Resource grid-connected pooling station-oriented resource selection method, system, equipment and medium

Abstract

The invention relates to the technical field of resource aggregation, and discloses a resource selection method, system, equipment and medium for a resource grid-connected pooling station, and the method comprises the steps: carrying out the quantification of the operation characteristics of each candidate power generation resource according to an operation characteristic index, obtaining the initial aggregation adaptability, and screening out a resource candidate set according to the initial aggregation adaptability; performing resource division on the resource candidate set based on a resource selection mechanism of cyclic fusion splitting to obtain a scheme set; respectively constructing a game cross decision model between the first resource selection scheme and the second resource selection scheme in the scheme set; and solving each game cross decision model to obtain the average game cross efficiency of each group of resource selection schemes, and obtaining an optimal resource selection scheme according to the average game cross efficiency. According to the method, safe grid connection of the new energy and the stored energy is ensured, and the aggregation potential of different resources is quantitatively evaluated, so that the cooperative complementation among the collected resources is realized, and the absorption capability of the power grid to the new energy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of resource aggregation, in particular to a resource selection method, system, device and medium for a resource grid-connected collection station. BACKGROUND

[0002] With the increasing proportion of clean energy such as wind power and photovoltaic power connected to the power system, on the one hand, the cost of power generation equipment continues to decline rapidly, and the proportion of access cost increases year by year, on the other hand, the proportion of clean energy access continues to increase, which brings serious challenges to the safe operation of the power system: on the power supply side, the multi-type clean energy generation main body mainly in the form of dispersed access presents the characteristics of many points, wide surfaces, variable output and poor regulation capacity; on the grid side, multi-type clean energy is connected in layers at different voltage levels, resulting in more complex uncertainty representation, higher influence degree and deeper interaction transmission.

[0003] The commonly used resource selection scheme at present is to select the collection resources by using data envelopment analysis method, but the existing scheme often separates the candidate screening and data envelopment analysis evaluation, and the weight of cross evaluation is often a free weight or a single upper and lower bound, which leads to that the aggregation scheme cannot effectively utilize the operation characteristics of the resources and form complementary benefits. SUMMARY

[0004] In order to solve the above technical problems, the present application provides a resource selection method, system, device and medium for a resource grid-connected collection station, which can realize accurate and efficient selection of aggregated resources of the collection station by accurately quantifying and evaluating the aggregation capacity of different resources, achieve the synergistic complementarity between the aggregated resources, and improve the technical effect of the power grid on the consumption capacity of new energy.

[0005] In a first aspect, the present application provides a resource selection method for a resource grid-connected collection station, the method comprising: quantifying the operation characteristics of each candidate power generation resource of the resource grid-connected collection station according to the operation characteristic index to obtain an initial aggregation applicability of each candidate power generation resource, and selecting a resource candidate set according to the initial aggregation applicability, wherein the candidate power generation resource includes new energy resources and energy storage resources; based on a resource selection mechanism of cyclic fusion and splitting, dividing the resource candidate set to obtain a scheme set containing a plurality of groups of resource selection schemes, and each group of resource selection schemes contains a plurality of candidate power generation resources; respectively, a game cross decision model between the first resource selection scheme and the second resource selection scheme in the scheme set is constructed, the game cross decision model takes maximization of game cross efficiency between the first resource selection scheme and the second resource selection scheme as a decision objective, and takes a secondary weight constraint as a constraint condition, the secondary weight constraint is generated based on the primary aggregated applicability of the candidate power generation resource in the second resource selection scheme; each game cross decision model is solved to obtain average game cross efficiency of each group of resource selection schemes, and an optimal resource selection scheme is obtained according to the average game cross efficiency.

[0006] Further, the step of quantifying the operation characteristics of each candidate power generation resource of the resource grid-connection collection station according to the operation characteristic index includes: According to the operation data of each candidate power generation resource, the operation characteristic index of each candidate power generation resource is calculated, the operation data includes capacity, power, response duration, output and adjustment upper and lower limit, and the operation characteristic index includes adjustment economy index, capacity contribution index, load tracking capability index and power supply capability index; The operation characteristic index of each candidate power generation resource is normalized and weighted summed to obtain the initial aggregated applicability of each candidate power generation resource.

[0007] Further, the resource selection mechanism based on cyclic fusion and splitting includes the step of performing resource division on the resource candidate set to obtain a scheme set containing several groups of resource selection schemes. Each candidate power generation resource in the resource candidate set is taken as a member, and the resource candidate set is initialized as several alliances, each alliance contains one or more members; According to the grid-connection controllability and overall benefit of the alliance, individual utility of the members in the alliance and overall aggregated utility of the alliance are obtained, and a potential function is defined based on the overall aggregated utility; The overall aggregated utility after alliance fusion is taken as a fusion criterion, the overall aggregated utility after alliance splitting is taken as a splitting criterion, and the iteration operation of fusion and splitting is performed on all alliances to obtain a scheme set, the scheme set contains several groups of resource selection schemes.

[0008] Further, the step of obtaining individual utility of each member in the alliance and overall aggregated utility of the alliance according to the grid-connection controllability and overall benefit of the alliance, and defining a potential function based on the overall aggregated utility includes: According to the load tracking capability of each member in the alliance, the disturbance of output fluctuation on the system and the access cost, the individual utility of each member is obtained; According to individual utility of each member in the alliance and alliance aggregate cost, the overall aggregate utility of the alliance is obtained, and the sum of the overall aggregate utility of all alliances is taken as a potential function.

[0009] Further, the constructing step of the game cross decision model comprises: Taking the first resource selection scheme as an evaluation object and the second resource selection scheme as an evaluated object, an object group is formed; According to a preset input-output index of the evaluated object, an index weight of the evaluation object to the evaluated object under evaluation, a game cross efficiency between the evaluation object and the evaluated object is obtained, the input-output index comprises an input index and an output index, and the index weight comprises an input index weight and an output index weight; According to the preliminary aggregate applicability of each candidate power generation resource in the evaluated object, a secondary weight constraint of the index weight of the evaluation object to the evaluated object under evaluation is generated, and the secondary weight constraint comprises a physical reasonable constraint and a fair benchmark constraint; A game cross decision model based on the object group is constructed with the maximum game cross efficiency between the evaluation object and the evaluated object as a decision target and the secondary weight constraint as a constraint condition.

[0010] Further, the step of generating the secondary weight constraint of the index weight of the evaluation object to the evaluated object under evaluation according to the preliminary aggregate applicability of each candidate power generation resource in the evaluated object comprises: According to the correlation between the input index of the evaluated object and the preliminary aggregate applicability of the candidate power generation resource in the evaluated object, a prior constraint factor is generated, and according to the output index of the evaluated object and the preliminary aggregate applicability of the candidate power generation resource in the evaluated object, a prior advantage factor is generated; According to the prior advantage factor, an upper and lower bound interval of the output index weight of the evaluation object to the evaluated object under evaluation is generated, and according to the prior constraint factor, an upper and lower bound interval of the input index weight of the evaluation object to the evaluated object under evaluation is generated; The upper and lower bound interval of the output index weight and the upper and lower bound interval of the input index weight are taken as the physical reasonable constraint; The sparse regular constraint of the index weight and the anchor constraint based on the prior factor are taken as the fair benchmark constraint, and the prior factor comprises the prior constraint factor and the prior advantage factor; The physical reasonable constraint and the fair benchmark constraint are taken as the secondary weight constraint of the index weight of the evaluation object to the evaluated object under evaluation.

[0011] Further, the step of solving each of the game cross decision models to obtain an average game cross efficiency of each group of resource selection schemes, and obtaining an optimal resource selection scheme according to the average game cross efficiency comprises: solving each of the game cross decision models corresponding to each object group to obtain an optimal game cross efficiency of each object group; dividing each object group into a plurality of object subsets according to the same evaluation object, and calculating an average value of the optimal game cross efficiency of each object subset to obtain an average game cross efficiency of a resource selection scheme corresponding to each evaluation object; taking the resource selection scheme with the highest average game cross efficiency as the optimal resource selection scheme.

[0012] In a second aspect, the present application provides a resource selection system for a resource grid collection station, which comprises: a candidate set generation module, configured to quantify the operation characteristics of each candidate power generation resource of the resource grid collection station according to the operation characteristic indicators to obtain an initial aggregation applicability of each candidate power generation resource, and to filter out a resource candidate set according to the initial aggregation applicability, wherein the candidate power generation resource comprises a new energy resource and an energy storage resource; a scheme generation module, configured to perform resource division on the resource candidate set based on a cyclic fusion and splitting resource selection mechanism to obtain a scheme set containing a plurality of groups of resource selection schemes, and each group of resource selection schemes contains a plurality of candidate power generation resources; a cross decision module, configured to construct a game cross decision model between a first resource selection scheme and a second resource selection scheme in the scheme set, wherein the game cross decision model maximizes the game cross efficiency between the first resource selection scheme and the second resource selection scheme as a decision target, and a secondary weight constraint is generated as a constraint condition based on the primary aggregation applicability of the candidate power generation resource in the second resource selection scheme; solving each of the game cross decision models to obtain an average game cross efficiency of each group of resource selection schemes, and obtaining an optimal resource selection scheme according to the average game cross efficiency.

[0013] In a third aspect, an embodiment of the present application further provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0014] In a fourth aspect, an embodiment of the present application further provides a computer readable storage medium, which stores a computer program, and the computer program is executable on the processor to implement the steps of the above method.

[0015] The application provides a resource selection method, system, device and medium for a resource grid-connected collection station. The application can integrate new energy resources and energy storage resources, quantitatively evaluate the aggregation potential of different power generation resources, quickly decide the selection problem of collection station members, realize the collaborative complementation between resources, and improve the accommodation capacity of the power grid to new energy through the primary aggregation applicability. Through the secondary weight constraint, the extreme weight problem of the traditional game data envelopment analysis model can be inhibited, the stability of the model can be improved, the decision result is more in line with the actual preference of the collection station, the accuracy of the decision result is improved, and thus efficient and low-cost energy collection is realized. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of the resource selection method for the resource grid-connected collection station in the embodiment of the application; Figure 2 is a resource selection aggregation operation curve diagram of the resource selection method for the resource grid-connected collection station in the spring scenario of the resource selection test in the embodiment of the application; Figure 3 is a structural diagram of the resource selection system for the resource grid-connected collection station in the embodiment of the application; Figure 4 is an internal structure diagram of the computer device in the embodiment of the application.

[0017] Reference signs: 10, candidate set generation module; 20, scheme generation module; 30, cross decision module. DETAILED DESCRIPTION

[0018] To make the purpose, technical scheme and advantages of the embodiments of the application clearer, the technical scheme in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0019] Please refer to Figure 1 The resource selection method for the resource grid-connected collection station provided in the first embodiment of the application includes steps S10-S40. Step S10: The operation characteristics of each candidate power generation resource of the resource grid-connected collection station are quantified according to the operation characteristic index, the initial aggregation applicability of each candidate power generation resource is obtained, and the resource candidate set is selected according to the initial aggregation applicability. The candidate power generation resources include new energy resources and energy storage resources. Step S20, based on the resource selection mechanism of the cyclic fusion and split, resource partitioning is performed on the resource candidate set to obtain a scheme set containing several groups of resource selection schemes, each group of resource selection schemes containing several candidate power generation resources; Step S30, a game cross decision model between the first resource selection scheme and the second resource selection scheme in the scheme set is respectively constructed, the game cross decision model taking the maximization of the game cross efficiency between the first resource selection scheme and the second resource selection scheme as a decision target, and taking a secondary weight constraint as a constraint condition, the secondary weight constraint being generated based on the primary aggregation applicability of the candidate power generation resources in the second resource selection scheme; Step S40, each game cross decision model is solved to obtain the average game cross efficiency of each group of resource selection schemes, and the optimal resource selection scheme is obtained according to the average game cross efficiency.

[0020] The application provides a resource selection method for a new energy and energy storage grid-connected collection station.

[0021] In a preferred embodiment, the application evaluates the potential of each power generation resource collection by constructing an evaluation index architecture for the operating characteristics of the power generation resources, and the specific steps include: According to the operating data of each candidate power generation resource, the operating characteristic index of each candidate power generation resource is calculated, the operating data including capacity, power, response duration, output and regulation upper and lower limits, and the operating characteristic index including regulation economy index, capacity contribution index, load tracking capability index and power supply capability index; The operating characteristic index of each candidate power generation resource is normalized and weighted summed to obtain the initial aggregation applicability of each candidate power generation resource.

[0022] In this embodiment, there are obvious operating differences among various power generation resources in terms of capacity, power, response duration, output and regulation upper and lower limits, in order to accurately quantify the operating differences of various resources, four types of operating characteristic indexes are set for a single power generation resource (referred to as a power generation unit), which are regulation economy index, capacity contribution index, load tracking capability index and power supply capability index, and each index is described below.

[0023] The regulation economy index of the power generation unit refers to the ratio between the scheduling cost and the total cost (the total cost of operation scheduling and the total cost of operation) of the power generation unit, and is used to measure the difference in resource regulation cost. The expression is: In the formula, RG represents new energy, ESS represents energy storage resource, The regulation economy index of the mth new energy power generation unit is represented by The regulation amount of the mth new energy power generation unit at the gth regulation is represented by The scheduling cost unit price of the mth new energy power generation unit at the gth regulation is represented by G m The regulation times of the mth new energy power generation unit is represented by The regulation economy index of the nth energy storage power generation unit is represented by The regulation amount of the nth energy storage power generation unit at the hth regulation is represented by The scheduling cost unit price of the nth energy storage power generation unit at the hth regulation is represented by G n The regulation times of the nth energy storage power generation unit is represented by. The regulation times are given by the new energy station and the energy storage station scheduling system respectively, and the total scheduling cost and the total operation cost can be calculated by the new energy station scheduling account and the energy storage system scheduling account.

[0024] The capacity contribution index of the power generation unit is used to represent the capacity contribution ability of the candidate power generation resources of the gathering station during the response to the grid regulation demand. The index can be obtained based on the ratio between the regulation power of the power generation unit responding to the grid regulation demand and the grid regulation demand capacity. The specific expression is: In the formula, The capacity contribution index of the mth new energy power generation unit is represented by g The duration of the mth new energy power generation unit at the gth regulation is represented by The regulation power of the mth new energy power generation unit responding to the grid regulation demand is represented by The capacity contribution index of the nth energy storage power generation unit is represented by The regulation power of the nth energy storage power generation unit responding to the grid regulation demand is represented by h The duration of the nth energy storage power generation unit at the hth regulation is represented by C re The grid regulation demand capacity is represented by.

[0025] The load tracking capability index of the power generation unit is used to represent the load tracking situation of each resource in the aggregation station, and the index takes the generation history information of the resource as the calculation data source, and the expression is: In the formula, The load tracking capability index of the mth new energy power generation unit is represented, and M represents the total number of new energy power generation units in the candidate power generation resource, The power grid load demand power is represented, and the specific is the day-ahead generation plan, The maximum power of the power grid load demand is represented, The load tracking capability index of the nth energy storage power generation unit is represented, and N represents the total number of energy storage power generation units in the candidate power generation resource.

[0026] The power supply capability index of the power generation unit is used to quantify the dispatchable potential of the dispatchable resource, and the main factors affecting the power supply capability are the remaining duration and the dispatchable capacity at the current time. The longer the remaining regulation time is, the greater the dispatchable capacity is, and the stronger the power supply capability is, so the index can be represented as: In the formula, The power supply capability index of the mth new energy power generation unit is represented, and △T m The remaining regulation time of the mth new energy power generation unit is represented, T C,m The response duration of the mth new energy power generation unit is represented, △C RG,m The remaining dispatchable capacity of the mth new energy power generation unit is represented, The maximum dispatchable capacity of the new energy power generation unit is represented, The power supply capability index of the nth energy storage power generation unit is represented, △T n The remaining regulation time of the nth energy storage power generation unit is represented, T C,n The response duration of the nth energy storage power generation unit is represented, △C ESS,n The remaining dispatchable capacity of the nth energy storage power generation unit is represented, The maximum dispatchable capacity of the energy storage power generation unit is represented.

[0027] Based on the above operation characteristic indexes, the evaluation index value of each candidate power generation resource i is calculated, and the index vector is obtained by normalization : In the formula, respectively represent the normalized adjustment economic indicators, capacity contribution indicators, load tracking ability indicators and power supply ability indicators of the i-th candidate power generation resource.

[0028] Then, according to the index weight of each operation characteristic index preset in advance, the operation characteristic indexes are weighted and summed to obtain the preliminary aggregation applicability of each candidate power generation resource. Preferably, the index weight can be determined by using entropy weight or analytic hierarchy process, and the index can be normalized by using quantile normalization method.

[0029] In the embodiment, the operation characteristic index calculates the adjustment economic indicators considering cost and other economic factors; calculates the capacity contribution indicators considering the rated output curves of various resources and the load curve in the region; calculates the load tracking ability indicators considering the actual output characteristics of various resources; and calculates the power supply ability indicators considering the response duration and output upper limit of various resources. Therefore, by quantitatively calculating the four types of standardized operation characteristic indexes of each candidate resource, the capacity, power, response duration, output upper and lower limits, and the instantaneous dynamic change behavior of the remaining adjustable amount can be all taken into account, reflecting the operation characteristics of different resources.

[0030] After obtaining the preliminary aggregation degree SUi of each candidate power generation resource i, the following steps are performed. i Then, according to the threshold value preset in advance, the candidate power generation resources are screened to form a resource candidate set of new energy and energy storage gathering and grid-connected stations. .

[0031] After obtaining the resource candidate set, the embodiment uses the mechanism of competition and cooperation of different power generation resources to construct a set of cyclic fusion-split resource selection mechanism. Through the resource division of the resource candidate set by the mechanism, a plurality of different resource selection schemes can be obtained. The specific steps of the resource selection mechanism include: Each candidate power generation resource in the resource candidate set is taken as a member, and the resource candidate set is initialized as a plurality of alliances, each alliance containing one or more members; According to the grid-connected controllability and overall benefit of the alliance, the individual utility of the members in the alliance and the overall aggregation utility of the alliance are obtained, and a potential function is defined based on the overall aggregation utility; The fusion criterion is the improvement of the overall aggregation utility after alliance fusion, the splitting criterion is the improvement of the overall aggregation utility after alliance splitting, and the potential function is taken as a monotonic improvement target. The iteration operation of fusion and splitting is performed on all alliances to obtain a scheme set, and the scheme set contains a plurality of resource selection schemes.

[0032] ​In this embodiment, each candidate power generation resource in the resource candidate set is taken as a member, and the resource candidate set is initialized into several mutually exclusive alliances according to the mutual exclusion principle, thus obtaining the initialized alliance set. , Indicates initialization of the alliance set The first in There are 10 alliances in this group. There are several alliances, each of which can contain one or more members. For ease of calculation, they can be divided independently according to individual resources during initialization, that is, each initialized alliance contains only one individual resource.

[0033] Then, based on the grid connection controllability of the alliance at the grid-connected aggregation station, the individual utility of each member within the alliance is defined; based on the overall benefit of the alliance, the overall aggregate utility of the alliance is defined; and based on the balance between grid connection synergy benefits and expansion costs, a potential function is defined. Specific steps include: The individual utility of each member is obtained based on their load tracking capabilities, the impact of power fluctuations on the system, and the access costs. Based on the individual utility of each member in the alliance and the alliance aggregation cost, the overall aggregation utility of the alliance is obtained, and the sum of the overall aggregation utilities of all alliances is used as the potential function.

[0034] In this embodiment, individual utility is used to measure the grid-connected controllability of alliance members in the grid-connected collection station. Grid-connected controllability is characterized by the member's ability to track the grid planning curve, the disturbance impact of output fluctuations on the system, and the economic cost of control. Specifically, the ability to track the grid planning curve is quantified by a load tracking function, which is constructed based on the average absolute percentage error between the load tracking deviation and the target output curve. Its formula is expressed as: In the formula, This represents the load tracking capability of the j-th member; MAPE j The mean absolute percentage error of the j-th member is represented by the mean absolute percentage error, which characterizes the ability of alliance members to track the power grid planning curve. This is a minimum value to prevent the denominator from being zero; T represents the total scheduling time. This represents the planned output of the j-th member at time t. Let represent the negative tracking deviation sequence of the j-th member at time t, which consists of the difference between the target output and the planned output of the resource.

[0035] The disturbance impact of output fluctuations on the system is quantified and calculated using a resource fluctuation function, which is constructed based on the absolute median deviation. The formula is as follows: wherein, denotes the volatility of the jth member, MAD j denotes the absolute median deviation of the jth member, the smaller the index, the better the stationarity; T denotes the total scheduling time, denotes the maximum output power of the jth member equivalent to the outside.

[0036] The economic cost of regulation is quantitatively calculated by the access cost function of the resource individual, and the access cost function includes the installation cost of the communication, measurement and control equipment involved in the resource aggregation process, and its formula is expressed as: wherein, B j denotes the resource access cost of the jth member, and the cost is negative to reflect that the lower the cost, the better, denotes the communication cost of the jth member, denotes the measurement cost of the jth member, denotes the installation cost of the control equipment of the jth member, , , are respectively the communication cost weight, the measurement cost weight and the installation cost weight, and the sum of the three weights is equal to 1.

[0037] In this embodiment, it is assumed that the communication, measurement and control equipment costs of the same type of resources are the same, which can be obtained by the new energy station and the storage system supply and dispatch center account, and the cost items are processed by monetization and unitization according to the equation.

[0038] After obtaining the above three quantitative indexes of grid-connected controllability, the additivity of the indexes is realized through robust dimensionless and same scale weighting, so the individual utility of the alliance member can be obtained: wherein, u j denotes the individual utility of the jth member, , , are respectively the load tracking weight, the system disturbance weight and the access cost weight, and the sum of the three weights is equal to 1.

[0039] The individual utility obtained by the weighted sum of the three quantitative indexes describes the three elements of grid-connected controllability, i.e. can follow, less disturbance and low cost, which is consistent with the operation goal of the collection station.

[0040] Based on the individual utility of the alliance member, for any alliance, the overall aggregation utility of the alliance is defined as the additive income of the individual utility in the alliance minus the aggregation cost, and its formula is expressed as: where S k denotes the kth coalition, U(S k ) denotes the overall aggregated utility of the kth coalition, denotes the preset penalty coefficient of aggregated cost, C agg (S k ) denotes the coalition aggregated cost of the kth coalition, meaning the coordination and access cost needed when merging a group of resources into a “coalition / cluster”; max and C min respectively denote the maximum and minimum values of the coalition aggregated cost in all coalitions; c0denotes the fixed coordination overhead, and c1denotes the access and operation unit cost, denotes the number of members in the kth coalition, denotes the access and operation overhead in terms of the number of members; len(S k ) denotes the average equivalent link length of the members in the kth coalition to the main sink node, which can be converted in combination with the geographical distance, communication link hop count, bandwidth / delay constraint, and c2denotes the unit cost of the connection link, denotes the overhead cost calculated in terms of the communication / control network scale, reflecting the cost brought by geographical dispersion, network level and delay constraint, wherein c0, c1and c2can be calculated through existing operation data.

[0041] Based on the overall aggregated utility of the coalition, the potential function is defined as the sum of the overall aggregated utilities of all coalitions, and the formula is represented as: wherein denotes the potential function of the coalition set H.

[0042] Based on the above formula, it can be known that the potential function can truly reflect the trade-off between grid-connected cooperation benefits and expansion costs, and the monotonicity and boundedness ensure the convergence of the potential function, thereby supporting the limited-step convergence of “feasible fusion / necessary split”.

[0043] The resource selection mechanism based on the cyclic fusion and split of the embodiment takes the overall aggregated utility improvement after coalition fusion as the fusion criterion, takes the overall aggregated utility improvement after coalition split as the split criterion, and takes the potential function as the monotonic improvement target, and alternately executes “feasible fusion” and “necessary split” in a limited-step cycle. Specifically, for any two disjoint coalitions , the difference between the overall aggregated utility after fusion of the two coalitions and the overall aggregated utility without fusion is calculated: wherein △U denotes the difference between the overall aggregated utility after fusion of the two coalitions and the overall aggregated utility without fusion, denotes a fusion threshold, the threshold being a value greater than or equal to zero, U(*) denotes the overall aggregated utility of the alliance.

[0044] If the above formula is true, it indicates that the overall aggregated utility of the alliance is improved after fusion, and then the alliance fusion is performed.

[0045] If any alliance There exists a subset such that , denotes a split threshold, the threshold being a value greater than or equal to zero, and it indicates that there exists a subset such that the overall aggregated utility is improved after member splitting, and then the alliance is split into and two alliances, composed of other members of except .

[0046] Since the penalty coefficient of the aggregated cost in the overall aggregated utility , the fixed coordination overhead c0, the access and operation unit cost c1 and the connection link unit cost c2 are bounded, the potential function is strictly increasing and has a limited upper bound, the difference of the potential function at each step is ΔΦ≥min(η,λ)>0, only fusion or splitting is performed at each round, avoiding immediate withdrawal leading to shock and avoiding dead loop, so as to converge in a limited number of steps.

[0047] Until the potential function cannot continue to improve, at which time the loop is terminated, and the system division state, that is, the alliance set after resource selection , denotes the kth alliance after resource selection, and the alliance set after resource selection has K alliances in total.

[0048] For the alliance set after resource selection, in combination with the primary aggregated applicability of each candidate power generation resource, a game cross decision model considering the secondary weight constraint of the primary aggregated applicability is constructed to make a decision on the resource selection scheme, and the specific steps of model construction include: The first resource selection scheme is taken as an evaluation object, and the second resource selection scheme is taken as an evaluated object to form an object group; According to a preset input and output index of the evaluated object and an index weight of the evaluation object to the evaluated object under evaluation, a game cross efficiency between the evaluation object and the evaluated object is obtained, the input and output index includes an input index and an output index, and the index weight includes an input index weight and an output index weight; According to the primary aggregation applicability of each candidate power generation resource in the evaluated object, a secondary weight constraint of an index weight of the evaluated object under evaluation of the evaluated object is generated, and the secondary weight constraint includes a physical reasonable constraint and a fair benchmark constraint; A game cross decision-making model based on object groups is constructed with maximization of game cross efficiency between the evaluated object and the evaluated object as a decision-making target and the secondary weight constraint as a constraint condition.

[0049] In the embodiment, a scheme set (i.e., a coalition set) containing several groups of resource selection schemes generated by the resource selection mechanism is taken as basic data, any resource selection scheme (i.e., any coalition) in the set is regarded as a decision-making unit (DMU), and input indexes and output indexes of each group of DMUs are constructed. Specifically, the input indexes of the kth resource selection scheme include an equivalent reserve occupancy rate of the candidate power generation resource in the kth resource selection scheme to the outside an expected scheduling cost and an equivalent fluctuation degree The three dimensions (m=3) are composed of the values, which can be evaluated by the scheduling center through the running output power curve and the cost of the kth coalition and are normalized to [0, 1]. The output indexes of the kth resource selection scheme include an equivalent load tracking degree of the candidate power generation resource in the kth resource selection scheme to the outside and an equivalent effective availability The two dimensions (r=2) are composed of the values, which can be evaluated by the scheduling center through the running output power curve of the kth coalition and are normalized to [0, 1].

[0050] Any group of resource selection schemes in the set is taken as a first resource selection scheme, any group of resource selection schemes is taken as a second resource selection scheme, the first resource selection scheme is taken as an evaluated object, and the second resource selection scheme is taken as an evaluated object to form an object group, that is, the evaluated object and the evaluated object can be the same scheme.

[0051] Taking an object group composed of a pth resource selection scheme (referred to as scheme p) and a kth resource selection scheme (referred to as scheme k) as an example, assuming that scheme p is an evaluation object and scheme k is an evaluated object, the game cross efficiency between scheme p and scheme k can be represented by the index weight of scheme p under the evaluation of scheme k and the input index and input index of the evaluation of scheme k. At this time, the game data envelopment analysis model can be used to establish the game cross decision model between scheme p and scheme k. The game cross decision model takes the index weight of scheme p under the evaluation of scheme k as a variable and maximizes the game cross efficiency between scheme p and scheme k as a decision target. Therefore, the model is actually a maximization linear programming problem: In the formula, u and v respectively represent the output index weight vector and the input index weight vector of scheme p under the evaluation of scheme k, u r and v m respectively represent the output index weight of the rth dimension and the input index weight of the mth dimension of scheme p under the evaluation of scheme k, y j,r represents the output index of the rth dimension of the kth resource selection scheme, and x j,m represents the input index of the mth dimension of the kth resource selection scheme. The constraint condition limits E p→k to be less than 1.

[0052] The output index weight vector and the input index weight vector that maximize the game cross efficiency are taken as the optimal index weight of scheme p under the evaluation of scheme k . Therefore, the optimal game cross efficiency of scheme p under the evaluation of scheme k is: In the formula, u and v respectively represent the optimal output index weight and the optimal input index weight of scheme p under the evaluation of scheme k.

[0053] The process of the game cross decision model for the game cross evaluation between DMUs is the process of mutual comparison and evaluation of different schemes under their respective input and output indexes. Preferably, the game data envelopment analysis model of the embodiment adopts an input-oriented CCR (Charnes-Cooper-Rhodes, CCR) model to calculate the game cross efficiency.

[0054] In the evaluation process, if no restrictions are made, each scheme can extremely amplify the output weight that is beneficial to itself, thereby affecting the accuracy of the evaluation result. In order to limit the emergence of extreme weights, the embodiment generates secondary weight constraints including physical reasonable constraints and fair benchmark constraints by means of the primary aggregated applicability, so as to realize the constraint on the index weight, and the specific steps include: According to the correlation between the input index of the evaluated object and the primary aggregated applicability of the candidate power generation resource in the evaluated object, a prior constraint factor is generated, and according to the output index of the evaluated object and the primary aggregated applicability of the candidate power generation resource in the evaluated object, a prior advantage factor is generated; According to the prior advantage factor, the upper and lower bound intervals of the output index weight under the evaluation of the evaluated object by the evaluation object are generated, and according to the prior constraint factor, the upper and lower bound intervals of the input index weight under the evaluation of the evaluated object by the evaluation object are generated; The upper and lower bound intervals of the output index weight and the upper and lower bound intervals of the input index weight are taken as the physical reasonable constraints; The sparse regular constraint of the index weight and the anchor point constraint based on the prior factor are taken as the fair benchmark constraints, and the prior factor includes the prior constraint factor and the prior advantage factor; The physical reasonable constraints and the fair benchmark constraints are taken as the secondary weight constraints of the index weight under the evaluation of the evaluated object by the evaluation object.

[0055] In the embodiment, the evaluated object and the evaluation object are both resource selection schemes. For each set of resource selection schemes, two prior factors based on the scheme level are calculated based on the input index, the output index and the primary aggregated applicability of the candidate power generation resource contained in the scheme, which are the prior advantage factor and the prior constraint factor respectively. The prior advantage factor represents the inherent advantage of the scheme in the output dimension (such as tracking / availability), and the larger the factor, the more advantageous the scheme is in the output dimension. The prior constraint factor represents the limitation of the scheme in the input dimension (such as cost / variation / reserve), and the larger the factor, the more limited the scheme is in the input dimension.

[0056] Still taking the evaluation of scheme p on scheme k as an example, the prior advantage factor is calculated based on the correlation coefficient between the primary aggregated applicability of each candidate power generation resource in scheme k and the output index of scheme k, and its expression is: In the formula, indicates the prior advantage factor of the output index of the kth resource selection scheme in the rth dimension, indicates the correlation coefficient between the output index in the rth dimension and the jth candidate power generation resource in the kth resource selection scheme, ;SUj represents the primary aggregated applicability of the jth candidate generation resource in the kth resource selection scheme.

[0057] The prior constraint factor is calculated based on the correlation coefficient between the primary aggregated applicability of each candidate generation resource in the kth scheme and the input indicator of the kth scheme, and its expression is: In the formula, represents the prior constraint factor of the input indicator in the mth dimension of the kth resource selection scheme, represents the correlation coefficient between the input indicator in the mth dimension and the jth candidate generation resource in the kth resource selection scheme, . and determined by the experience of power dispatch experts.

[0058] The logical mapping rule of the prior factor to the physically reasonable interval of the weight is: through the prior advantage factor, the upper and lower limit values of the output indicator weight under the evaluation of scheme p to scheme k are corrected; through the prior constraint factor, the upper and lower limit values of the input indicator weight under the evaluation of scheme p to scheme k are corrected, so as to encourage each scheme to tend to aggregation and improve the aggregation utility, and the specific expression is: In the formula, u r and v m respectively represent the output indicator weight in the rth dimension and the input indicator weight in the mth dimension under the evaluation of scheme p to scheme k, and respectively represent the lower limit and the upper limit of the output indicator weight in the rth dimension, and respectively represent the lower limit and the upper limit of the input indicator weight in the mth dimension, all represent the slope coefficient.

[0059] Preferably, and take values of 0.05 and 0.95 respectively, and take values of 0.05 and 0.95 respectively, the slope coefficient is set to take a value range of [0, 0.3] according to experience, and the slope coefficient is used for contraction and expansion of the physically reasonable interval, so as to reflect the importance of the primary aggregated applicability in the evaluation stage.

[0060] Further, in order to prevent the emergence of extreme weights during the game cross-efficiency mutual evaluation process, the embodiment further establishes a fair benchmark constraint interval for the index weight through sparse regularization constraint of the index weight and anchor point constraint based on the prior factor, wherein for the sparse regularization constraint, by setting a control sparsity, the sum of the first norm of the input index weight and the first norm of the output index weight is limited within the control sparsity range; for the anchor point constraint, the limit value of the index weight (including the input index weight and the output index weight) is set according to the anchor point, and the anchor point is calculated based on the benchmark value of the index weight and the prior factor, and for the output index weight, the anchor point tolerance is calculated according to the benchmark value of the output index weight, the prior advantage factor and the anchor point tolerance, and for the input index weight, the anchor point tolerance is calculated according to the benchmark value of the input index weight, the prior constraint factor and the anchor point tolerance, and the specific expression is: In the formula, u and v respectively represent the output index weight vector and the input index weight vector under the evaluation of scheme p to scheme k; respectively represent the anchor point value of the output index weight in the rth dimension and the anchor point value of the input index weight in the mth dimension under the evaluation of scheme p to scheme k, represents the anchor point tolerance; respectively represent the benchmark coefficient of the output index in the rth dimension and the benchmark coefficient of the input index in the mth dimension, the benchmark coefficient is a preset value, representing the benchmark preference of the industry in the evaluation process (such as tracking ≥ available), and ; represents the benchmark value of the output index weight in the rth dimension; represents the benchmark value of the input index in the mth dimension, and the benchmark value is a preset value; represents the control sparsity, which is a constant fixed value, and preferably, 1.9; represents the first norm; u r represents the output index weight in the rth dimension under the evaluation of scheme p to scheme k; v m represents the input index weight in the mth dimension under the evaluation of scheme p to scheme k.

[0061] The above physical reasonable interval and fair benchmark constraint interval are used as the secondary weight constraint of the physical reasonable constraint and the fair benchmark constraint, and the above maximum linear programming problem can be represented as: Through the above steps, the game cross decision model corresponding to each object group can be established, and then the solver is used to solve each game cross decision model, and according to the solving result, the average game cross efficiency of each group of resource selection schemes is calculated, that is, the optimal resource selection scheme is sorted and selected, and the specific steps include: solving the game cross decision model corresponding to each object group, to obtain the optimal game cross efficiency of each object group; According to the same evaluation object, each object group is divided into multiple object subsets, and the average value of the optimal game cross efficiency of each object subset is calculated respectively, to obtain the average game cross efficiency of the resource selection scheme corresponding to each evaluation object; The resource selection scheme with the highest average game cross efficiency is taken as the optimal resource selection scheme.

[0062] In this embodiment, any group of resource selection schemes is taken as the evaluation object, and any group of resource selection schemes is taken as the evaluated object, to obtain multiple object groups. For each object group, a game cross decision model is established, and each game cross decision model is solved respectively to obtain the optimal game cross efficiency of each object group, that is, the optimal game cross efficiency between the evaluation object and the evaluated object in each object group.

[0063] According to the same evaluation object, each object group is divided into object subsets based on each evaluation object. For example, the evaluation object of each object group included in the object subset of scheme p is scheme p.

[0064] Then the average value of the optimal game cross efficiency is calculated for each evaluation subset, so as to obtain the average game cross efficiency of the resource selection scheme corresponding to the evaluation object, such as the average game cross efficiency of scheme p : In the formula, K represents the total number of resource selection schemes, that is, the total number of alliance members in the alliance set.

[0065] The average game cross efficiency of each resource selection scheme is sorted, and the resource selection scheme with the highest average game cross efficiency is selected as the optimal resource selection scheme, and the resource collection and regulation of the grid collection station is performed according to the optimal resource selection scheme.

[0066] This embodiment introduces a two-level weight constraint in the existing game cross efficiency model, which not only encourages the scheme to tend to aggregation and improves the physical rationality of the index weight interval, but also avoids the appearance of extreme index weight, affecting the fairness of the evaluation result, so that the cross evaluation result between decision units is more accurate. Through the game cross mechanism, the self-interest preference is converted into a consistent public measurement for all schemes, and the stable sorting is obtained after averaging. Therefore, the highest average game cross efficiency means that the efficiency average is recognized by all schemes with their respective acceptable evaluation criteria, indicating that the resource selection scheme is the optimal and most recognized scheme.

[0067] In order to verify the resource aggregation effect of the resource selection method provided in the embodiment, a resource selection test is performed by taking the actual resource data of a certain place aggregation station as an example. The test performs resource selection in the scenarios of one year and typical days in four seasons. First, a program is written in the Matlab environment to realize the resource selection method provided in the embodiment, and a Gurobi mathematical programming optimizer is used for solving.

[0068] The overall aggregation utility of the resource selection schemes in different scenarios obtained by solving is shown in Table 1 as follows: Table 1 Aggregation utility table of resource selection in different scenarios According to Table 1, the resource selection mechanism provided in the embodiment can quickly perform resource selection, realize the coordination and complementation between resources, and improve the consumption capacity of the power grid for new energy.

[0069] Taking the spring scenario as an example, the normalized input indicators and output indicators of the three resource selection schemes in the spring scenario are shown in Table 2 as follows: Table 2 Index summary table of resource selection schemes in spring scenario The interval calculation of the second weight constraint is performed for the three resource selection schemes in the spring scenario, and the optimal game cross efficiency is calculated based on the constraint condition. The results are as follows: taking scheme 1 as the evaluation object, and schemes 1, 2 and 3 as the evaluated objects, the optimal game cross efficiencies obtained are 1.000, 0.918 and 0.935 respectively; taking scheme 2 as the evaluation object, and schemes 1, 2 and 3 as the evaluated objects, the game cross efficiencies obtained are 0.942, 1.000 and 0.951 respectively; taking scheme 3 as the evaluation object, and schemes 1, 2 and 3 as the evaluated objects, the game cross efficiencies obtained are 0.965, 0.972 and 1.000 respectively. Then the average cross efficiencies of schemes 1, 2 and 3 are 0.951, 0.964 and 0.979 respectively, and the scheme ranking is scheme 3> scheme 2> scheme 1, so scheme 3 is optimal in the spring scenario.

[0070] In order to verify the aggregation effect of each resource selection scheme in the spring scenario, simulation is performed to obtain the aggregation operation curves of the grid-connected aggregation station in the spring scenario under the three resource selection schemes as shown in Figure 2 Figure 2 ​It can be seen that in the three schemes under the spring scene, the operation curve of scheme 3 has the best tracking effect on the load demand curve, can timely respond to the demand of the power grid, and will not cause power deviation. At the same time, the equivalent effective available power of scheme 3 is the highest, which is used to support the power grid. Although the curve fluctuation of scheme 3 is slightly inferior to the other two schemes, the standby power of scheme 3 is the best, and the preparation for the power grid is the best. The above results are consistent with the cross efficiency results based on the secondary constraint condition.

[0071] The resource selection method provided by the application can consider the complementary effect of distributed energy, and based on the sorting determination condition of average game cross efficiency, the finally formed collection station can participate in scheduling with output characteristics more consistent with load scheduling requirements.

[0072] The resource selection method for resource grid-connected collection stations provided by the embodiment can integrate new energy resources and energy storage resources, quantitatively evaluate the aggregation potential of different power generation resources, quickly decide the selection problem of collection station members, realize the synergistic complementation between resources, and improve the consumption capacity of the power grid for new energy; through the secondary weight constraint, the extreme weight problem of the traditional game data envelopment analysis model can be inhibited, the stability of the model is improved, the decision result is more consistent with the actual preference of the collection station, the accuracy of the decision result is improved, and thus efficient and low-cost energy collection is realized.

[0073] Please refer to Figure 3 , based on the same inventive concept, the second embodiment of the application provides a resource selection system for resource grid-connected collection stations, comprising: A candidate set generation module 10 is configured to quantitatively evaluate the operation characteristics of each candidate power generation resource of the resource grid-connected collection station according to the operation characteristic index, obtain an initial aggregation applicability of each candidate power generation resource, and select a resource candidate set according to the initial aggregation applicability, wherein the candidate power generation resource includes new energy resources and energy storage resources. A scheme generation module 20 is configured to perform resource division on the resource candidate set based on a cyclic fusion and splitting resource selection mechanism, and obtain a scheme set containing a plurality of groups of resource selection schemes, each group of resource selection schemes containing a plurality of candidate power generation resources. A cross decision module 30 is configured to construct a game cross decision model between a first resource selection scheme and a second resource selection scheme in the scheme set, wherein the game cross decision model maximizes the game cross efficiency between the first resource selection scheme and the second resource selection scheme as a decision target, and a secondary weight constraint is used as a constraint condition, and the secondary weight constraint is generated based on the initial aggregation applicability of the candidate power generation resource in the second resource selection scheme. Solving each of the game cross decision models, average game cross efficiency of each group of resource selection schemes is obtained, and the optimal resource selection scheme is obtained according to the average game cross efficiency.

[0074] The technical features and technical effects of the resource selection system for the resource grid-connected collection station proposed in the embodiments of the application are the same as those of the method proposed in the embodiments of the application, and are not repeated here. Each module in the resource selection system for the resource grid-connected collection station can be realized by software, hardware, and combinations thereof, in whole or in part. Each module can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory in the computer device in software form, so as to be called and executed by the processor to perform the operations corresponding to each module.

[0075] In addition, the embodiments of the application also propose a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0076] Please refer to Figure 4 , an internal structure diagram of a computer device in an embodiment, which can be a terminal or a server. The computer device includes a processor, a memory, a network interface, a display, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The network interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement the resource selection method for the resource grid-connected collection station. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball, or touchpad arranged on the shell of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0077] Those skilled in the art can understand Figure 4 the structure shown in the figure, only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have the same component arrangement.

[0078] In addition, the embodiments of the application also propose a computer readable storage medium, which stores a computer program, and the computer program is executed by the processor to implement the steps of the above method.

[0079] To sum up, the resource selection method, system, device and medium for the resource grid-connected collection station proposed in the embodiments of the present application, the method quantifies the operation characteristics of each candidate power generation resource of the resource grid-connected collection station according to the operation characteristic index, obtains the initial aggregation applicability of each candidate power generation resource, and screens out a resource candidate set according to the initial aggregation applicability, the candidate power generation resource including new energy resources and energy storage resources; based on the resource selection mechanism of cyclic fusion and splitting, the resource candidate set is divided into resource selection schemes, each group of resource selection schemes including a plurality of candidate power generation resources; a game cross decision model between a first resource selection scheme and a second resource selection scheme in the scheme set is constructed respectively, the game cross decision model taking the maximization of the game cross efficiency between the first resource selection scheme and the second resource selection scheme as the decision target, and taking a two-level weight constraint as the constraint condition, the two-level weight constraint being generated based on the primary aggregation applicability of the candidate power generation resource in the second resource selection scheme; each game cross decision model is solved to obtain the average game cross efficiency of each group of resource selection schemes, and the optimal resource selection scheme is obtained according to the average game cross efficiency. The present application accurately quantifies and evaluates the aggregation ability of different resources, realizes the accurate and efficient selection of the aggregation resources of the collection station, achieves the synergy and complementarity between the aggregation resources, thereby improving the accommodation capacity of the power grid to new energy and realizing efficient and low-cost energy collection.

[0080] Each embodiment in the specification is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment focuses on the difference from other embodiments. In particular, the system embodiment is basically similar to the method embodiment, so the description is relatively simple, and the relevant parts can be referred to the part of the method embodiment. It should be noted that, for the sake of brevity of the description, not all possible combinations of the technical features of the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the specification.

[0081] The above-described embodiments only express several preferred embodiments of the present application, and the description is more specific and detailed, but it should not be understood as limiting the scope of the patent. It should be noted that, for ordinary skilled in the art, without departing from the technical principles of the present application, some improvements and replacements can be made, and these improvements and replacements should be considered as the protection scope of the present application. Therefore, the protection scope of the patent of the present application should be subject to the protection scope of the claims.

Claims

1. A resource selection method for resource grid-connected aggregation stations, characterized in that, include: The operational characteristics of each candidate power generation resource in the resource grid-connected aggregation station are quantified according to the operational characteristic indicators to obtain the initial aggregation applicability of each candidate power generation resource. Based on the initial aggregation applicability, a resource candidate set is selected, which includes new energy resources and energy storage resources. Based on the resource selection mechanism of cyclic fusion and splitting, the resource candidate set is divided into resource partitions to obtain a scheme set containing several sets of resource selection schemes, and each set of resource selection schemes contains several candidate power generation resources. A game-theoretic cross-decision model is constructed between the first resource selection scheme and the second resource selection scheme in the scheme set. The game-theoretic cross-decision model takes maximizing the game-theoretic cross-decision efficiency between the first resource selection scheme and the second resource selection scheme as the decision objective, and uses secondary weight constraints as constraints. The secondary weight constraints are generated based on the primary aggregation applicability of the candidate power generation resources in the second resource selection scheme. Solve each of the aforementioned game-theoretic cross-decision models to obtain the average game-theoretic cross-efficiency of each resource selection scheme, and obtain the optimal resource selection scheme based on the average game-theoretic cross-efficiency.

2. The resource selection method for resource grid-connected aggregation stations according to claim 1, characterized in that, The step of quantifying the operational characteristics of each candidate power generation resource at the resource grid-connected aggregation station based on operational characteristic indicators to obtain the initial aggregation applicability of each candidate power generation resource includes: Based on the operational data of each candidate power generation resource, the operational characteristic indicators of each candidate power generation resource are calculated. The operational data includes capacity, power, response duration, output and upper and lower limits of regulation. The operational characteristic indicators include regulation economy indicators, capacity contribution indicators, load tracking capability indicators and power supply capability indicators. The operational characteristic indicators of each candidate power generation resource are normalized and weighted and summed to obtain the initial aggregate applicability of each candidate power generation resource.

3. The resource selection method for resource grid-connected aggregation stations according to claim 1, characterized in that, The resource selection mechanism based on cyclic fusion and splitting, which divides the resource candidate set into resource partitions to obtain a set of schemes containing several resource selection schemes, includes the following steps: Using each candidate power generation resource in the resource candidate set as a member, the resource candidate set is initialized into several alliances, each alliance containing one or more members; Based on the grid connection controllability and overall benefits of the alliance, the individual utility of the members within the alliance and the overall aggregate utility of the alliance are obtained, and a potential function is defined based on the overall aggregate utility. Using the improvement of the overall aggregation utility after alliance merging as the merging criterion and the improvement of the overall aggregation utility after alliance splitting as the splitting criterion, and using the potential function as the monotonically improving target, iterative operations of merging and splitting are performed on all alliances to obtain a set of solutions, which includes several sets of resource selection solutions.

4. The resource selection method for resource grid-connected aggregation stations according to claim 3, characterized in that, The steps of obtaining the individual utility of each member and the overall aggregate utility of the alliance based on the grid connection controllability and overall benefits of the alliance, and defining a potential function based on the overall aggregate utility, include: The individual utility of each member is obtained based on their load tracking capabilities, the impact of power fluctuations on the system, and the access costs. Based on the individual utility of each member in the alliance and the alliance aggregation cost, the overall aggregation utility of the alliance is obtained, and the sum of the overall aggregation utilities of all alliances is used as the potential function.

5. The resource selection method for resource grid-connected aggregation stations according to claim 1, characterized in that, The steps for constructing the game-theoretic cross-decision model include: The first resource selection scheme is used as the evaluation object, and the second resource selection scheme is used as the evaluated object, forming an object group; Based on the preset input and output indicators of the evaluated object and the indicator weights of the evaluation object on the evaluated object, the game cross efficiency between the evaluation object and the evaluated object is obtained. The input and output indicators include input indicators and output indicators, and the indicator weights include input indicator weights and output indicator weights. Based on the primary aggregation applicability of each candidate power generation resource in the evaluated object, a secondary weight constraint is generated for the indicator weights of the evaluated object under the evaluation of the evaluated object. The secondary weight constraint includes physical rationality constraint and fair benchmark constraint. With the goal of maximizing the cross-game efficiency between the evaluation object and the evaluated object, and with the secondary weight constraints as the constraints, a cross-game decision-making model based on object groups is constructed.

6. The resource selection method for resource grid-connected aggregation stations according to claim 5, characterized in that, The step of generating secondary weight constraints on the indicator weights of the evaluated object based on the primary aggregation applicability of each candidate power generation resource in the evaluated object includes: Based on the correlation between the input indicators of the evaluated object and the primary aggregation applicability of the candidate power generation resources in the evaluated object, a prior constraint factor is generated, and based on the output indicators of the evaluated object and the primary aggregation applicability of the candidate power generation resources in the evaluated object, a prior advantage factor is generated. Based on the prior advantage factor, the upper and lower bounds of the output indicator weights of the evaluation object in evaluating the evaluated object are generated, and based on the prior constraint factor, the upper and lower bounds of the input indicator weights of the evaluation object in evaluating the evaluated object are generated. The upper and lower bounds of the output index weights and the upper and lower bounds of the input index weights are used as physical reasonable constraints. The sparse regularization constraint of the index weight and the anchor constraint based on the prior factor are used as the fairness benchmark constraint. The prior factor includes the prior constraint factor and the prior advantage factor. The physical rationality constraint and the fairness benchmark constraint are used as secondary weight constraints for the indicator weights of the evaluation object on the evaluated object.

7. The resource selection method for resource grid-connected aggregation stations according to claim 5, characterized in that, The steps of solving each of the aforementioned game-theoretic cross-decision models to obtain the average game-theoretic cross-efficiency of each resource selection scheme, and obtaining the optimal resource selection scheme based on the average game-theoretic cross-efficiency, include: Solve the game cross-decision model corresponding to each object group to obtain the optimal game cross-efficiency for each object group; Based on the same evaluation object, each object group is divided into multiple object subsets, and the average value of the optimal game cross efficiency of each object subset is calculated to obtain the average game cross efficiency of the resource selection scheme corresponding to each evaluation object. The resource selection scheme with the highest average game cross-efficiency is taken as the optimal resource selection scheme.

8. A resource selection system for resource grid-connected aggregation stations, characterized in that, include: The candidate set generation module is used to quantify the operating characteristics of each candidate power generation resource of the resource grid-connected aggregation station according to the operating characteristic indicators, obtain the initial aggregation applicability of each candidate power generation resource, and filter out the resource candidate set according to the initial aggregation applicability. The candidate power generation resources include new energy resources and energy storage resources. The scheme generation module is used to divide the resource candidate set into resources based on the resource selection mechanism of cyclic fusion and splitting, and obtain a scheme set containing several sets of resource selection schemes, each set of resource selection schemes containing several candidate power generation resources. The cross-decision module is used to construct game-theoretic cross-decision models between the first resource selection scheme and the second resource selection scheme in the scheme set. The game-theoretic cross-decision model takes maximizing the game-theoretic cross-decision efficiency between the first resource selection scheme and the second resource selection scheme as the decision objective and uses secondary weight constraints as constraints. The secondary weight constraints are generated based on the primary aggregation applicability of the candidate power generation resources in the second resource selection scheme. Solve each of the aforementioned game-theoretic cross-decision models to obtain the average game-theoretic cross-efficiency of each resource selection scheme, and obtain the optimal resource selection scheme based on the average game-theoretic cross-efficiency.

9. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images