Multi-objective scheduling strategy selection method and system considering source load uncertainty

Abstract

The embodiment of the application discloses a multi-target scheduling strategy selection method and system considering source load uncertainty, the method comprises the following steps: obtaining target virtual power plant data; determining the evaluation target corresponding to the target virtual power plant data to obtain a plurality of evaluation targets; constructing a multi-target optimization scheduling model according to the plurality of evaluation targets; performing fuzzy processing on the multi-target optimization scheduling model to obtain a satisfaction model; inputting the target virtual power plant data into the satisfaction model to obtain a plurality of scheduling strategies; comprehensively evaluating the plurality of scheduling strategies according to the plurality of evaluation targets to obtain a plurality of evaluation results; selecting a target evaluation result from the plurality of evaluation results; selecting the scheduling strategy corresponding to the target evaluation result to obtain a target optimal scheduling strategy. By adopting the embodiment of the application, the multi-target scheduling strategy considering economy, energy and environment is selected when the virtual power plant faces source load uncertainty.

Description

TECHNICAL FIELD

[0001] The application relates to the field of power system dispatching selection, in particular to a multi-objective dispatching strategy selection method and system considering source-load uncertainty. BACKGROUND

[0002] Under the background of vigorously developing renewable energy, the concept of virtual power plant (VPP) is proposed. As an innovative energy management form, VPP integrates distributed power generation resources, energy storage devices and controllable loads into a virtual unified body, realizing effective management and dispatching of dispersed energy. However, the volatility and uncertainty of distributed power generation resources, as well as the continuous change of power load demand, pose great challenges to the reliability and economy of VPP.

[0003] In the research field of VPP optimal dispatching, some studies use robust optimization and stochastic optimization methods to reduce the risk brought by uncertainty, while other studies use prediction technology to improve the prediction accuracy of renewable energy output and load demand. Although these methods improve the dispatching efficiency of VPP to some extent, they often only focus on specific aspects such as economy or reliability of the power system, lack comprehensive consideration of economic, energy and environmental multi-objectives, and are difficult to maintain the reliability and economy of the system at the same time.

[0004] Therefore, how to select an economic, energy and environmental multi-objective dispatching strategy for VPP in the face of source-load uncertainty has become a problem to be solved. SUMMARY

[0005] The embodiments of the application provide a multi-objective dispatching strategy selection method and system considering source-load uncertainty. By establishing a multi-objective dispatching model considering economic, energy and environmental factors, and using a fuzzy membership function method to convert the multi-objective dispatching model into a comprehensive satisfaction model, at the same time, according to the actual operation requirements of the power system, setting constraint conditions, and solving the comprehensive satisfaction model, a plurality of dispatching strategies are obtained. Based on the robust optimization model, a dispatching strategy with a satisfaction degree greater than a certain preset value is selected from the plurality of dispatching strategies, a plurality of optimal dispatching strategies are obtained, and finally the optimal optimal dispatching strategy is selected from the plurality of optimal dispatching strategies. The maximum economy, optimization of energy efficiency and minimization of environmental impact of VPP under source-load uncertainty are realized, and at the same time, the stability and reliability of the power system are ensured.

[0006] In a first aspect, the embodiments of the application provide a multi-objective dispatching strategy selection method considering source-load uncertainty, comprising:

[0007] Obtaining target virtual power plant data;

[0008] Determining evaluation targets corresponding to the target virtual power plant data to obtain multiple evaluation targets;

[0009] Constructing a multi-objective optimization scheduling model according to the multiple evaluation objectives;

[0010] Performing fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model;

[0011] Inputting the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies;

[0012] The multiple scheduling strategies are comprehensively evaluated according to the multiple evaluation targets to obtain multiple evaluation results, a target evaluation result is selected from the multiple evaluation results, and the scheduling strategy corresponding to the target evaluation result is selected to obtain the target optimal scheduling strategy.

[0013] In a second aspect, an embodiment of the present application provides a multi-objective scheduling strategy selection system considering source-load uncertainty, including: a data acquisition module, a model construction module, a model processing module, a strategy solution module, and a strategy evaluation module, wherein:

[0014] The data acquisition module is used to acquire target virtual power plant data;

[0015] The data acquisition module is further configured to determine an evaluation target corresponding to the target virtual power plant data, and obtain multiple evaluation targets;

[0016] The model building module is used to build a multi-objective optimization scheduling model according to the multiple evaluation objectives;

[0017] The model processing module is used to perform fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model;

[0018] The strategy solving module is used to input the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies;

[0019] The strategy evaluation module is used to comprehensively evaluate the multiple scheduling strategies according to the multiple evaluation targets to obtain multiple evaluation results, select a target evaluation result from the multiple evaluation results, select the scheduling strategy corresponding to the target evaluation result, and obtain the target optimal scheduling strategy.

[0020] In a third aspect, an embodiment of the present application provides an electronic device comprising: a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps in the first aspect of the embodiment of the present application.

[0021] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program for electronic data exchange, and the computer program causes a computer to perform some or all of the steps described in the first aspect of the embodiments of the present application.

[0022] In a fifth aspect, an embodiment of the present application provides a computer program product, which comprises a non-transitory computer readable storage medium storing a computer program. The computer program is operable to cause a computer to perform some or all of the steps described in the first aspect of the embodiments of the present application. The computer program product can be a software installation package.

[0023] By implementing the embodiments of the present application, the target virtual power plant data is acquired; the evaluation target corresponding to the target virtual power plant data is determined to obtain multiple evaluation targets; a multi-target optimization scheduling model is constructed according to the multiple evaluation targets; a satisfaction model is obtained by fuzzy processing the multi-target optimization scheduling model; the target virtual power plant data is input into the satisfaction model to obtain multiple scheduling strategies; the multiple scheduling strategies are comprehensively evaluated according to the multiple evaluation targets to obtain multiple evaluation results; a target evaluation result is selected from the multiple evaluation results; a scheduling strategy corresponding to the target evaluation result is selected to obtain a target optimal scheduling strategy, which realizes the selection of a multi-target scheduling strategy considering economy, energy and environment comprehensively when the virtual power plant faces source-load uncertainty. BRIEF DESCRIPTION OF DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the background art, the drawings needed to be used in the embodiments of the present application or the background art will be described below.

[0025] Figure 1 is a flowchart of a multi-target scheduling strategy selection method considering source-load uncertainty provided by the embodiments of the present application;

[0026] Figure 2 is a structural diagram of a multi-target scheduling strategy selection system considering source-load uncertainty provided by the embodiments of the present application;

[0027] Figure 3 is a structural diagram of an electronic device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0028] In order for those skilled in the art to better understand the technical scheme of the present application, the technical scheme in the embodiments of the present application will be clearly and completely described in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0029] The terms "first", "second", and the like in the specification and claims of the present application and the above-described drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the listed steps or units, but can optionally include steps or units not listed, or can optionally include other steps or units inherent to the process, method, product, or device.

[0030] In this document, the term "embodiment" means that the specific features, structures, or characteristics described in connection with the embodiment can be included in at least one embodiment of the present application. The appearance of this phrase in various places in the specification does not necessarily mean the same embodiment is referred to, nor does it mean that independent or alternative embodiments are mutually exclusive or alternative to each other. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0031] The related content, concepts, meanings, technical problems, technical schemes, beneficial effects, etc. involved in the embodiments of the present application are described below.

[0032] First, some terms involved in the present application are explained and described:

[0033] Virtual power plant: Virtual power plant is a new type of power resource management and optimization method, which aggregates distributed power generation resources (such as wind power, solar photovoltaic), energy storage systems, controllable loads and electric vehicles, etc. through advanced information communication technology and software system, realizes the coordinated optimization operation of these resources, and helps to improve energy utilization efficiency and system operation stability.

[0034] Source-load uncertainty: Source-load uncertainty mainly refers to the uncertainty of power generation (power source) and power demand (load) in the power system. This uncertainty may be caused by the volatility of renewable energy (such as changes in wind speed, solar intensity), changes in load demand (affected by weather, time, economic activity, etc.), prediction errors, equipment failures and maintenance, and market and price fluctuations, etc. These uncertain factors can be addressed through prediction technology, dispatching strategy and reserve resources.

[0035] Membership function: A membership function refers to the membership of an element to a fuzzy set, which is used to describe the degree to which an element belongs to a certain fuzzy set. It mainly maps elements to the interval [0, 1], where 0 represents that the element does not belong to the fuzzy set at all, 1 represents that the element completely belongs to the fuzzy set, and the values between 0 and 1 represent different degrees of belonging to the fuzzy set.

[0036] CRITIC weight method: CRITIC weight method is based on the comparison of evaluation indexes and the conflict between indexes to comprehensively measure the objective weight of indexes. It considers the variability of indexes and the correlation between indexes, and not the larger the number is, the more important it is. It fully utilizes the objective properties of data for scientific evaluation.

[0037] Self-Confident Double-Hierarchy Linguistic Preference Relation (SC-DHLPR): Self-Confident Double-Hierarchy Linguistic Preference Relation (SC-DHLPR) is a method for group decision-making, which expresses the preferences of decision-makers through double-layer linguistic terminology sets, and assigns a self-confidence to each Double-Hierarchy Linguistic Terminology (DHLT) to reflect the decision-makers' confidence in their preferences. This method helps to more objectively reflect the actual preferences of individuals in the group decision-making process and assists in determining the weights of decision-makers, thereby affecting the final decision-making result.

[0038] Self-confidence: Self-confidence usually refers to the individual's confidence in their judgment or preference. In the process of decision-making and evaluation, self-confidence can help quantify the individual's trust level in their choices or evaluations, thereby giving the individual's opinions appropriate weight in group decision-making.

[0039] TOPSIS method: TOPSIS method (Technique for Order Preference by Similarity to Ideal Solution, TOPSIS) is an approximation of ideal solution ranking method, i.e. the distance method between superior and inferior solutions, which is a commonly used comprehensive evaluation method. Its results can accurately reflect the gap between each evaluation scheme.

[0040] Please refer to Figure 1 , Figure 1 is a flowchart of a multi-objective scheduling strategy selection method considering source-load uncertainty provided by the embodiments of the present application, which includes but is not limited to the following steps:

[0041] S101, obtaining target virtual power plant data.

[0042] Among them, the virtual power plant (VPP) is a new form of power resource management, and the target virtual power plant data refers to the specific information set used to build and optimize the virtual power plant operation model. The target virtual power plant data can include the following data: real-time power generation data of distributed power generation facilities (such as the output power of wind turbines and photovoltaics), the storage capacity and charging and discharging status of energy storage equipment, demand response data of controllable loads (such as adjustable loads in the power system), and power grid operation data (such as electricity price information and load demand forecasts).

[0043] Specifically, smart meters and sensors can be used to collect real-time data from distributed generation facilities, energy storage equipment, and controllable loads. This collected data can be integrated and pre-processed to form the target virtual power plant data. By analyzing the target virtual power plant data, effective management and optimized scheduling of distributed energy resources can be achieved.

[0044] S102: Determine the evaluation targets corresponding to the target virtual power plant data to obtain multiple evaluation targets.

[0045] Specifically, in VPP operations and management, evaluation objectives are key metrics for measuring performance and benefits, typically including Virtual Power Plant Consolidated Profit (VPPCP), Renewable Energy Usage Rate (REUR), and Carbon Net Processing Capacity (CNPC). By analyzing VPP scheduling strategies based on multiple evaluation objectives, a comprehensive assessment is ensured from economic, environmental, and energy perspectives.

[0046] The comprehensive profit of the virtual power plant is the comprehensive profit of the VPP, which is mainly calculated from the operating income and costs incurred during the operation of the VPP. The calculation formula is as follows:

[0047] P VPPCP =R VPP -[C dev +C buy +C d-r ]

[0048] In the above formula, P VPPCP is the comprehensive profit of the virtual power plant. When solving the scheduling strategy, the maximum value is usually taken as the solution target; R VPP The income of VPP operators; C dev is the equipment operating cost of VPP; C buy is the energy purchase cost of VPP; Cd-r The demand response cost of the VPP. The cost in the above is the cost generated by the VPP in operation.

[0049] For the income of the VPP operator, it mainly includes the income generated by selling electric energy to the power grid company, power generation group, etc. and the renewable energy subsidy, and its calculation formula is as follows:

[0050]

[0051] In the above formula, R sell is the income of the VPP operator selling electric energy; R RE-sub is the renewable energy subsidy; T is the operation cycle of the VPP dispatching; is the electricity price on the grid at t time; is the electricity quantity sold at t time; β RE-sub is the price of the renewable energy subsidy; is the renewable energy generation quantity at t time.

[0052] For the equipment operation cost, it mainly includes the charge and discharge cost of the energy storage system and the operation cost of the wind turbine and photovoltaic equipment, and its calculation formula is as follows:

[0053]

[0054] In the above formula, h es is the cost coefficient of the energy storage system, which is generally a default coefficient set in advance; P es,pro (t) is the discharge power of the energy storage system at t time; P es,abo (t) is the charge power of the energy storage system at t time; h pv is the cost coefficient of the distributed photovoltaic equipment in the VPP; h wind is the cost coefficient of the distributed wind turbine equipment in the VPP; P pv (t) is the actual power generation of the photovoltaic equipment at t time; P wind (t) is the actual power generation of the wind turbine equipment at t time.

[0055] For the energy purchase cost, it mainly includes the energy cost consumed by the VPP in dispatching, which can include the cost of consuming electric energy and natural gas, and the calculation formula of the energy purchase cost is as follows:

[0056]

[0057] In the above formula, h e,buy (t) is the electricity price at t time; P e,buy (t) is the power of the purchased electric energy at t time; h g,buy (t) is the natural gas purchase price at t time; Pg,buy (t) is the gas purchase amount at time t.

[0058] For the demand response cost, it mainly includes the cost generated by the VPP in response to the scheduling strategy, and its calculation formula is as follows:

[0059]

[0060] In the above formula, h d-r (t) is the subsidy cost coefficient of the user load participating in demand response in the VPP at time t, which is mainly determined according to the way of participating in demand response by the user, and is generally set in advance; P d-r (t) is the power of the user load participating in demand response at time t.

[0061] Among them, the renewable energy utilization rate is the utilization rate of photovoltaic, wind turbine and other renewable energy after the VPP scheduling, and its calculation formula is as follows:

[0062]

[0063] In the above formula, A RE is the renewable energy utilization rate, which is expressed in the form of ratio, and the maximum value is usually taken as the solving target when solving the scheduling strategy; P RE,real is the actual use amount of renewable energy; P RE,all is the total output of renewable energy.

[0064] Among them, the carbon net emission amount is the difference between the carbon dioxide emission and absorption amount under the scheduling of the VPP, which is used to represent the net emission when the VPP interacts with carbon dioxide, and its calculation formula is as follows:

[0065]

[0066] In the above formula, is the carbon net emission amount, and the minimum value is usually taken as the solving target when solving the scheduling strategy; C release is the carbon emission amount; C abs is the carbon absorption amount; is the natural gas usage amount at time t; p gas is the coefficient of carbon dioxide generated when natural gas is used; u is a carbon fixation system, which can be an ecological system such as green land, forest and lake, and is used to absorb carbon dioxide; is the amount of carbon dioxide absorbed by different carbon fixation systems per unit area at time t; s u is the area of the carbon fixation system.

[0067] Specifically, after determining the evaluation targets corresponding to the evaluation VPP scheduling strategy according to the target virtual power plant data, a plurality of evaluation targets are obtained, each of which has a corresponding calculation formula, and the actual target value of each evaluation target can be obtained through the calculation formula. At the same time, a multi-objective optimization scheduling model for determining the VPP can be constructed according to the plurality of evaluation targets, and by solving the multi-objective optimization scheduling model, the objectives such as profit maximization, renewable energy use maximization and carbon emission minimization can be achieved.

[0068] S103, constructing a multi-objective optimization scheduling model according to the plurality of evaluation targets.

[0069] Specifically, three key evaluation targets required for evaluating the scheduling strategy of the VPP are determined, i.e., virtual power plant comprehensive profit, renewable energy use rate and carbon net emission amount, and the three key evaluation targets are used as the basis for constructing the multi-objective optimization scheduling model. The multi-objective optimization scheduling model includes a target function defining each evaluation target and a constraint condition for each evaluation target. By solving the multi-objective optimization scheduling model and considering the plurality of evaluation targets and constraint conditions, an optimal or approximately optimal solution can be found, and an optimal VPP scheduling strategy considering the plurality of evaluation targets can be achieved.

[0070] Optionally, the step S103 of constructing a multi-objective optimization scheduling model according to the plurality of evaluation targets can further include the following steps:

[0071] A31, obtaining a target function corresponding to each evaluation target in the plurality of evaluation targets, to obtain a plurality of target functions;

[0072] A32, determining a constraint condition corresponding to each target function according to the plurality of target functions, to obtain a plurality of constraint conditions;

[0073] A33, constructing the multi-objective optimization scheduling model according to the plurality of target functions and the plurality of constraint conditions.

[0074] Specifically, when constructing the multi-objective optimization scheduling model, a target function corresponding to each evaluation target in the plurality of evaluation targets can be obtained, to obtain a plurality of target functions. The target function is a calculation formula corresponding to each evaluation target, and for each target function, profit maximization, renewable energy use maximization and carbon emission minimization are used as the solving objectives when solving the scheduling strategy.

[0075] Further, a constraint condition corresponding to each target function can be determined according to the plurality of target functions, to obtain a plurality of constraint conditions, wherein the plurality of constraint conditions include at least one of the following: power balance constraint, renewable energy output constraint, energy storage system related constraint and demand response constraint.

[0076] wherein the power balance constraint is used to ensure the balance between the power input and output of the VPP, i.e. the total power generated by the VPP at any time (including renewable energy, discharging of the energy storage system, power purchase, etc.) is equal to the total power consumption of the VPP (including load demand, charging of the energy storage system, power sale to the grid, etc.), and the constraint condition is as follows:

[0077] P pv P wind P d－r P e,buy P es,pro P es,abo P load,e P e,sell P

[0078] In the above formula, P pv (t) is the actual power generation of the photovoltaic device at time t; P wind (t) is the actual power generation of the wind turbine at time t; P d-r (t) is the power of user load demand response at time t; P e,buy (t) is the power of purchased electricity at time t; P es,pro (t) is the discharging power of the energy storage system at time t; P es,abo (t) is the charging power of the energy storage system at time t; P load,e (t) is the load demand power at time t; and P e,sell (t) is the power of power sale to the grid.

[0079] wherein the renewable energy output constraint is used to ensure that the power generation of renewable energy such as photovoltaic and wind turbine within the VPP is within the predicted maximum power range, and by predicting the maximum power range of renewable energy, the volatility and uncertainty of renewable energy can be effectively managed, and the constraint condition is as follows:

[0080]

[0081] In the above formula, P pv,pre (t) is the predicted power generation of photovoltaic within the VPP at time t; and P wind,pre (t) is the predicted power generation of wind turbine within the VPP at time t.

[0082] wherein the energy storage system related constraint includes the energy storage system capacity and state of charge constraint, and the energy storage system power and charging / discharging state constraint.

[0083] For the energy storage system capacity and state of charge constraint, the constraint condition is as follows:

[0084]

[0085] In the above formula, E(t) is the state of charge of the energy storage system at time t; η es is the charge-discharge coefficient of the energy storage system; P es,abo (t) is the charging power of the energy storage system at time t; P es,pro (t) is the discharging power of the energy storage system at time t; E(24) is the state of charge of the energy storage system at the 24th time unit (usually 24 time units per day); E i (0) is the initial state of charge of the energy storage system at the initial time of a day; E min is the minimum state of charge of the energy storage system; E max is the maximum state of charge of the energy storage system in a safe state.

[0086] For the power and charge-discharge state constraints of the energy storage system, the constraint conditions are as follows:

[0087]

[0088] In the above formula, P es,max is the maximum charge-discharge power of the energy storage system, and P es,max is usually represented by M in the big M method, which is used for linearization of the nonlinear constraint; U es,abo (t) is the charging state of the energy storage system at time t, and is 1 when charging and 0 when not charging; U es,pro (t) is the discharging state of the energy storage system at time t, and is 1 when discharging and 0 when not discharging; the charging state and the discharging state of the energy storage system cannot be 1 at the same time.

[0089] The demand response constraints include demand response capacity constraints and demand response speed constraints.

[0090] For the demand response capacity constraints, the constraint conditions are as follows:

[0091] {P d-r,valley,max (t)≤P d-r (t)≤P d-r,peak,max (t)

[0092] In the above formula, P d-r,valley,max (t) is the maximum power of the user in the VPP at time t to participate in the valley filling type demand response (i.e., to increase power consumption when the power demand is low); P d-r,peak,max (t) is the maximum power of the user in the VPP at time t to participate in the peak cutting type demand response (i.e., to reduce power consumption when the power demand is high).

[0093] For the demand response speed constraints, the constraint conditions are as follows:

[0094] -P d-r,limit ≤P d-r (t)-P d-r (t-1)≤P d-r,limit

[0095] In the above formula, P d-r,limit is the maximum power change limit of the power load participating in demand response, which is generally set by default or in advance by the system.

[0096] Further, the multiple objective functions and the multiple constraint conditions are integrated into a multi-objective optimization scheduling model, which can simultaneously consider multiple evaluation targets and multiple constraint conditions, and then solve multiple or optimal feasible scheduling strategies to balance the economy, renewable energy utilization rate and environmental impact of the virtual power plant.

[0097] S104, fuzzy processing the multi-objective optimization scheduling model to obtain a satisfaction model.

[0098] Specifically, in the multi-objective optimization problem, the objective functions and the constraint conditions often have uncertainty and fuzziness, and direct processing may lead to a model that is too complex or difficult to solve. Therefore, the multi-objective optimization scheduling model is fuzzy processed to obtain a single-objective optimization scheduling model, i.e., a satisfaction model, to improve the practicability and flexibility of the model.

[0099] Optionally, the step S104 of fuzzy processing the multi-objective optimization scheduling model to obtain a satisfaction model can further include the following steps:

[0100] A41, determining a membership function corresponding to each objective function in the multiple objective functions to obtain multiple membership functions;

[0101] A42, obtaining a weight coefficient corresponding to each evaluation target to obtain multiple weight coefficients;

[0102] A43, constructing the satisfaction model according to the multiple membership functions and the multiple weight coefficients.

[0103] Specifically, the membership function corresponding to each objective function in the multiple objective functions can be determined by a pre-set fuzzy processing tool to obtain multiple membership functions. These membership functions map the values of the objective functions to the interval [0, 1], where 1 represents the optimal state of the objective function value and 0 represents the worst state of the objective function value. Then, a weight coefficient corresponding to each evaluation target is obtained to obtain multiple weight coefficients, and the sum of the multiple weight coefficients is 1. According to the multiple membership degrees and the multiple weight coefficients, a satisfaction model can be constructed, which is as follows:

[0104]

[0105] In the above formula, Z is the satisfaction calculated by the satisfaction model; ψ1 is the weight coefficient corresponding to the comprehensive profit of the virtual power plant; χ1(P VPPCP ) is the membership function corresponding to the comprehensive profit target function of the virtual power plant; ψ2 is the weight coefficient corresponding to the renewable energy usage rate; χ2(A RE ) is the membership function corresponding to the carbon net emission target function.

[0106] Optionally, the step A41 of determining the membership function corresponding to each target function in the plurality of target functions to obtain a plurality of membership functions can further include the following steps.

[0107] B41, obtaining a reference target function; the reference target function is any target function in the plurality of target functions;

[0108] B42, determining the maximum value and the minimum value of the reference target function according to the target virtual power plant data;

[0109] B43, inputting the maximum value and the minimum value into a preset fuzzy membership function to obtain the membership function corresponding to the reference target function.

[0110] Specifically, any target function in the plurality of target functions is obtained as a reference target function for main analysis, and the data range of the reference target function can be determined according to the target virtual power plant data, and the maximum value and the minimum value of the reference target function can be determined according to the data range.

[0111] Further, the maximum value and the minimum value are input into a preset fuzzy membership function to obtain the membership function corresponding to the reference target function, wherein if the target function corresponding to the virtual power plant comprehensive profit or the renewable energy usage rate is selected as the reference target function, the greater the target function value is, the better the scheduling performance is, and then an upward half Γ-type membership function can be used for fuzzy analysis of the reference target function to obtain the membership function of the virtual power plant comprehensive profit or the renewable energy usage rate, and if the carbon net emission is selected as the reference target function, the smaller the target function value is, the better the scheduling performance is, and then a downward half Γ-type membership function can be used for fuzzy analysis of the reference target function to obtain the membership function of the carbon net emission.

[0112] The upward half Γ-type membership function is as follows:

[0113]

[0114] In the above formula, k(f) is the membership function; f is the target function value; f​max is the maximum value of the objective function.

[0115] wherein the decreasing half Γ membership function is shown as follows:

[0116]

[0117] In the above formula, f min is the minimum value of the objective function.

[0118] Optionally, the step A42 of obtaining the weight coefficients corresponding to the plurality of evaluation targets to obtain the plurality of weight coefficients can further include the following steps:

[0119] C41, performing objective weight analysis on the plurality of evaluation targets based on an objective weighting method to obtain a plurality of objective weights;

[0120] C42, determining the decision weight corresponding to each evaluation target in the plurality of evaluation targets according to the self-confidence double-layer language preference relationship to obtain a plurality of decision weights;

[0121] C43, multiplying the plurality of objective weights and the plurality of decision weights one by one to obtain a plurality of combined weights;

[0122] C44, performing normalization processing on the plurality of combined weights to obtain the plurality of weight coefficients.

[0123] Specifically, the objective weight analysis is performed on the plurality of evaluation targets based on the objective weighting method to obtain a plurality of objective weights. The objective weighting method can adopt the CRITIC weight method, which mainly performs objective weight analysis according to the evaluation target values of different evaluation objects in the plurality of evaluation targets, and determines the objective weight of each evaluation target by calculating the standard deviation and correlation coefficient of different evaluation targets. Then, the decision weight corresponding to each evaluation target in the plurality of evaluation targets is determined according to the self-confidence double-layer language preference relationship to obtain a plurality of decision weights, and the plurality of objective weights and the plurality of decision weights are combined, i.e. the objective weight and the decision weight are multiplied one by one to obtain a plurality of combined weights. Finally, the plurality of combined weights are normalized to obtain a plurality of weight coefficients, which are used to balance the importance of each evaluation target in the satisfaction model.

[0124] Optionally, the step C42 of determining the decision weight corresponding to each evaluation target in the plurality of evaluation targets according to the self-confidence double-layer language preference relationship to obtain a plurality of decision weights can further include the following steps:

[0125] D41, obtaining a preset double-layer language term set; the preset double-layer language term set includes a plurality of double-layer language terms;

[0126] D42. Evaluate the multiple evaluation targets according to the preset two-level language terminology set to obtain multiple evaluation results;

[0127] D43, determining the confidence level corresponding to each of the multiple evaluation results to obtain multiple confidence levels;

[0128] D44. Perform a weight analysis on each of the multiple evaluation targets based on the multiple evaluation results and the multiple confidence levels to obtain multiple decision weights.

[0129] Specifically, a preset two-level language term set is obtained, which includes multiple two-level language terms. The two-level language terms can be terms such as good, average, and poor, and each two-level language term corresponds to a weight coefficient. Multiple evaluation targets are evaluated according to the preset level language term set to obtain multiple evaluation results, that is, a weight coefficient is assigned to each evaluation target. Then, for each evaluation result, a confidence level can be assigned to each evaluation result. The confidence level can be preset or determined by the decision maker's preference. A weight analysis is performed on each evaluation target in the multiple evaluation targets based on the multiple evaluation results and the multiple confidence levels, that is, the weight of each evaluation target is calculated based on the evaluation results and the confidence level, that is, the decision weight. The decision weight is used to reflect the relative importance of each evaluation target.

[0130] Optionally, the decision weight of each evaluation target may be obtained through a subjective weight analysis method, which may be a hierarchical analysis method or the like.

[0131] Optionally, taking into account source-load uncertainty, the satisfaction model can be robustly optimized, ensuring that the dispatch strategy solution is robust to multiple uncertainties under uncertain conditions. Furthermore, the wind turbine and photovoltaic outputs, as well as the load, are considered as uncertain parameters, and a power difference model is constructed to analyze the uncertainty of the difference between actual and predicted outputs.

[0132] The power difference model is as follows:

[0133]

[0134] In the above formula, is the predicted value of power difference; P e,t is the actual value of the power difference; α e is the fluctuation amplitude, which is a non-negative parameter used to quantify the power difference prediction value The uncertainty of the prediction value is higher. is the uncertainty set of the power difference, which means that the actual value of the power difference P e,t The value range of is the predicted value of photovoltaic output; is the predicted value of wind turbine output; is the load output forecast value.

[0135] Furthermore, a robust model can be set so that under the robust model, the satisfaction of the scheduling strategy solved according to the satisfaction model is not lower than the satisfaction threshold, that is, a feasible scheduling strategy can be obtained. The robust model is as follows:

[0136]

[0137] In the above formula, α is the possible value of the fluctuation range, and the maximum value can be taken as the fluctuation range; f1 is the acceptable satisfaction; f0 is the optimal target value of the satisfaction model; β CO is the deviation factor, which is a parameter between 0 and 1, indicating the degree of deviation of f0 above f1.

[0138] Specifically, based on the power difference model and the robust model, it can be determined that the satisfaction model can obtain a feasible scheduling strategy under multiple uncertainties, that is, the power difference is When the range changes arbitrarily, the satisfaction of the scheduling strategy is not less than (1-β CO )f0, then the scheduling strategy is a feasible solution.

[0139] S105: Input the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies.

[0140] Specifically, the membership of different evaluation targets is first calculated. The evaluation target values ​​of different evaluation targets can be determined based on the target virtual power plant data. The evaluation target values ​​are input into the membership function corresponding to each evaluation target to obtain the membership corresponding to different evaluation targets. The satisfaction of the scheduling strategy obtained under the input parameters is obtained based on the weight coefficients corresponding to multiple evaluation targets. If the satisfaction is greater than the preset satisfaction threshold, it can be determined that the scheduling strategy is a feasible scheduling strategy.

[0141] In one possible embodiment, different input parameters can be adjusted according to the satisfaction model. Since the weight coefficient corresponding to each evaluation target has been determined, under different management, such as different power generation combinations and scheduling sequences, different energy storage system charging and discharging strategies, different demand response and load management strategies, and different power trading and price negotiation strategies, the membership value corresponding to each evaluation target is different, so multiple scheduling strategies can be obtained.

[0142] Optionally, the above step S105, inputting the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies, may further include the following steps:

[0143] A51, inputting the target virtual power plant data into the plurality of constraint conditions to obtain a plurality of reference scheduling strategies; each reference scheduling strategy comprises reference evaluation target values of the plurality of evaluation targets;

[0144] A52, selecting any reference scheduling strategy from the plurality of reference scheduling strategies to obtain a target reference scheduling strategy;

[0145] A53, obtaining reference evaluation target values of each evaluation target of the target reference scheduling strategy to obtain a plurality of reference evaluation target values;

[0146] A54, inputting the plurality of reference evaluation target values into the membership function corresponding to each evaluation target to obtain a plurality of membership values;

[0147] A55, determining a target satisfaction degree of the reference scheduling strategy according to the plurality of membership values;

[0148] A56, when the target satisfaction degree is greater than a preset satisfaction threshold, determining that the target reference scheduling strategy is a feasible scheduling strategy;

[0149] A57, determining the plurality of scheduling strategies according to the feasible scheduling strategy.

[0150] Specifically, target virtual power plant data is input into a plurality of constraint conditions to obtain a plurality of reference scheduling strategies, each reference scheduling strategy comprises reference evaluation target values of a plurality of evaluation targets, any reference scheduling strategy is selected from the plurality of reference scheduling strategies to obtain a target reference scheduling strategy, reference evaluation target values of each evaluation target of the target reference scheduling strategy are obtained to obtain a plurality of reference evaluation target values, the plurality of reference evaluation target values are input into the membership function corresponding to each evaluation target to obtain a plurality of membership values, a target satisfaction degree of the reference scheduling strategy is determined according to the plurality of membership values, when the target satisfaction degree is greater than a preset satisfaction threshold, it can be determined that the target reference scheduling strategy is a feasible scheduling strategy, and through the above calculation method, a plurality of scheduling strategies can be determined from the plurality of reference scheduling strategies.

[0151] S106, comprehensively evaluating the plurality of scheduling strategies according to the plurality of evaluation targets to obtain a plurality of evaluation results, selecting a target evaluation result from the plurality of evaluation results, selecting a scheduling strategy corresponding to the target evaluation result to obtain a target optimal scheduling strategy.

[0152] Specifically, the optimized TOPSIS method can be used to evaluate each scheduling strategy to select the optimal scheduling strategy. The optimized TOPSIS method is used to comprehensively evaluate the multiple scheduling strategies according to the multiple evaluation targets to obtain multiple evaluation results, the evaluation results are used to represent the merit values of different scheduling strategies, one target evaluation result is selected from the multiple evaluation results, the target evaluation result is closest to the optimal scheme, and the scheduling strategy corresponding to the target evaluation result can obtain the target optimal scheduling strategy.

[0153] Optionally, the step S106 of comprehensively evaluating the multiple scheduling strategies according to the multiple evaluation targets to obtain multiple evaluation results can further include the following steps:

[0154] A61, constructing a target decision matrix according to the multiple scheduling strategies; each scheduling strategy includes evaluation target values of the multiple evaluation targets;

[0155] A62, performing normalization processing on the target decision matrix to obtain a target standard matrix;

[0156] A63, constructing a weighted standard matrix according to the multiple weight coefficients and the target standard matrix;

[0157] A64, determining ideal solutions of each evaluation target according to the weighted standard matrix to obtain positive ideal solutions and negative ideal solutions; the positive ideal solution is a maximum value in the multiple evaluation target values corresponding to the evaluation target; and the negative ideal solution is a minimum value in the multiple evaluation target values corresponding to the evaluation target;

[0158] A65, determining distances between the evaluation target values corresponding to the evaluation target of each scheduling strategy in the multiple scheduling strategies and the positive ideal solution to obtain multiple first distances;

[0159] A66, determining distances between the evaluation target values corresponding to the evaluation target of each scheduling strategy in the multiple scheduling strategies and the negative ideal solution to obtain multiple second distances;

[0160] A67, determining evaluation results of each scheduling strategy in the multiple scheduling strategies according to the multiple first distances and the multiple second distances to obtain multiple evaluation results.

[0161] Specifically, a target decision matrix is constructed according to the multiple scheduling strategies, a row of the target decision matrix is each scheduling strategy, and a column of the target decision matrix is an evaluation target value of each evaluation. Then, in order to eliminate the influence of different evaluation target dimensions and orders of magnitude, normalization processing is performed on the target decision matrix to obtain a target standard matrix, and then a weight coefficient of each evaluation target is applied to process the target standard matrix, that is, each evaluation target value is multiplied by the weight coefficient corresponding to the evaluation target, and then a weighted standard matrix is constructed.

[0162] Further, the ideal solution of each evaluation target is determined according to the weighted normalized matrix, to obtain a positive ideal solution and a negative ideal solution, wherein the positive ideal solution is the maximum value of the plurality of evaluation target values corresponding to the evaluation target, and the negative ideal solution is the minimum value of the plurality of evaluation target values corresponding to the evaluation target. Then, the distance between the evaluation target value corresponding to the evaluation target of each scheduling strategy in the plurality of scheduling strategies and the positive ideal solution is determined to obtain a plurality of first distances, and the distance between the evaluation target value corresponding to the evaluation target of each scheduling strategy in the plurality of scheduling strategies and the negative ideal solution is determined to obtain a plurality of second distances.

[0163] The calculation formula of the first distance is as follows:

[0164]

[0165] In the above formula, is the first distance, that is, the distance between each evaluation target in the scheduling strategy and the positive ideal solution; j is the jth evaluation target; m is the total of m evaluation targets; z ij is the value of the jth evaluation target of the ith scheduling strategy in the weighted normalized matrix; is the positive ideal solution; n is the total of n scheduling strategies to be evaluated.

[0166] The calculation formula of the first distance is as follows:

[0167]

[0168] In the above formula, is the second distance, that is, the distance between each evaluation target in the scheduling strategy and the negative ideal solution; is the negative ideal solution.

[0169] Further, the evaluation result of each scheduling strategy in the plurality of scheduling strategies is determined according to the plurality of first distances and the plurality of second distances. The evaluation result of the scheduling strategy is determined by the first distance and the second distance. The smaller the first distance is, the closer to the positive ideal solution it is. The smaller the second distance is, the closer to the negative ideal solution it is. Therefore, the scheduling strategy closest to the positive ideal solution and farthest from the negative ideal solution is the optimal scheduling strategy, that is, the best evaluation result.

[0170] The calculation formula of the evaluation result is as follows:

[0171]

[0172] In the above formula, C i is the evaluation result of the ith scheduling strategy, and its value range is [0, 1]. The closer to 1, the better the scheduling strategy is.

[0173] Optionally, the step A62 of normalizing the target decision matrix to obtain a target standard matrix can further include the following steps:

[0174] B61, inputting each evaluation target value in the target decision matrix into a membership function corresponding to the evaluation target to obtain a plurality of standard memberships;

[0175] B61, constructing the target standard matrix according to the plurality of standard memberships.

[0176] Specifically, in order to reduce the influence of data dimension on the calculation result, each matrix element in the target decision matrix can be normalized according to the membership function of each evaluation target, that is, each evaluation target value in the target decision matrix is input into the membership function corresponding to the evaluation target to obtain a plurality of standard memberships, the value range of the standard membership is [0, 1], and therefore, the target standard matrix can be constructed according to the plurality of standard memberships.

[0177] In the embodiment of the present application, according to the plurality of evaluation results, the scheduling strategy closest to the positive ideal solution and farthest from the negative ideal solution can be selected as the target optimal scheduling strategy, the target optimal scheduling strategy has the maximum satisfaction degree in the plurality of scheduling strategies, and the maximization of comprehensive benefits can be realized.

[0178] In summary, by implementing the embodiments of the present application, the target virtual power plant data can be obtained, the evaluation targets corresponding to the target virtual power plant data are determined to obtain a plurality of evaluation targets, a multi-objective optimization scheduling model is constructed according to the plurality of evaluation targets, a satisfaction degree model is obtained by fuzzy processing the multi-objective optimization scheduling model, the target virtual power plant data is input into the satisfaction degree model to obtain a plurality of scheduling strategies, the plurality of scheduling strategies are comprehensively evaluated according to the plurality of evaluation targets to obtain a plurality of evaluation results, a target evaluation result is selected from the plurality of evaluation results, a scheduling strategy corresponding to the target evaluation result is selected to obtain a target optimal scheduling strategy, and the multi-objective scheduling strategy considering economy, energy and environment is selected when the virtual power plant faces source-load uncertainty.

[0179] Please refer to Figure 2 , Figure 2 is a structure diagram of a multi-objective scheduling strategy selection system considering source-load uncertainty provided by the embodiments of the present application, the multi-objective scheduling strategy selection system considering source-load uncertainty 200 includes a data acquisition module 201, a model construction module 202, a model processing module 203, a strategy solving module 204, and a strategy evaluation module 205, wherein,

[0180] The data acquisition module 201 is configured to acquire target virtual power plant data.

[0181] The data acquisition module 201 is further configured to determine evaluation targets corresponding to the target virtual power plant data, and obtain a plurality of evaluation targets.

[0182] The model construction module 202 is configured to construct a multi-objective optimization scheduling model according to the plurality of evaluation targets.

[0183] The model processing module 203 is configured to perform fuzzy processing on the multi-objective optimization scheduling model, and obtain a satisfaction model.

[0184] The strategy solving module 204 is configured to input the target virtual power plant data into the satisfaction model, and obtain a plurality of scheduling strategies.

[0185] The strategy evaluation module 205 is configured to comprehensively evaluate the plurality of scheduling strategies according to the plurality of evaluation targets, obtain a plurality of evaluation results, select a target evaluation result from the plurality of evaluation results, select a scheduling strategy corresponding to the target evaluation result, and obtain a target optimal scheduling strategy.

[0186] Optionally, in the aspect of constructing the multi-objective optimization scheduling model according to the plurality of evaluation targets, the model construction module 202 is further configured to:

[0187] acquire a target function corresponding to each evaluation target in the plurality of evaluation targets, and obtain a plurality of target functions;

[0188] determine a constraint condition corresponding to each target function according to the plurality of target functions, and obtain a plurality of constraint conditions;

[0189] construct the multi-objective optimization scheduling model according to the plurality of target functions and the plurality of constraint conditions.

[0190] Optionally, in the aspect of performing fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model, the model processing module 203 is further configured to:

[0191] determine a membership function corresponding to each target function in the plurality of target functions, and obtain a plurality of membership functions;

[0192] acquire a weight coefficient corresponding to a plurality of evaluation targets, and obtain a plurality of weight coefficients;

[0193] construct the satisfaction model according to the plurality of membership functions and the plurality of weight coefficients.

[0194] Optionally, in the aspect of determining a membership function corresponding to each target function in the plurality of target functions to obtain a plurality of membership functions, the model processing module 203 is further configured to:

[0195] obtaining a reference objective function; the reference objective function is any objective function in the plurality of objective functions;

[0196] determining a maximum value and a minimum value of the reference objective function according to the target virtual power plant data;

[0197] inputting the maximum value and the minimum value into a preset fuzzy membership function to obtain a membership function corresponding to the reference objective function.

[0198] Optionally, in the aspect of obtaining a plurality of weight coefficients corresponding to a plurality of evaluation objectives, the model processing module 203 is further specifically configured to:

[0199] performing objective weight analysis on the plurality of evaluation objectives based on an objective weighting method to obtain a plurality of objective weights;

[0200] determining a decision weight corresponding to each evaluation objective in the plurality of evaluation objectives according to a self-confidence double-layer language preference relationship to obtain a plurality of decision weights;

[0201] multiplying the plurality of objective weights and the plurality of decision weights one by one to obtain a plurality of combined weights;

[0202] performing normalization processing on the plurality of combined weights to obtain the plurality of weight coefficients.

[0203] Optionally, in the aspect of determining a decision weight corresponding to each evaluation objective in the plurality of evaluation objectives according to a self-confidence double-layer language preference relationship to obtain a plurality of decision weights, the model processing module 203 is further specifically configured to:

[0204] obtaining a preset double-layer language term set; the preset double-layer language term set includes a plurality of double-layer language terms;

[0205] evaluating the plurality of evaluation objectives according to the preset double-layer language term set to obtain a plurality of evaluation results;

[0206] determining a self-confidence degree corresponding to each evaluation result in the plurality of evaluation results to obtain a plurality of self-confidence degrees;

[0207] performing weight analysis on each evaluation objective in the plurality of evaluation objectives according to the plurality of evaluation results and the plurality of self-confidence degrees to obtain a plurality of decision weights.

[0208] Optionally, in the aspect of inputting the target virtual power plant data into the satisfaction model to obtain a plurality of scheduling strategies, the strategy solving module 204 is further specifically configured to:

[0209] inputting the target virtual power plant data into the plurality of constraint conditions to obtain a plurality of reference scheduling strategies; each reference scheduling strategy comprises reference evaluation target values of the plurality of evaluation targets;

[0210] selecting any reference scheduling strategy from the plurality of reference scheduling strategies to obtain a target reference scheduling strategy;

[0211] obtaining reference evaluation target values of each evaluation target of the target reference scheduling strategy to obtain a plurality of reference evaluation target values;

[0212] inputting the plurality of reference evaluation target values into a membership function corresponding to each evaluation target to obtain a plurality of membership values;

[0213] determining a target satisfaction degree of the reference scheduling strategy according to the plurality of membership values;

[0214] when the target satisfaction degree is greater than a preset satisfaction threshold, determining that the target reference scheduling strategy is a feasible scheduling strategy;

[0215] determining the plurality of scheduling strategies according to the feasible scheduling strategy.

[0216] Optionally, in the comprehensive evaluation of the plurality of scheduling strategies according to the plurality of evaluation targets to obtain a plurality of evaluation results, the strategy evaluation module 205 is further specifically configured to:

[0217] constructing a target decision matrix according to the plurality of scheduling strategies; each scheduling strategy comprises evaluation target values of the plurality of evaluation targets;

[0218] performing normalization processing on the target decision matrix to obtain a target standard matrix;

[0219] constructing a weighted standard matrix according to the plurality of weight coefficients and the target standard matrix;

[0220] determining ideal solutions of each evaluation target according to the weighted standard matrix to obtain positive ideal solutions and negative ideal solutions; the positive ideal solution is a maximum value in the plurality of evaluation target values corresponding to the evaluation target; and the negative ideal solution is a minimum value in the plurality of evaluation target values corresponding to the evaluation target;

[0221] determining distances between evaluation target values corresponding to each scheduling strategy in the plurality of scheduling strategies and the positive ideal solution to obtain a plurality of first distances;

[0222] determining distances between evaluation target values corresponding to each scheduling strategy in the plurality of scheduling strategies and the negative ideal solution to obtain a plurality of second distances;

[0223] The evaluation results of each scheduling strategy in the plurality of scheduling strategies are determined according to the plurality of first distances and the plurality of second distances, and a plurality of evaluation results are obtained.

[0224] Optionally, after the normalization processing on the target decision matrix, a target standard matrix is obtained, and the strategy evaluation module 205 is further specifically configured to:

[0225] Each evaluation target value in the target decision matrix is input into a membership function corresponding to the evaluation target, and a plurality of standard memberships are obtained.

[0226] The target standard matrix is constructed according to the plurality of standard memberships.

[0227] The multi-objective scheduling strategy selection system 200 considering source load uncertainty described in the present application can obtain target virtual power plant data, determine evaluation targets corresponding to the target virtual power plant data, obtain a plurality of evaluation targets, construct a multi-objective optimization scheduling model according to the plurality of evaluation targets, perform fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model, input the target virtual power plant data into the satisfaction model to obtain a plurality of scheduling strategies, comprehensively evaluate the plurality of scheduling strategies according to the plurality of evaluation targets to obtain a plurality of evaluation results, select a target evaluation result from the plurality of evaluation results, select a scheduling strategy corresponding to the target evaluation result, and obtain a target optimal scheduling strategy, thereby realizing selection of a multi-objective scheduling strategy considering economy, energy and environment comprehensively for a virtual power plant in the face of source load uncertainty.

[0228] Please refer to Figure 3 , Figure 3 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. The electronic device can include a processor, a memory, a communication interface and one or more programs. The processor, the memory and the communication interface can be connected to each other through a bus. The one or more programs are stored in the memory and configured to be executed by the processor. In the embodiment of the present application, the program includes instructions for performing the following steps:

[0229] Obtain target virtual power plant data.

[0230] Determine evaluation targets corresponding to the target virtual power plant data to obtain a plurality of evaluation targets.

[0231] Construct a multi-objective optimization scheduling model according to the plurality of evaluation targets.

[0232] Perform fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model.

[0233] Input the target virtual power plant data into the satisfaction model to obtain a plurality of scheduling strategies.

[0234] According to the multiple evaluation targets, the multiple scheduling strategies are comprehensively evaluated to obtain multiple evaluation results, one target evaluation result is selected from the multiple evaluation results, a scheduling strategy corresponding to the target evaluation result is selected, and a target optimal scheduling strategy is obtained.

[0235] The electronic device described in the application can obtain target virtual power plant data, determine evaluation targets corresponding to the target virtual power plant data to obtain multiple evaluation targets, construct a multi-target optimization scheduling model according to the multiple evaluation targets, perform fuzzy processing on the multi-target optimization scheduling model to obtain a satisfaction model, input the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies, comprehensively evaluate the multiple scheduling strategies according to the multiple evaluation targets to obtain multiple evaluation results, select one target evaluation result from the multiple evaluation results, select a scheduling strategy corresponding to the target evaluation result, and obtain a target optimal scheduling strategy, so that the virtual power plant can select a multi-target scheduling strategy that comprehensively considers economy, energy and environment when facing source-load uncertainty.

[0236] The embodiment of the application further provides a computer readable storage medium, wherein the computer readable storage medium stores a computer program for electronic data exchange, and the computer program causes a computer to execute part or all steps of any method described in the above method embodiments, and the computer includes the electronic device.

[0237] The embodiment of the application further provides a computer program product, and the computer program product includes a non-transitory computer readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute part or all steps of any method described in the above method embodiments. The computer program product can be a software installation package, and the computer includes the electronic device.

[0238] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be implemented by a computer program to instruct relevant hardware to complete, the program can be stored in a computer readable storage medium, and the program can include the processes of the above-mentioned method embodiments when executed. The storage medium includes a ROM or a random storage memory RAM, a magnetic disk or an optical disk and various program code storage media.

[0239] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in RAM, flash memory, ROM, EPROM, EEPROM, registers, hard disk, a removable media, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium can be integral to the processor. The processor and the storage medium can reside in an ASIC. Alternatively, the processor and the storage medium can be located in a terminal device or an access device. The processor and the storage medium can also be located in any other

[0240] Those skilled in the art should clearly understand that, in one or more examples described above, the functions described in the embodiments of the present application can be implemented entirely or partially by software, hardware, firmware, or any combination thereof. When implemented by software, the functions can be implemented in the form of a computer program product entirely or partially. The computer program product includes one or more computer instructions. When loaded and executed on a computer, the computer instructions entirely or partially generate the processes or functions described in the embodiments of the present application. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer instructions can be transferred from one website, computer, server, or data center to another via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible by a computer or a data storage device such as a server, data center, etc. that includes one or more available media sets. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., digital video disc (DVD)), or a semiconductor medium (e.g., solid state disk (SSD)), etc.

[0241] The various modules / units included in the various devices and products described in the above embodiments can be software modules / units or hardware modules / units, or partially software modules / units and partially hardware modules / units. For example, for the various devices and products applied to or integrated into a chip, the various modules / units included therein can all be implemented in the form of hardware such as circuitry, or at least some of the modules / units can be implemented in the form of software programs running on a processor integrated in the chip, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuitry; for the various devices and products applied to or integrated into a chip module, the various modules / units included therein can all be implemented in the form of hardware such as circuitry, and different modules / units can be located in the same component (e.g., a chip, a circuit module, etc.) or different components of the chip module, or at least some of the modules / units can be implemented in the form of software programs running on a processor integrated in the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuitry; for the various devices and products applied to or integrated into a terminal device, the various modules / units included therein can all be implemented in the form of hardware such as circuitry, and different modules / units can be located in the same component (e.g., a chip, a circuit module, etc.) or different components of the terminal device, or at least some of the modules / units can be implemented in the form of software programs running on a processor integrated in the terminal device, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuitry.

[0242] The above detailed description of the specific embodiments of the present application has been presented for the purposes of illustration and description. It is not intended to be exhaustive or to limit the application to the precise form described, and many modifications, equivalents and alternatives shown. The detailed description is not intended to limit the scope of the present application. Instead, the scope of the present application is defined by the appended claims, reasonably construed in light of the prior art.

Claims

1. A multi-objective scheduling strategy selection method considering source-load uncertainty, characterized by: The method comprises: Obtain target virtual power plant data; Determining evaluation targets corresponding to the target virtual power plant data to obtain multiple evaluation targets; Constructing a multi-objective optimization scheduling model according to the multiple evaluation objectives; Performing fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model; Inputting the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies; Comprehensively evaluating the multiple scheduling strategies according to the multiple evaluation targets to obtain multiple evaluation results, selecting a target evaluation result from the multiple evaluation results, and selecting a scheduling strategy corresponding to the target evaluation result to obtain a target optimal scheduling strategy; The multi-objective optimization scheduling model is constructed according to the multiple evaluation objectives, including: Obtaining an objective function corresponding to each of the multiple evaluation objectives to obtain multiple objective functions; Determining a constraint condition corresponding to each objective function according to the multiple objective functions to obtain multiple constraint conditions; Constructing the multi-objective optimization scheduling model according to the multiple objective functions and the multiple constraints; The fuzzy processing of the multi-objective optimization scheduling model to obtain a satisfaction model includes: Determining a membership function corresponding to each of the multiple objective functions to obtain multiple membership functions; Obtain weight coefficients corresponding to multiple evaluation targets to obtain multiple weight coefficients; Constructing the satisfaction model according to the multiple membership functions and the multiple weight coefficients; The step of determining the membership function corresponding to each of the multiple objective functions to obtain multiple membership functions includes: Obtaining a reference objective function; the reference objective function is any objective function among the multiple objective functions; Determining the maximum value and the minimum value of the reference objective function according to the target virtual power plant data; Inputting the maximum value and the minimum value into a preset fuzzy membership function to obtain a membership function corresponding to a reference objective function; The step of obtaining weight coefficients corresponding to multiple evaluation targets to obtain multiple weight coefficients includes: Performing objective weight analysis on the multiple evaluation targets based on an objective weighting method to obtain multiple objective weights; Determining a decision weight corresponding to each of the multiple evaluation targets based on the confident two-layer language preference relationship to obtain multiple decision weights; Multiplying the multiple objective weights and the multiple decision weights one by one to obtain multiple combined weights; Normalizing the multiple combined weights to obtain the multiple weight coefficients; The step of determining a decision weight corresponding to each of the multiple evaluation targets based on the confident two-layer language preference relationship to obtain multiple decision weights includes: Obtaining a preset two-level language term set; the preset two-level language term set includes a plurality of two-level language terms; Evaluate the multiple evaluation targets according to the preset two-level language term set to obtain multiple evaluation results; Determining a confidence level corresponding to each of the plurality of evaluation results to obtain a plurality of confidence levels; A weight analysis is performed on each of the multiple evaluation targets according to the multiple evaluation results and the multiple confidence levels to obtain multiple decision weights.

2. The method according to claim 1, wherein The target virtual power plant data is input into the satisfaction model to obtain multiple scheduling strategies, including: Inputting the target virtual power plant data into the multiple constraints to obtain multiple reference scheduling strategies; each reference scheduling strategy includes reference evaluation target values ​​of the multiple evaluation targets; Select any reference scheduling strategy from the multiple reference scheduling strategies to obtain a target reference scheduling strategy; Obtaining a reference evaluation target value for each evaluation target of the target reference scheduling strategy to obtain a plurality of reference evaluation target values; Inputting the multiple reference evaluation target values ​​into the membership function corresponding to each evaluation target to obtain multiple membership values; determining a target satisfaction level of the reference scheduling strategy according to the multiple membership values; When the target satisfaction is greater than a preset satisfaction threshold, determining that the target reference scheduling strategy is a feasible scheduling strategy; The multiple scheduling strategies are determined according to the feasible scheduling strategies.

3. The method according to claim 2, wherein The comprehensive evaluation of the multiple scheduling strategies according to the multiple evaluation objectives is performed to obtain multiple evaluation results, including: Constructing a target decision matrix according to the plurality of scheduling strategies; each scheduling strategy includes evaluation target values ​​of the plurality of evaluation targets; Normalizing the target decision matrix to obtain a target standard matrix; constructing a weighted norm matrix based on the plurality of weight coefficients and the target norm matrix; Determine an ideal solution for each evaluation target according to the weighted norm matrix to obtain a positive ideal solution and a negative ideal solution; the positive ideal solution is the maximum value among multiple evaluation target values ​​corresponding to the evaluation target; the negative ideal solution is the minimum value among multiple evaluation target values ​​corresponding to the evaluation target; Determine the distance between the evaluation target value corresponding to the evaluation target of each scheduling strategy in the multiple scheduling strategies and the positive ideal solution to obtain multiple first distances; Determine the distance between the evaluation target value corresponding to the evaluation target of each scheduling strategy in the multiple scheduling strategies and the negative ideal solution to obtain multiple second distances; An evaluation result of each of the multiple scheduling strategies is determined according to the multiple first distances and the multiple second distances to obtain multiple evaluation results.

4. The method according to claim 3, wherein The target decision matrix is ​​normalized to obtain a target standard matrix, including: Inputting each evaluation target value in the target decision matrix into the membership function corresponding to the evaluation target to obtain multiple standard memberships; The target standard matrix is ​​constructed according to the multiple standard memberships.

5. A multi-objective scheduling strategy selection system considering source-load uncertainty, characterized by: The system includes: a data acquisition module, a model building module, a model processing module, a strategy solving module, and a strategy evaluation module, wherein: The data acquisition module is used to acquire target virtual power plant data; The data acquisition module is further configured to determine an evaluation target corresponding to the target virtual power plant data, and obtain multiple evaluation targets; The model building module is used to build a multi-objective optimization scheduling model according to the multiple evaluation objectives; The model processing module is used to perform fuzzy processing on the multi-objective optimization scheduling model to obtain a satisfaction model; The strategy solving module is used to input the target virtual power plant data into the satisfaction model to obtain multiple scheduling strategies; The strategy evaluation module is used to comprehensively evaluate the multiple scheduling strategies according to the multiple evaluation targets to obtain multiple evaluation results, select a target evaluation result from the multiple evaluation results, select the scheduling strategy corresponding to the target evaluation result, and obtain the target optimal scheduling strategy; The multi-objective optimization scheduling model is constructed according to the multiple evaluation objectives, including: Obtaining an objective function corresponding to each of the multiple evaluation objectives to obtain multiple objective functions; Determining a constraint condition corresponding to each objective function according to the multiple objective functions to obtain multiple constraint conditions; Constructing the multi-objective optimization scheduling model according to the multiple objective functions and the multiple constraints; The fuzzy processing of the multi-objective optimization scheduling model to obtain a satisfaction model includes: Determining a membership function corresponding to each of the multiple objective functions to obtain multiple membership functions; Obtain weight coefficients corresponding to multiple evaluation targets to obtain multiple weight coefficients; Constructing the satisfaction model according to the multiple membership functions and the multiple weight coefficients; The step of determining the membership function corresponding to each of the multiple objective functions to obtain multiple membership functions includes: Obtaining a reference objective function; the reference objective function is any objective function among the multiple objective functions; Determining the maximum value and the minimum value of the reference objective function according to the target virtual power plant data; Inputting the maximum value and the minimum value into a preset fuzzy membership function to obtain a membership function corresponding to a reference objective function; The step of obtaining weight coefficients corresponding to multiple evaluation targets to obtain multiple weight coefficients includes: Performing objective weight analysis on the multiple evaluation targets based on an objective weighting method to obtain multiple objective weights; Determining a decision weight corresponding to each of the multiple evaluation targets based on the confident two-layer language preference relationship to obtain multiple decision weights; Multiplying the multiple objective weights and the multiple decision weights one by one to obtain multiple combined weights; Normalizing the multiple combined weights to obtain the multiple weight coefficients; The step of determining a decision weight corresponding to each of the multiple evaluation targets based on the confident two-layer language preference relationship to obtain multiple decision weights includes: Obtaining a preset two-level language term set; the preset two-level language term set includes a plurality of two-level language terms; Evaluate the multiple evaluation targets according to the preset two-level language term set to obtain multiple evaluation results; Determining a confidence level corresponding to each of the plurality of evaluation results to obtain a plurality of confidence levels; A weight analysis is performed on each of the multiple evaluation targets according to the multiple evaluation results and the multiple confidence levels to obtain multiple decision weights.

Images