k,T is the ending capacity of shared energy storage k; Step 1.2: Establish a user operation control model. The objective function is to minimize the cost of purchasing electricity from the grid. The constraints are the charging and discharging power constraints of the shared energy storage leased by the user and the user power balance constraints. Objective function: Constraints: Where C user,i is the cost of purchasing electricity from the grid for user i, I is the total number of users, is the electricity purchase price during period t, is the power purchased by user i from the upper power grid during period t, is the total load of user i in period t, is the photovoltaic output used by user i in period t, is the wind power output used by user i in period t, Leasing shared energy storage for users to charge their own power, Lease shared energy storage for users to discharge their own power, The photovoltaic output of user i in period t, The wind power output allocated to user i in period t, The maximum power of the shared energy storage leased to user i to charge itself, The maximum power that user i can discharge from the shared energy storage he rents, x i,t It is a 0-1 variable to ensure that charging and discharging are not performed at the same time; Where K e is the total amount of shared energy storage; Step 2: Different shared energy storage types have different operational control results. Based on the system model, set shared energy storage selection indicators from two aspects: shared energy storage and users, to evaluate the advantages and disadvantages of different shared energy storage types; Step 2.1: Based on the operational control results obtained in the first step, establish a corresponding indicator system for shared energy storage, including the annual investment and operating costs of each shared energy storage, the cost of purchasing electricity and charging, the income from leasing shared energy storage to users, and the residual value income. Some indicators are as follows: Where, D is the k-year investment and operation cost of shared energy storage, ay The number of days per year that shared energy storage operates, is the electricity purchase cost for shared energy storage in k years, T is the total number of time periods in a day, is the abandoned wind or solar power output purchased by the shared energy storage k from the user during period t, μ is the unit price of the abandoned wind or solar power purchased by the shared energy storage from the user, is the power purchased by shared energy storage k from the grid during period t, σ t is the price of electricity purchased from the grid during period t by the shared energy storage, is the income from leasing shared energy storage k to users, M ess is the unit price for leasing shared energy storage, is the residual income of shared energy storage k, is the initial investment cost of shared energy storage k, R v,k is the residual value rate of shared energy storage k; Step 2.2: Based on the operational control results obtained in the first step, establish a corresponding indicator system for users, including the user's income from reducing electricity purchase costs from the grid, the income from selling abandoned wind or solar energy, and the cost of leasing shared energy storage. Some indicators are as follows: Where, Reduce the cost of purchasing electricity from the power grid for user i, The power purchased by user i from the upper grid during period t when he does not rent the shared energy storage. The income from selling abandoned wind or solar energy for user i, The cost of leasing shared energy storage for user i; Step 3: Shared energy storage and user multi-criteria compromise solution ranking method, evaluate the shared energy storage selection indicators from two aspects, shared energy storage and user, using the multi-criteria compromise solution ranking method to obtain the evaluation scores; Step 3.1: Build evaluation models for each shared energy storage and user, establish a set of shared energy storage type alternatives, a set of evaluation attributes, and a weight vector of the attributes; Take a shared energy storage or user as an example, let P = {1, 2, ..., p}, Q = {1, 2, ..., q}; A={A1,A2,…,A p } represents a set of p shared energy storage type schemes, each of which includes a battery type, and there is no intersection between the schemes, where A s represents the s-th shared energy storage type scheme, s∈P; C={C1,C2,…,C q } represents a set of q attributes, i.e., the corresponding indicators, and the indicators obtained in step 2 are sorted, where C j represents the jth attribute, i.e. the jth indicator, j∈Q; ω={ω1,ω2,…,ω q } represents the weight vector of the attribute, where ω j For attribute C j The weights satisfy ω j ≥0 and ω j The sum of is equal to 1; x s,j For Plan A s For attribute C j As a result, the decision evaluation matrix is shown in the following table. Each shared energy storage and user builds their own evaluation model according to this step; Decision Evaluation Matrix Step 3.2: Standardize the shared energy storage and user indicators obtained in step 2 to eliminate the impact of dimension on the final results; Take a shared energy storage or user as an example, let x s,j As its decision indicator, r s,j The result after the index is standardized is the following formula to standardize the index: If the shared energy storage and the user's decision indicator x s,j For benefit indicators: If the shared energy storage and the user's decision indicator x s,j For cost indicators: in Each shared energy storage and user standardizes their respective indicators according to this step; Step 3.3: Determine the group utility and individual regret of each shared energy storage selection scheme for each shared energy storage and user entity, and calculate the value of the decision-making index for each shared energy storage selection scheme. The smaller the index value, the better the scheme. The optimal scheme for different entities may be different. Take a shared energy storage or user as an example, let and The maximum and minimum values of each column in its standardized decision evaluation matrix are: Calculate the group utility value S of its shared energy storage selection scheme s and individual regret value R s : According to the results of the group utility value and individual regret value of the shared energy storage selection scheme, the value of the decision index of the shared energy storage selection scheme is calculated. The smaller the index value, the better the scheme: in, In the formula, v represents the coefficient of the shared energy storage selection decision mechanism. If v>0.5, it means that the decision is made according to the decision mechanism of maximizing the group effect of shared energy storage selection. If v<0.5, it means that the decision is made according to the decision mechanism of minimizing the individual regret value of shared energy storage selection. If v=0.5, it means that the decision is made according to the decision mechanism of maximizing the group effect and minimizing the individual regret value of shared energy storage selection, which are equally important. Finally, the index value of each shared energy storage selection scheme can be obtained, and the smaller the index value, the better. Each shared energy storage and user follows this step to obtain their own index value for each shared energy storage selection scheme; Step 4: Combine the shared energy storage obtained in step 3 with the user's index values for each shared energy storage selection scheme, and select the optimal shared energy storage type after comprehensive consideration of all subjects; Step 4.1: Based on the index values of each shared energy storage and user for each shared energy storage selection scheme obtained in step 3, solve the basic probability distribution of each one respectively; A is a shared energy storage selection scheme. For each shared energy storage and user, the basic probability distribution on A is an m-function that satisfies: m(φ)=0; Among them, taking a shared energy storage or user as an example, let a1, a2, ..., a p Represents the index value for each shared energy storage selection scheme, where a s represents the index value of the s-th shared energy storage type scheme, s∈P, b1,b2,…,b p represents the basic probability distribution for each shared energy storage selection scheme, where b s Denotes the basic probability distribution of the s-th shared energy storage type scheme, s∈P, then: Each shared energy storage and user obtains their own basic probability distribution for each shared energy storage selection scheme according to this step; Step 4.2: According to the synthesis rule of evidence theory, the basic probability distribution of each shared energy storage and user is integrated to obtain the final result, thereby obtaining the selection of shared energy storage; Let the total number of shared energy storage and users be n; According to the results of evidence theory fusion, the shared energy storage type corresponding to the smaller value in the result is selected. This shared energy storage type is the optimal solution.

Images