Virtual power plant cluster resource screening method and system based on multi-energy collaboration

Abstract

The invention discloses a virtual power plant cluster resource screening method and system based on multi-energy collaboration, and relates to the technical field of intelligent power grids, and the method comprises the steps: dividing resource equipment in a virtual power plant according to the control authority, analyzing the response electric quantity provided by uncontrollable resources, and building a response characteristic model; analyzing an expected value and a random deviation value of user response electric quantity by utilizing a response characteristic model, constructing a virtual power plant resource optimization combination model according to an analysis result, setting a constraint condition according to an analysis result of resource response characteristics, and determining a control attribute sequence of decision variables through an angle and distance dual mechanism; and formulating a user response plan and a resource combination strategy according to the sequence. According to the method, by optimizing resource management, maximizing expected benefits and adopting flexible response strategies and mixed decision variable analysis, the operation efficiency and economic benefits of the virtual power plant are effectively improved, and meanwhile stable operation of a power system is facilitated.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a virtual power plant cluster resource screening method and system based on multi-energy collaboration. Background Art

[0002] Virtual power plant is a technical means to centrally manage and coordinate the dispatch of multiple distributed energy sources, demand response, energy storage equipment, etc., aiming to enable these small-scale distributed resources to achieve stable, adjustable and economical power generation capacity similar to traditional large-scale power plants under the dispatch of virtual power plants. VPP can not only sell electricity to the power market, but also provide power system auxiliary services such as frequency regulation, peak regulation, emergency standby, etc. to the power grid.

[0003] The basic operation of a VPP is to dispatch various resources under its jurisdiction to meet power demand and optimize power system performance. However, due to the large number and variety of distributed resources, as well as the large amount of operating data and high real-time requirements, the resource scheduling of VPPs poses an extremely high challenge. In addition, since VPP resources usually come from dispersed individual users, the power demand and response behavior of these users are highly uncertain, which also brings risks to the stable operation of VPPs.

[0004] To address this problem, existing technical solutions include dividing different resources into controllable resources and uncontrollable resources according to their characteristics; dividing users into demand response users and free response users according to whether the response power they provide is uncertain; considering uncertainty, using risk management models to optimize resource combinations, etc. Under the VPP management model, operators determine the response power of each user by weighing internal resources and external demand, and achieve global optimization by reasonably optimizing the user's power purchase plan. Summary of the invention

[0005] In view of the above-mentioned existing problems, the present invention provides a virtual power plant cluster resource screening method and system based on multi-energy collaboration, so as to solve the problems in the prior art of how to reasonably divide and manage controllable resources and uncontrollable resources, how to build a reasonable resource optimization combination model, how to effectively deal with the risks caused by market rules and user response behaviors, how to utilize hybrid decision variable analysis technology and how to formulate and implement effective user response strategies under the premise of meeting the stable operation of the power system and market rules.

[0006] In order to solve the above technical problems, a virtual power plant cluster resource screening method based on multi-energy collaboration is proposed, including:

[0007] The resource equipment in the virtual power plant is divided according to the control authority, and the response power provided by the uncontrollable resources is analyzed and a response characteristic model is established; the expected value and random deviation of the user's response power are analyzed using the response characteristic model, and a virtual power plant resource optimization combination model is constructed based on the analysis results; constraint conditions are set according to the analysis results of the resource response characteristics, and the control attribute ranking of the decision variables is determined through the dual mechanisms of angle and distance, and user response plans and resource combination strategies are formulated based on the ranking.

[0008] As a preferred solution of the virtual power plant cluster resource screening method based on multi-energy collaboration described in the present invention, the division according to control authority includes dividing the resources into controllable and uncontrollable categories according to the control authority of the virtual power plant over the resource equipment.

[0009] The uncontrollable resources are included in the contract signed between the virtual power plant and the user. The user has not entrusted the control authority of the terminal device to the virtual power plant and is regarded as a demand response user. The response power provided by the user is uncertain. As a risky asset, the response action of the demand response user is an incentive-based demand response. As an aggregator, the virtual power plant needs to pay attention to the actual response effect of the user.

[0010] As a preferred solution of the virtual power plant cluster resource screening method based on multi-energy collaboration described in the present invention, wherein: the establishment of the response characteristic model includes identifying the contract characteristics of demand response users, analyzing the electricity consumption characteristics of demand response users, and requiring demand response users to submit electricity purchase demand forecast curves and response ranges to the virtual power plant before the scheduling cycle, establishing a response characteristic model of demand response users, analyzing the historical response deviation electricity of risky asset users, and analyzing the correlation between each user's return standard deviation and risky assets, and understanding the consistency and complementarity of response errors between different users based on the correlation analysis.

[0011] The response characteristic model is expressed as:

[0012]

[0013] in, is the actual response power change of the user in the tth time period, is the lower limit of the response power achieved by the user in the tth time period, is the upper limit of the response power achieved by the user in the tth time period, is the predicted power consumption curve of the i-th user in the k-th time period, * is the target value, C i,DR,k is the total cost of the user in the kth time period, T is the total number of time periods, i is the user index, and k is the time period index.

[0014] Among them, after considering the uncertainties Rewritten as:

[0015]

[0016] in, is the actual response power change after rewriting, is the expected value of the response power allocated to the user, The random deviation of the user's response power.

[0017] The correlation analysis includes analyzing the historical response deviation of risky asset users to obtain the standard deviation of each user's income σ i and correlations between risky assets It means that the response errors of risky asset user i and risky asset user j are consistent, and over-response and under-response will occur at the same time; It indicates that the response errors between users are complementary.

[0018] As a preferred solution of the virtual power plant cluster resource screening method based on multi-energy collaboration described in the present invention, wherein: the construction of the virtual power plant resource optimization combination model includes defining the goal of the virtual power plant resource optimization combination problem, establishing the objective function of the virtual power plant optimization combination model, determining the expected income, and decomposing the expected cost for analysis.

[0019] The decomposition of expected costs includes decomposing expected costs into incentive costs, uncertain costs and access costs; the optimization combination model formula is:

[0020] max:E(R p )=C in -E(C cost )

[0021] Among them, E(R p ) is the expected revenue of the virtual power plant after combination, C in is the expected revenue of the virtual power plant, E(C cost ) is the expected cost.

[0022] The expected revenue formula of the virtual power plant is:

[0023] C in =(C r -C w ) aim,t Δt

[0024] Among them, C r is the retail electricity price, C w is the wholesale electricity price in the electricity market, P aim,t is the target output power at time point t, and Δt is the time interval.

[0025] The expected cost components are expressed as:

[0026] E(C cost )=E(f1)+E(f2)+f3

[0027]

[0028] Among them, E(C cost ) is the expected cost, f1 is the incentive cost, f2 is the uncertain cost, f3 is the access cost, E(f1) is the expected incentive cost, E(f2) is the expected uncertain cost, i, j, t and k are variable indices, C i,j,k is the unit response incentive cost of interactive resources, c er is the unit deviation penalty price, P aim,t is the target output power at time t, P vpp,t is the actual output power at time point t, Δt is the time interval, λ is the access cost required for a single resource to participate in regulation, V i,j,k It is a 0 / 1 variable.

[0029] As a preferred solution of the virtual power plant cluster resource screening method based on multi-energy collaboration described in the present invention, the setting of constraint conditions includes setting risk and utility constraints, resource response characteristic constraints, equation constraints and application constraints of virtual power plant aggregation external characteristics according to the analysis results of resource response characteristics, and defining asset portfolio weights according to the expected return expression to solve the optimal resource combination plan.

[0030] The constraints are expressed as:

[0031]

[0032] in, is the risk and utility constraints for the optimal combination solution, is the resource response characteristic constraint, Equality constraints for the aggregation of external characteristics of the virtual power plant, To apply constraints.

[0033] As a preferred solution of the virtual power plant cluster resource screening method based on multi-energy collaboration described in the present invention, the application constraints include: the virtual power plant participates in the energy market bidding as a third-party independent entity, submits an aggregated power baseline based on power generation forecast and load forecast as a bidding curve, compares the bids of the participating entities, obtains the clearing power curve, determines the power baseline forecast value and the clearing power curve, selects the user combination participating in the response according to the clearing result, and formulates a user response plan, considers external characteristic requirements, and uses external characteristic requirements as application constraints for virtual power plants to participate in the electric energy market, refers to the implementation rules of the electric power spot market transactions, formulates the VPP internal settlement principles, introduces penalty costs in the objective function of the virtual power plant internal settlement principles to characterize the size of the revenue recovery in the market settlement, and the virtual power plant determines the deviation of the unbalanced electricity in the electricity market according to the market trading rules.

[0034] The deviation of the electricity market is expressed as:

[0035] -ε VPP ≤P VPP,t Δt-P aim,t Δt≤ε VPP

[0036] Among them, P aim,t is the target output power at time t, P vpp,t is the actual output power at time t, Δt is the time interval, ε VPP is the maximum unbalanced charge error.

[0037] As a preferred solution of the virtual power plant cluster resource screening method based on multi-energy collaboration described in the present invention, wherein: the formulation of user response plans and resource combination strategies includes setting constraint conditions according to the analysis results of resource response characteristics, and determining the control attribute ranking of decision variables by using a hybrid decision variable analysis method using a dual mechanism of angle and distance, and formulating user response plans and resource combination strategies according to the ranking;

[0038] The hybrid decision variable analysis method includes clustering the weight vector, selecting two candidate solutions from the neighborhood of the weight center, perturbing the two candidate solutions 5 times to obtain a perturbed solution set, normalizing the perturbed solution set to obtain two lines, calculating the CPS values ​​of the two perturbed solution sets using the CPS formula, taking the average of the two CPS values ​​to obtain the CPS value of the decision variable, and obtaining all the CPS values ​​of the weight center, sorting the CPS values, obtaining the control attribute sorting of the decision variable, and judging whether the decision variable tends to control the convergence or diversity of the population;

[0039] The CPS formula is expressed as:

[0040]

[0041] Among them, θ i,j is the angle between the normalized straight line of the perturbation solution and the current weight center vector, CPS i,j is the CPS value, d i,j It is the distance between the two points with the farthest distance between the perturbation solutions in the direction of the weight center vector.

[0042] As a preferred solution of the virtual power plant cluster resource screening system based on multi-energy collaboration described in the present invention, it is characterized by including a resource partitioning module, a response characteristic analysis module, a resource optimization combination model building module, a constraint condition setting and solving module and a resource combination strategy formulation module.

[0043] The resource division module is used to divide the resource equipment in the virtual power plant into controllable and uncontrollable categories according to control authority, and to establish a response characteristic model of uncontrollable resources to analyze the response power provided.

[0044] The response characteristic analysis module is used to analyze the expected value and random deviation of the user's response power by using the established response characteristic model, identify the contract characteristics and power consumption characteristics of the demand response user, submit the power purchase demand forecast curve and response range, analyze the historical response deviation power of risky asset users, and the consistency and complementarity of response errors between users.

[0045] The resource optimization combination model building module is used to define the goal of the virtual power plant resource optimization combination problem, establish an objective function, determine expected income, decompose expected costs, and build an optimization combination model.

[0046] The constraint setting and solving module is used to set risk and utility constraints, resource response characteristic constraints, equation constraints and application constraints of virtual power plant aggregation external characteristics according to the analysis results of resource response characteristics, define asset portfolio weights, and solve the optimal resource combination plan.

[0047] The resource combination strategy formulation module is used to determine the control attribute ranking of decision variables through a mixed decision variable analysis method, and formulate a user response plan and a resource combination strategy according to the ranking result.

[0048] A computer device includes a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the method described in the method for virtual power plant cluster resource screening based on multi-energy collaboration are implemented.

[0049] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the method described in the method for virtual power plant cluster resource screening based on multi-energy collaboration are implemented.

[0050] Beneficial effects of the present invention: The present invention divides the resource equipment in the virtual power plant according to the control authority in detail, and establishes a response characteristic model for uncontrollable resources, thereby clarifying the resource classification, which is helpful to adopt targeted scheduling strategies, improve the efficiency of resource management and the predictability of response power; the response characteristic model is used to analyze the expected value and random deviation of the user's response power, and a resource optimization combination model is constructed to more accurately predict user behavior, reduce uncertainty risks, and optimize the resource combination effect; by setting constraints based on response characteristic analysis and using the dual mechanism of angle and distance to determine the order of decision variable control attributes, an accurate user response plan and resource combination strategy are formulated, which improves the accuracy and flexibility of decision-making, optimizes resource scheduling, and reduces operation costs. operating costs, thereby enhancing the overall benefits of virtual power plants; identifying the contractual characteristics and electricity consumption characteristics of demand response users, analyzing the historical response deviation electricity, and deeply understanding user behavior, thus improving model accuracy, achieving risk diversification, and enhancing the stability and reliability of virtual power plants; defining the objective function and decomposing the expected cost through mathematical modeling, finding the optimal resource combination plan, and maximizing economic benefits; setting risk and utility constraints, defining asset portfolio weights, ensuring the practical feasibility of the resource combination plan, and improving resource utilization efficiency and market competitiveness; setting application constraints and introducing penalty costs ensure market compliance, incentivize accurate responses, and improve market performance; the hybrid decision variable analysis method improves the scientific nature and effectiveness of decision-making, and ensures the market adaptability and leadership of virtual power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:

[0052] Figure 1 An overall flow chart of a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0053] Figure 2 A schematic diagram of a CPS update disturbance mechanism for a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0054] Figure 3 A schematic diagram of the MDA-MOEA algorithm code for a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0055] Figure 4A schematic diagram of the solution set of 300 variables on DTLZ1 of a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0056] Figure 5 A schematic diagram of the solution set of 300 variables on DTLZ3 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided in one embodiment of the present invention.

[0057] Figure 6 A schematic diagram of the solution set of 300 variables on DTLZ6 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided in one embodiment of the present invention.

[0058] Figure 7 A schematic diagram of the solution set of 300 variables on DTLZ7 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0059] Figure 8 A schematic diagram of the solution set of 300 variables on WFG2 of a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0060] Fig. 9 A schematic diagram of the solution set of 300 variables on WFG5 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided as an embodiment of the present invention.

[0061] Fig.10 A schematic diagram of the solution set of 1000 variables on DTLZ1 of a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0062] Fig.11 A schematic diagram of the solution set of 1000 variables on DTLZ3 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided as an embodiment of the present invention.

[0063] Fig.12 A schematic diagram of the solution set of 1000 variables on DTLZ6 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided as an embodiment of the present invention.

[0064] Fig.13 A schematic diagram of the solution set of 1000 variables on DTLZ7 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided by an embodiment of the present invention.

[0065] Fig.14 A schematic diagram of the solution set of 1000 variables on WFG2 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided as an embodiment of the present invention.

[0066] Fig.15A schematic diagram of the solution set of 1000 variables on WFG5 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided as an embodiment of the present invention.

[0067] Fig.16 A schematic diagram of the solution set of 1000 variables on WFG8 for a virtual power plant cluster resource screening method based on multi-energy collaboration provided as an embodiment of the present invention.

[0068] Fig.17 A system solution flow chart of a virtual power plant cluster resource screening system based on multi-energy collaboration provided for an embodiment of the present invention. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0070] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0071] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor is it an embodiment that is mutually exclusive with other embodiments, either individually or selectively.

[0072] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0073] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0074] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0075] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a virtual power plant cluster resource screening method based on multi-energy collaboration, including:

[0076] S1: Divide the resource equipment in the virtual power plant according to the control authority, analyze the response power provided by the uncontrollable resources and establish a response characteristic model.

[0077] The division according to control authority includes dividing the resources into controllable and uncontrollable categories according to the control authority of the virtual power plant over the resource equipment.

[0078] The uncontrollable resources are included in the contract signed between the virtual power plant and the user. The user has not entrusted the control authority of the terminal device to the virtual power plant and is regarded as a demand response user. The response power provided by the user is uncertain. As a risky asset, the response action of the demand response user is an incentive-based demand response. As an aggregator, the virtual power plant needs to pay attention to the actual response effect of the user.

[0079] Furthermore, the establishment of the response characteristic model includes identifying the contract characteristics of demand response users, analyzing the electricity consumption characteristics of demand response users, and requiring demand response users to submit electricity purchase demand forecast curves and response ranges to the virtual power plant before the scheduling cycle, establishing a response characteristic model for demand response users, analyzing the historical response deviation electricity of risky asset users, and analyzing the correlation between each user's return standard deviation and risky assets, and based on the correlation analysis, understanding the consistency and complementarity of response errors between different users.

[0080] The response characteristic model is expressed as:

[0081]

[0082] in, is the actual response power change of the user in the tth time period, is the lower limit of the response power achieved by the user in the tth time period, is the upper limit of the response power achieved by the user in the tth time period, is the predicted power consumption curve of the i-th user in the k-th time period, * is the target value, C i,DR,k is the total cost of the user in the kth time period, T is the total number of time periods, i is the user index, and k is the time period index.

[0083] Among them, after considering the uncertainties Rewritten as:

[0084]

[0085] in, is the actual response power change after rewriting, is the expected value of the response power allocated to the user, The random deviation of the user's response power.

[0086] The correlation analysis includes analyzing the historical response deviation of risky asset users to obtain the standard deviation of each user's income σ i and correlations between risky assets It means that the response errors of risky asset user i and risky asset user j are consistent, and over-response and under-response will occur at the same time; It indicates that the response errors between users are complementary.

[0087] S2: Use the response characteristic model to analyze the expected value and random deviation of the user's response power, and build a virtual power plant resource optimization combination model based on the analysis results.

[0088] Furthermore, the construction of the virtual power plant resource optimization combination model includes defining the goal of the virtual power plant resource optimization combination problem, establishing the objective function of the virtual power plant optimization combination model, determining the expected income, and decomposing the expected cost for analysis.

[0089] The decomposition of expected costs includes decomposing expected costs into incentive costs, uncertain costs and access costs; the optimization combination model formula is:

[0090] max:E(R p )=C in -E(C cost )

[0091] Among them, E(R p ) is the expected revenue of the virtual power plant after combination, C in is the expected revenue of the virtual power plant, E(C cost ) is the expected cost.

[0092] The expected revenue formula of the virtual power plant is:

[0093] C in =(C r -C w ) aim,t Δt

[0094] Among them, C r is the retail electricity price, C w is the wholesale electricity price in the electricity market, P aim,t is the target output power at time point t, and Δt is the time interval.

[0095] The expected cost components are expressed as:

[0096] E(C cost )=E(f1)+E(f2)+f3

[0097]

[0098] Among them, E(C cost ) is the expected cost, f1 is the incentive cost, f2 is the uncertain cost, f3 is the access cost, E(f1) is the expected incentive cost, E(f2) is the expected uncertain cost, i, j, t and k are variable indices, C i,j,k is the unit response incentive cost of interactive resources, c er is the unit deviation penalty price, P aim,t is the target output power at time t, P vpp,t is the actual output power at time point t, Δt is the time interval, λ is the access cost required for a single resource to participate in regulation, V i,j,k It is a 0 / 1 variable.

[0099] It should be noted that the setting of constraints includes setting risk and utility constraints, resource response characteristic constraints, equality constraints and application constraints of virtual power plant aggregation external characteristics according to the analysis results of resource response characteristics, and defining asset portfolio weights according to the expected return expression to solve the optimal resource combination plan.

[0100] The constraints are expressed as:

[0101]

[0102] in, is the risk and utility constraints for the optimal combination solution, is the resource response characteristic constraint, Equality constraints for the aggregation of external characteristics of the virtual power plant, To apply constraints.

[0103] According to the expected return expression of the portfolio, the weights of various asset combinations are expressed as follows:

[0104]

[0105] Among them, E(C cost ) is the expected cost, is the asset portfolio, w n,s,i is the asset portfolio weight.

[0106] Furthermore, the application constraints include: the virtual power plant participates in the energy market bidding as a third-party independent entity, submits an aggregated power baseline based on power generation forecasts and load forecasts as a bidding curve, compares the bids of the participating entities, obtains the clearing power curve, determines the power baseline forecast value and the clearing power curve, selects a user combination participating in the response based on the clearing results, and formulates a user response plan, considers external characteristic demands, and uses external characteristic demands as application constraints for the virtual power plant to participate in the electric energy market, refers to the implementation rules of the electricity spot market transactions, formulates the VPP internal settlement principles, introduces penalty costs into the objective function of the virtual power plant internal settlement principles to characterize the size of the revenue recovery in the market settlement, and the virtual power plant determines the deviation of the unbalanced electricity in the electricity market based on the market trading rules.

[0107] The deviation of the electricity market is expressed as:

[0108] -ε VPP ≤P VPP,t Δt-P aim,t Δt≤ε VPP

[0109] Among them, P aim,t is the target output power at time t, P vpp,t is the actual output power at time t, Δt is the time interval, ε VPP is the maximum unbalanced charge error.

[0110] S3: Set constraints based on the analysis results of resource response characteristics, and determine the control attribute ranking of decision variables through the dual mechanism of angle and distance, and formulate user response plans and resource combination strategies based on the ranking.

[0111] Furthermore, the formulation of user response plans and resource combination strategies includes setting constraints based on the analysis results of resource response characteristics, and determining the control attribute ranking of decision variables through a hybrid decision variable analysis method using a dual mechanism of angle and distance, and formulating user response plans and resource combination strategies based on the ranking.

[0112] like Figure 2 The hybrid decision variable analysis method includes clustering the weight vector, selecting two candidate solutions from the neighborhood of the weight center, perturbing the two candidate solutions 5 times to obtain a perturbed solution set, and normalizing the perturbed solution set to obtain two lines, using the CPS formula to calculate the CPS values ​​of the two perturbed solution sets, taking the average of the two CPS values ​​to obtain the CPS value of the decision variable, and obtaining all the CPS values ​​of the weight center, sorting the CPS values, obtaining the control attribute sorting of the decision variables, and judging whether the decision variables tend to control the convergence or diversity of the population.

[0113] The CPS formula is expressed as:

[0114]

[0115] Among them, θ i,j is the angle between the normalized straight line of the perturbation solution and the current weight center vector, CPS i,j is the CPS value, d i,j It is the distance between the two points with the farthest distance between the perturbation solutions in the direction of the weight center vector.

[0116] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0117] Example 2, reference Figure 3-16 , which is an embodiment of the present invention, provides a virtual power plant cluster resource screening method based on multi-energy collaboration. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0118] Figure 3 This is the code for the MDA-MOEA algorithm. First, the population P and reference point are randomly initialized to generate a uniform weight vector, and then CPSIndex is obtained. In the population evolution stage, the parent is randomly selected for each weight vector, and the offspring individuals are generated using a genetic algorithm. Then the reference point and neighborhood are updated, and the population is updated using a boundary crossover aggregation method based on a penalty factor until the algorithm termination condition is met.

[0119] Figure 4-16 A schematic diagram of experimental comparison of algorithms based on large-scale multi-objective optimization to verify the algorithm's solution performance; first, the experimental parameter settings are introduced, followed by the experimental results in 100 dimensions, 300 dimensions, and 1000 dimensions, and finally the impact of different aggregation function methods on the algorithm performance is compared.

[0120] Experimental parameters were set as follows: the population size N was set to 100, the maximum number of objective function evaluations was set to 100000, 300000, and 10000000 respectively when the decision dimensions were 100, 300, and 1000. The IGD results of the algorithms were compared after 20 independent runs. The algorithms compared were LMEA, MOEA / DVA, NAS-MOEA, and LSMOEA / D. All experiments were run on Intel Xeon Gold 5218R, and all the compared algorithms were implemented using MATLAB. The following table shows the basic parameter settings of the algorithms:

[0121]

[0122] For MDA-MOEA, the neighborhood size NS is set to 0.1N, the number of cluster centers C is set to 10, the number of solution selections in the mixed decision variable analysis process is set to 2, the number of perturbations for each solution is set to 6, and binary crossover and polynomial mutation methods are used in the population evolution process.

[0123] For LMEA, the number of solutions selected in decision variable clustering is set to 2, the number of perturbations NCA for each solution is set to 4, and the number of decision variable correlation analyses is set to 6.

[0124] For MOEA / DVA, the maximum number of decision variable correlation analyses is set to 6, and the number of selected solutions is set to 20.

[0125] For NAS-MOEA, the number of decision variable correlation analyses in the decision variable analysis stage is set to 6, the number of solution selections in the decision variable analysis is set to 20, and in the evolutionary convergence population stage, the initial value of the mutation operator f during differential evolution is set to 0.5, the crossover operator is 1, the size of the neighborhood is set to 0.1N, the probability of each solution being replaced by a child solution is 0.01N, and the probability of selecting a parent solution from the neighborhood is set to 0.9.

[0126] For LSMOEA / D, the neighborhood size is set to 0.1N, the size of the subgroup is set to D / 20, the number of solutions selected during the analysis is set to 2, and the number of perturbations for each solution is set to 4.

[0127] Experimental results on 100-dimensional problems: The proposed algorithms are at the same order of magnitude as the comparison algorithms on the DTLZ and WFG problems with 100 decision variables on most data sets, and the calculation differences between different algorithms for 100-dimensional decision variables are not much. Specifically, MDA-MOEA outperforms all comparison algorithms on 4 of the 14 given data sets, namely DTLZ4, DTLZ6, WFG1 and WFG4, and LMEA outperforms other algorithms on 9 of the 14 data sets. The IGD results of MOEA / DVA and NAS-MOEA on the 14 given data sets are worse than MDA-MOEA. The IGD results of MDA-MOEA and LSMOEA / D on the 14 given data sets are mostly similar, but MDA-MOEA is slightly better than LSMOEA / D.

[0128] The main reason why the IGD values ​​of various algorithms are not much different when the number of decision variables is 100 is that the amount of data of 100-dimensional decision variables is relatively small compared to the number of decision variables of LSMOPs. Therefore, in the case of 100 dimensions, the advantages of MDA-MOEA and MOEA / DVA, LMEA, NAS-MOEA, LSMOEA / D and other algorithms cannot be reflected. In some data sets, MDA-MOEA's performance is even slightly worse than other algorithms. In mixed variable analysis, MDA-MOEA innovatively changes the binary classification qualitative method commonly used by the algorithm in analyzing variable control properties to a quantitative method of a series of CPS values, which changes the way variables are identified in the subsequent optimization process, and still shows good performance when the amount of data is small.

[0129] like Figure 4-9 ,Experimental results on 300-dimensional problems: MDA-MOEA achieved good results on DTLZ1-DTLZ7 and WFG2-WFG8 problems. The results of LMEA, MOEA / DVA and NAS-MOEA on all data sets were worse than MDA-MOEA. The IGD value difference on some data sets was up to three orders of magnitude. LSMOEA / D had the best results only on WFG1 problem.

[0130] MDA-MOEA uses a boundary cross aggregation function with a penalty factor to update the population. Compared with the weighted summation method, the boundary cross aggregation method with a penalty factor can handle multi-objective problems with non-convex Pareto frontiers well. The defect that the boundary cross aggregation method must satisfy some equality constraints can also be well avoided. Therefore, the boundary cross aggregation method with a penalty factor has a wider applicability and shows good optimization effects.

[0131] like Figure 10-16,Experimental results on 1000-dimensional problems: MDA-MOEA achieved good results on problems such as DTLZ4-DTLZ7, WFG1, WFG3-5 and WFG8. NAS-MOEA achieved the best optimization effect on DTLZ1 and DTLZ3. For DTLZ2, the average optimization effect of LSMOEA / D was better than that of MDA-MOEA, but it was slightly worse than MDA-MOEA for the optimal IGD result.

[0132] MOEA / DVA, LMEA, and NAS-MOEA require more iterations to obtain a more convergent and evenly distributed population, while MDA-MOEA uses the MOEA / D framework to update the population through the neighborhood, and can converge faster with a relatively small number of iterations. The main reason is that the population update using the MOEA / D framework selects individuals in the neighborhood each time, which consumes less computing resources than the non-dominated sorting method used in the comparison algorithm. In addition, in the early stage of population evolution, the decision variables with smaller CPS values ​​are optimized, which are more inclined to evolve population convergence. Under the influence of these two factors, MDA-MOEA can evolve individuals with better convergence in a shorter number of iterations, and the performance of the algorithm can therefore be improved.

[0133] The figure shows the average final population solution set obtained by running different algorithms 20 times on six test functions such as DTLZ1 and DTLZ3 with 1000 decision variables. The black curve and three-dimensional mesh diagram represent PF, and the gray circles represent the real individuals obtained by running each algorithm. It can be seen that the overall performance of the MDA-MOEA algorithm on each test function is better than that of other algorithms, and as the dimension of decision variables increases, its performance advantage becomes more obvious. This is because MDA-MOEA can balance the relationship between population convergence and distribution during the iteration process. The population distribution obtained by MDA-MOEA on WFG5 is similar to that of LSMOEA / D, but it is better than LMEA, MOEA / DVA and NAS-MOEA in terms of distribution uniformity.

[0134] Performance comparison of different aggregation functions: On DTLZ-DTLZ and WFG4, the IGD results of MDA-MOEA are all better than those of the comparison algorithms on problems with different dimensions of 100-1000. Only on DTLZ5, WFG3, WFG7 and WFG9, the IGD results of MDA-MOEA-TE are better than those of MDA-MOEA. In addition, the IGD values ​​obtained by using these two algorithms in different dimensions have their own advantages and disadvantages. But overall, the results obtained by using the boundary crossing aggregation method based on the penalty factor are slightly better than those obtained by using the Chebyshev aggregation method. In 20 of the 48 test problems, the MDAMOEA algorithm is better than the MDA-MOEA-TE algorithm. In the comparison of algorithm running time, it can be seen that the running time required for MDA-MOEA is shorter for 46 of the 48 test functions, and the running time required for MDA-MOEA-TE is only better than that of MDA-MOEA on the 1000-dimensional WGF6 test function. Therefore, when applying the MOEA / D framework, a boundary cross aggregation method based on penalty factors is used to improve the overall performance of the algorithm.

[0135] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0136] Example 3, reference Fig.17 , which is the third embodiment of the present invention, provides a virtual power plant cluster resource screening system based on multi-energy collaboration, including a resource partitioning module 100, a response characteristic analysis module 200, a resource optimization combination model building module 300, a constraint setting and solving module 400 and a resource combination strategy formulation module 500.

[0137] The resource division module 100 is used to divide the resource equipment in the virtual power plant into controllable and uncontrollable categories according to the control authority, and to establish a response characteristic model of uncontrollable resources to analyze the response power provided.

[0138] The response characteristic analysis module 200 is used to analyze the expected value and random deviation of the user's response power by using the established response characteristic model, identify the contract characteristics and power consumption characteristics of the demand response user, submit the power purchase demand forecast curve and response range, analyze the historical response deviation power of risky asset users, and the consistency and complementarity of response errors between users.

[0139] The resource optimization combination model building module 300 is used to define the goal of the virtual power plant resource optimization combination problem, establish an objective function, determine expected revenue, decompose expected costs, and build an optimization combination model.

[0140] The constraint setting and solving module 400 is used to set risk and utility constraints, resource response characteristic constraints, equation constraints and application constraints of virtual power plant aggregation external characteristics according to the analysis results of resource response characteristics, define asset portfolio weights, and solve the optimal resource combination plan.

[0141] The resource combination strategy formulation module 500 is used to determine the control attribute ranking of decision variables through a hybrid decision variable analysis method, and formulate a user response plan and a resource combination strategy according to the ranking result.

[0142] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0143] Embodiment 4, the fourth embodiment of the present invention, is different from the first three embodiments in that:

[0144] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program codes.

[0145] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0146] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0147] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. A virtual power plant cluster resource screening method based on multi-energy collaboration, characterized by: include, Divide the resource equipment in the virtual power plant according to the control authority, analyze the response power provided by the uncontrollable resources and establish a response characteristic model; The response characteristic model is used to analyze the expected value and random deviation of the user's response power, and a virtual power plant resource optimization combination model is constructed based on the analysis results; Constraints are set based on the analysis results of resource response characteristics, and the control attribute ranking of decision variables is determined through the dual mechanisms of angle and distance. User response plans and resource combination strategies are formulated based on the ranking.

2. The virtual power plant cluster resource screening method based on multi-energy collaboration according to claim 1 is characterized in that: The division according to control authority includes dividing the resources into controllable and uncontrollable categories according to the control authority of the virtual power plant over the resource equipment; The uncontrollable resources are included in the contract signed between the virtual power plant and the user. The user has not entrusted the control authority of the terminal device to the virtual power plant and is regarded as a demand response user. The response power provided by the user is uncertain. As a risky asset, the response action of the demand response user is an incentive-based demand response. As an aggregator, the virtual power plant needs to pay attention to the actual response effect of the user.

3. The virtual power plant cluster resource screening method based on multi-energy collaboration according to claim 2 is characterized in that: The establishment of the response characteristic model includes identifying the contract characteristics of demand response users, analyzing the electricity consumption characteristics of demand response users, and requiring demand response users to submit electricity purchase demand forecast curves and response ranges to the virtual power plant before the scheduling cycle, establishing a response characteristic model for demand response users, analyzing the historical response deviation electricity of risky asset users, and analyzing the correlation between the standard deviation of each user's income and risky assets, and understanding the consistency and complementarity of response errors between different users based on the correlation analysis; The response characteristic model is expressed as: in, is the actual response power change of the user in the tth time period, is the lower limit of the response power achieved by the user in the tth time period, is the upper limit of the response power achieved by the user in the tth time period, is the predicted power consumption curve of the i-th user in the k-th time period, * is the target value, C i,DR,k is the total cost of the user in the kth time period, T is the total number of time periods, i is the user index, and k is the time period index; Among them, after considering the uncertainties Rewritten as: in, is the actual response power change after rewriting, is the expected value of the response power allocated to the user, The random deviation of the user's response power; The correlation analysis includes analyzing the historical response deviation of risky asset users to obtain the standard deviation of each user's income σ i and correlations between risky assets It means that the response errors of risky asset user i and risky asset user j are consistent, and over-response and under-response will occur at the same time; It indicates that the response errors between users are complementary.

4. The virtual power plant cluster resource screening method based on multi-energy collaboration according to claim 3 is characterized in that: The construction of the virtual power plant resource optimization combination model includes defining the goal of the virtual power plant resource optimization combination problem, establishing the objective function of the virtual power plant optimization combination model, determining the expected income, and decomposing the expected cost for analysis; The decomposition of expected costs includes decomposing expected costs into incentive costs, uncertain costs and access costs; the optimization combination model formula is: max:E(R p )=C in -E(C cost ) Among them, E(R p ) is the expected revenue of the virtual power plant after combination, C in is the expected revenue of the virtual power plant, E(C cost ) is the expected cost; The expected revenue formula of the virtual power plant is: C in =(C r -C w )P aim,t Δt Among them, C r is the retail electricity price, C w is the wholesale electricity price in the electricity market, P aim,t is the target output power at time point t, Δt is the time interval; The expected cost components are expressed as: E(C cost )=E(f1)+E(f2)+f3 Among them, E(C cost ) is the expected cost, f1 is the incentive cost, f2 is the uncertain cost, f3 is the access cost, E(f1) is the expected incentive cost, E(f2) is the expected uncertain cost, i, j, t and k are variable indices, C i,j,k is the unit response incentive cost of interactive resources, c er is the unit deviation penalty price, P aim,t is the target output power at time t, P vpp,t is the actual output power at time point t, Δt is the time interval, λ is the access cost required for a single resource to participate in regulation, V i,j,k It is a 0 / 1 variable.

5. The virtual power plant cluster resource screening method based on multi-energy collaboration according to claim 4 is characterized in that: The setting of constraint conditions includes setting risk and utility constraints, resource response characteristic constraints, equality constraints and application constraints of virtual power plant aggregation external characteristics according to the analysis results of resource response characteristics, and defining asset portfolio weights according to the expected return expression to solve the optimal resource portfolio solution; The constraints are expressed as: in, is the risk and utility constraints for the optimal combination solution, is the resource response characteristic constraint, Equality constraints for the aggregation of external characteristics of the virtual power plant, To apply constraints.

6. The virtual power plant cluster resource screening method based on multi-energy collaboration according to claim 5 is characterized in that: The application constraints include that the virtual power plant participates in the energy market bidding as a third-party independent entity, submits an aggregated power baseline based on power generation forecast and load forecast as a bidding curve, compares the bids of the participating entities, obtains a clearing power curve, determines the power baseline forecast value and the clearing power curve, selects a user combination participating in the response based on the clearing result, and formulates a user response plan, considers external characteristic requirements, and uses external characteristic requirements as application constraints for the virtual power plant to participate in the electric energy market, refers to the implementation rules of the electric power spot market transaction, formulates the VPP internal settlement principle, introduces a penalty cost into the objective function of the virtual power plant internal settlement principle to characterize the size of the revenue recovery in the market settlement, and determines the deviation of the unbalanced electricity in the electricity market based on the market transaction rules; The deviation of the electricity market is expressed as: -e VPP ≤P VPP,t Δt-P aim,t Δt≤ε VPP Among them, P aim,t is the target output power at time t, P vpp,t is the actual output power at time t, Δt is the time interval, ε VPP is the maximum unbalanced charge error.

7. The virtual power plant cluster resource screening method based on multi-energy collaboration according to claim 6 is characterized in that: The formulating of the user response plan and resource combination strategy includes setting constraint conditions according to the analysis results of the resource response characteristics, and determining the control attribute ranking of the decision variables by the angle and distance dual mechanism through the hybrid decision variable analysis method, and formulating the user response plan and resource combination strategy according to the ranking; The hybrid decision variable analysis method includes clustering the weight vector, selecting two candidate solutions from the neighborhood of the weight center, perturbing the two candidate solutions 5 times to obtain a perturbed solution set, normalizing the perturbed solution set to obtain two lines, calculating the CPS values ​​of the two perturbed solution sets using the CPS formula, taking the average of the two CPS values ​​to obtain the CPS value of the decision variable, and obtaining all the CPS values ​​of the weight center, sorting the CPS values, obtaining the control attribute sorting of the decision variable, and judging whether the decision variable tends to control the convergence or diversity of the population; The CPS formula is expressed as: Among them, θ i,j is the angle between the normalized straight line of the perturbation solution and the current weight center vector, CPS i,j is the CPS value, d i,j It is the distance between the two points with the farthest distance between the perturbation solutions in the direction of the weight center vector.

8. A system using the virtual power plant cluster resource screening method based on multi-energy collaboration as claimed in any one of claims 1 to 7, characterized in that: It includes resource partitioning module, response characteristic analysis module, resource optimization combination model building module, constraint condition setting and solving module and resource combination strategy formulation module; The resource division module is used to divide the resource equipment in the virtual power plant into controllable and uncontrollable categories according to the control authority, and to establish a response characteristic model of uncontrollable resources to analyze the response power provided; The response characteristic analysis module is used to analyze the expected value and random deviation of the user's response power by using the established response characteristic model, identify the contract characteristics and power consumption characteristics of the demand response user, submit the power purchase demand forecast curve and response range, analyze the historical response deviation power of risky asset users, and the consistency and complementarity of response errors between users; The resource optimization combination model building module is used to define the goal of the virtual power plant resource optimization combination problem, establish an objective function, determine the expected income, decompose the expected cost, and build an optimization combination model; The constraint condition setting and solving module is used to set risk and utility constraints, resource response characteristic constraints, equality constraints and application constraints of virtual power plant aggregation external characteristics according to the analysis results of resource response characteristics, define asset portfolio weights, and solve the optimal resource combination plan; The resource combination strategy formulation module is used to determine the control attribute ranking of decision variables through a mixed decision variable analysis method, and formulate a user response plan and a resource combination strategy according to the ranking result.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the virtual power plant cluster resource screening method based on multi-energy collaboration described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the virtual power plant cluster resource screening method based on multi-energy collaboration described in any one of claims 1 to 7 are implemented.

Images