Active power distribution network reconstruction scheme determination method considering demand response under uncertainty

Abstract

The invention relates to an active power distribution network reconstruction scheme determination method and device considering demand response under uncertainty, computer equipment, a computer readable storage medium and a computer program product, and can be applied to the technical field of electric power. The method comprises the following steps: acquiring historical operation data of a power generation unit and a load unit in a target power distribution network; the power generation unit comprises a wind power generation unit and a photovoltaic power generation unit; based on the historical operation data, constructing a multi-dimensional joint probability distribution model of the target power distribution network; the multi-dimensional joint probability distribution model represents an incidence relation between the power generation unit and the load unit; generating a typical scene set of the target power distribution network according to the multi-dimensional joint probability distribution model; based on the typical scene set, constructing a two-stage optimization model of the target power distribution network; and determining a reconstruction scheme of the target power distribution network according to the two-stage optimization model. By adopting the method, the accuracy of determining the reconstruction scheme of the power distribution network can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power, and in particular to a method and device for determining a reconstruction scheme of an active distribution network under uncertainty considering demand response, a computer device, a computer readable storage medium, and a computer program product. BACKGROUND

[0002] In the technical field of electric power systems, distribution network reconstruction is an important means to improve the reliability of distribution network operation, but with the access of new energy such as wind power and photovoltaic power, the operation scenarios of distribution networks are becoming increasingly complex.

[0003] In existing distribution network reconstruction schemes, the correlation between generation units and load units is often ignored, resulting in low accuracy in determining the reconstruction scheme of the distribution network. SUMMARY

[0004] Therefore, it is necessary to provide a method and device for determining a reconstruction scheme of an active distribution network under uncertainty considering demand response, which can improve the accuracy of determining the reconstruction scheme of the distribution network.

[0005] In a first aspect, the present application provides a method for determining a reconstruction scheme of an active distribution network under uncertainty considering demand response. The method comprises:

[0006] obtaining historical operation data of generation units and load units in a target distribution network; the generation units include wind power generation units and photovoltaic power generation units;

[0007] based on the historical operation data, constructing a multi-dimensional joint probability distribution model of the target distribution network; the multi-dimensional joint probability distribution model represents the correlation between the generation units and the load units;

[0008] generating a typical scenario set of the target distribution network according to the multi-dimensional joint probability distribution model;

[0009] based on the typical scenario set, constructing a two-stage optimization model of the target distribution network; the two-stage optimization model includes first-stage decision variables and second-stage decision variables; the first-stage decision variables are used to determine the topology structure and demand response capacity of the target distribution network; the second-stage decision variables are used to determine the control information of the generation units and the control information of the load units under each typical scenario in the typical scenario set;

[0010] determining the reconstruction scheme of the target distribution network according to the two-stage optimization model.

[0011] In one of the embodiments, the constructing the multi-dimensional joint probability distribution model of the target power distribution network based on the historical operation data comprises:

[0012] performing edge probability distribution construction processing on the power generation unit data and the load unit data in the historical operation data to obtain edge probability distribution information corresponding to each unit;

[0013] constructing the multi-dimensional joint probability distribution model based on the edge probability distribution information.

[0014] In one of the embodiments, the constructing the multi-dimensional joint probability distribution model based on the edge probability distribution information comprises:

[0015] constructing a tree structure model according to the edge probability distribution information; the tree structure model is composed of multiple tree structures;

[0016] determining corresponding coupling information for each edge in the tree structure model; the coupling information is used to represent a conditional dependence relationship;

[0017] determining the multi-dimensional joint probability distribution model according to the tree structure model and the coupling information.

[0018] In one of the embodiments, the generating the typical scenario set of the target power distribution network according to the multi-dimensional joint probability distribution model comprises:

[0019] generating an initial scenario set of the target power distribution network according to the multi-dimensional joint probability distribution model;

[0020] performing clustering processing on the initial scenario set to obtain the typical scenario set.

[0021] In one of the embodiments, the determining the reconstruction scheme of the target power distribution network according to the two-stage optimization model comprises:

[0022] performing model fusion processing on the two-stage optimization model to obtain a fusion model of the target power distribution network;

[0023] determining the reconstruction scheme of the target power distribution network according to the fusion model.

[0024] In one of the embodiments, after determining the reconstruction scheme of the target power distribution network according to the two-stage optimization model, the method further comprises:

[0025] performing prediction processing on the reconstruction scheme to obtain a prediction result of the reconstruction scheme;

[0026] The prediction result comprises at least one of an operation cost, a load shedding rate, and a new energy curtailment rate.

[0027] In a second aspect, the present application provides a device for determining a reconstruction scheme of an active power distribution network considering demand response under uncertainty. The device comprises:

[0028] a data acquisition module configured to acquire historical operation data of generation units and load units in a target power distribution network; the generation units comprise wind power generation units and photovoltaic power generation units;

[0029] a first construction module configured to construct a multi-dimensional joint probability distribution model of the target power distribution network based on the historical operation data; the multi-dimensional joint probability distribution model represents an association relationship between the generation units and the load units;

[0030] a scenario generation module configured to generate a typical scenario set of the target power distribution network according to the multi-dimensional joint probability distribution model;

[0031] a second construction module configured to construct a two-stage optimization model of the target power distribution network based on the typical scenario set; the two-stage optimization model comprises first-stage decision variables and second-stage decision variables; the first-stage decision variables are used to determine a topology structure and a demand response capacity of the target power distribution network; the second-stage decision variables are used to determine control information of the generation units and control information of the load units under each typical scenario in the typical scenario set;

[0032] a scheme determination module configured to determine a reconstruction scheme of the target power distribution network according to the two-stage optimization model.

[0033] In a third aspect, the present application provides a computer device. The computer device comprises a memory and a processor; the memory stores a computer program; and the processor implements the following steps when executing the computer program:

[0034] acquiring historical operation data of generation units and load units in a target power distribution network; the generation units comprise wind power generation units and photovoltaic power generation units;

[0035] constructing a multi-dimensional joint probability distribution model of the target power distribution network based on the historical operation data; the multi-dimensional joint probability distribution model represents an association relationship between the generation units and the load units;

[0036] generating a typical scenario set of the target power distribution network according to the multi-dimensional joint probability distribution model;

[0037] constructing a two-stage optimization model of the target power distribution network based on the typical scenario set; the two-stage optimization model includes first-stage decision variables and second-stage decision variables; the first-stage decision variables are used to determine a topology structure and a demand response capacity of the target power distribution network; the second-stage decision variables are used to determine control information of the power generation units and control information of the load units under each typical scenario in the typical scenario set;

[0038] determining a reconstruction scheme of the target power distribution network according to the two-stage optimization model.

[0039] In a fourth aspect, the present application also provides a computer readable storage medium. The computer readable storage medium has a computer program stored thereon, and the computer program, when executed by a processor, implements the following steps:

[0040] obtaining historical operation data of power generation units and load units in a target power distribution network; the power generation units include wind power generation units and photovoltaic power generation units;

[0041] constructing a multi-dimensional joint probability distribution model of the target power distribution network based on the historical operation data; the multi-dimensional joint probability distribution model represents an association relationship between the power generation units and the load units;

[0042] generating a typical scenario set of the target power distribution network according to the multi-dimensional joint probability distribution model;

[0043] constructing a two-stage optimization model of the target power distribution network based on the typical scenario set; the two-stage optimization model includes first-stage decision variables and second-stage decision variables; the first-stage decision variables are used to determine a topology structure and a demand response capacity of the target power distribution network; the second-stage decision variables are used to determine control information of the power generation units and control information of the load units under each typical scenario in the typical scenario set;

[0044] determining a reconstruction scheme of the target power distribution network according to the two-stage optimization model.

[0045] In a fifth aspect, the present application also provides a computer program product. The computer program product includes a computer program, and the computer program, when executed by a processor, implements the following steps:

[0046] obtaining historical operation data of power generation units and load units in a target power distribution network; the power generation units include wind power generation units and photovoltaic power generation units;

[0047] constructing a multi-dimensional joint probability distribution model of the target power distribution network based on the historical operation data; the multi-dimensional joint probability distribution model represents an association relationship between the power generation units and the load units;

[0048] generating a set of typical scenarios of the target power distribution network according to the multi-dimensional joint probability distribution model;

[0049] constructing a two-stage optimization model of the target power distribution network based on the set of typical scenarios; the two-stage optimization model comprises first-stage decision variables and second-stage decision variables; the first-stage decision variables are used to determine a topology structure and a demand response capacity of the target power distribution network; the second-stage decision variables are used to determine control information of the power generation units and control information of the load units under each typical scenario in the set of typical scenarios;

[0050] determining a reconstruction scheme of the target power distribution network according to the two-stage optimization model.

[0051] The method, device, computer equipment, computer readable storage medium and computer program product for determining a reconstruction scheme of an active power distribution network considering demand response under uncertainty, obtain historical operation data of generation units and load units in a target power distribution network; the generation units include wind power generation units and photovoltaic power generation units; based on the historical operation data, a multi-dimensional joint probability distribution model of the target power distribution network is constructed; the multi-dimensional joint probability distribution model represents the correlation between the generation units and the load units; according to the multi-dimensional joint probability distribution model, a typical scenario set of the target power distribution network is generated; based on the typical scenario set, a two-stage optimization model of the target power distribution network is constructed; the two-stage optimization model includes first-stage decision variables and second-stage decision variables; the first-stage decision variables are used to determine the topology structure and demand response capacity of the target power distribution network; the second-stage decision variables are used to determine the control information of the generation units and the control information of the load units under each typical scenario in the typical scenario set; and according to the two-stage optimization model, a reconstruction scheme of the target power distribution network is determined. This scheme can accurately represent the correlation between the wind power generation units, the photovoltaic power generation units and the load units by obtaining the historical operation data of the wind power generation units, the photovoltaic power generation units and the load units in the target power distribution network and constructing a multi-dimensional joint probability distribution model, which is beneficial to truly depict the time-space coupling characteristics of various uncertain factors in the target power distribution network, thereby avoiding the distortion problem of uncertainty description caused by treating various uncertain factors independently; the typical scenario set can effectively cover the main characteristics of the uncertainty space with a limited number of typical scenarios according to the multi-dimensional joint probability distribution model, which is beneficial to reasonably reflect the uncertainty of the wind power generation units, the photovoltaic power generation units and the load units in the optimization model; by constructing the two-stage optimization model and taking the topology structure and the demand response capacity as the first-stage decision variables and taking the generation unit control information and the load unit control information under each typical scenario as the second-stage decision variables, the coordinated optimization of the day-ahead topology reconstruction decision and the real-time operation control decision can be realized, which is beneficial to comprehensively consider the topology structure adjustment and the demand response resource allocation of the power distribution network under the influence of uncertainty, thereby being beneficial to improving the accuracy of determining the reconstruction scheme of the power distribution network. BRIEF DESCRIPTION OF DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments or the related art, the drawings needed to be used in the embodiments or the related art description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without any creative effort.

[0053] Figure 1This is a flowchart illustrating a method for determining an active distribution network reconfiguration scheme considering demand response under uncertainty in one embodiment.

[0054] Figure 2 This is a flowchart illustrating the steps involved in building a model in one embodiment;

[0055] Figure 3 This is a flowchart illustrating the steps for determining a model in one embodiment;

[0056] Figure 4 This is a structural block diagram of an active distribution network reconfiguration scheme determination device considering demand response under uncertainty in one embodiment;

[0057] Figure 5 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0059] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0060] In one exemplary embodiment, such as Figure 1 As shown, a method for determining an active distribution network reconfiguration scheme considering demand response under uncertainty is provided. This embodiment illustrates the application of this method to a terminal as an example. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and is implemented through interaction between the terminal and the server. The terminal can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc.; the server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. In this embodiment, the method includes the following steps:

[0061] Step S101: Obtain historical operating data of power generation units and load units in the target distribution network; power generation units include wind power generation units and photovoltaic power generation units;

[0062] Step S102: Based on historical operating data, construct a multidimensional joint probability distribution model of the target distribution network; the multidimensional joint probability distribution model characterizes the relationship between generation units and load units;

[0063] Step S103: Generate a set of typical scenarios for the target distribution network based on the multidimensional joint probability distribution model;

[0064] Step S104: Based on the typical scenario set, construct a two-stage optimization model for the target distribution network. The two-stage optimization model includes first-stage decision variables and second-stage decision variables. The first-stage decision variables are used to determine the topology and demand response capacity of the target distribution network. The second-stage decision variables are used to determine the control information of the generation unit and the control information of the load unit under each typical scenario in the typical scenario set.

[0065] Step S105: Determine the reconfiguration scheme of the target distribution network based on the two-stage optimization model.

[0066] The target distribution network can be an active distribution network to be optimized by topology reconfiguration, which is connected to distributed new energy power generation units such as wind power generation units and photovoltaic power generation units, as well as multiple load units.

[0067] Among them, the power generation unit can be the power supply equipment in the target distribution network, including new energy power generation equipment such as wind power generation units and photovoltaic power generation units, as well as conventional distributed power generation equipment such as distributed generators.

[0068] The load unit can be the electrical load node in the target distribution network, including the active load and reactive load at each node.

[0069] Historical operating data can include historical output data of wind power generation units, historical output data of photovoltaic power generation units, and historical load data of load units, which are used to establish probability distribution models for each variable.

[0070] Among them, the multidimensional joint probability distribution model can be a multidimensional joint distribution function describing the output of wind power generation units, photovoltaic power generation units and load units, constructed using the R-Vine Copula theory. This model can characterize the complex spatiotemporal coupling relationship and correlation between wind power generation units, photovoltaic power generation units and load units.

[0071] Among them, the typical scenario set can be a set of scenarios with a moderate number and each with a probability weight, obtained by generating a large number of initial scenarios from the multidimensional joint probability distribution model through Monte Carlo sampling and then reducing the number of scenarios using the mean clustering algorithm. This set is used to efficiently represent the uncertainty space in the optimization model.

[0072] The two-stage optimization model can be a stochastic optimization model, which includes a first-stage decision and a second-stage decision. The first stage makes decisions before the uncertainty is realized, and the second stage makes operational optimization decisions for each typical scenario after the uncertainty is realized.

[0073] The decision variables in the first stage can be decision parameters that need to be determined before the uncertainty is realized, including the switching status of each branch in the topology of the target distribution network and the demand response capacity agreements signed by each node in each time period.

[0074] The decision variables in the second stage can be the operational control parameters that need to be determined after the implementation of each typical scenario in the typical scenario set, including the active and reactive power output of the power generation unit, the actual load shedding amount of the load unit, and the actual demand response execution power.

[0075] The topology can be the network connection form of the target distribution network, which is determined by the switch status of each branch. A switch status of 1 indicates that the line switch is closed, and a switch status of 0 indicates that the line switch is open.

[0076] The demand response capacity can be the value of the demand response capacity agreement signed by each node with the grid operator a day before, representing the maximum load power that the node can reduce in real time.

[0077] The control information can be the operating control parameters of the power generation unit and the load unit under typical scenarios. For the power generation unit, it includes active power output and reactive power output. For the load unit, it includes the actual demand response power and load shedding amount.

[0078] The reconfiguration scheme can be the optimal distribution network operation strategy obtained by solving a two-stage optimization model, including the optimal topology of the target distribution network, the demand response capacity protocol of each node, and the power output of the generation unit and the control strategy of the load unit under each typical scenario.

[0079] Optionally, the terminal acquires historical operating data of wind power generation units, photovoltaic power generation units, and load units in the target distribution network. It then processes this historical operating data using kernel density estimation to establish marginal probability distributions for the output of wind power generation units, photovoltaic power generation units, and load units. Based on the vine structure dependency function theory, the marginal probability distributions of wind power generation units, photovoltaic power generation units, and load units are combined to construct a multidimensional joint probability distribution model characterizing the relationships between these units. This multidimensional joint probability distribution model can depict the spatiotemporal coupling and correlation between wind power generation units, photovoltaic power generation units, and load units. Finally, a massive number of initial scenarios are generated from the multidimensional joint probability distribution model using Monte Carlo sampling, and then... The mean clustering algorithm reduces the number of initial scenarios to obtain a set of typical scenarios with appropriate probability weights. A two-stage optimization model is constructed based on the typical scenario set. The first-stage decision variables of the two-stage optimization model include the switching status of each branch in the target distribution network and the demand response capacity protocol of each node. The second-stage decision variables include the active and reactive power output of the generation unit and the actual load shedding and actual demand response execution power of the load unit under each typical scenario. The two-stage optimization model is transformed into a mixed integer second-order cone programming model and solved to obtain the optimal topology of the target distribution network, the demand response capacity protocol of each node, and the control strategies of the generation unit and the load unit under each typical scenario, which serve as the reconfiguration scheme of the target distribution network.

[0080] In the aforementioned method for determining the reconfiguration scheme of an active distribution network considering demand response under uncertainty, historical operating data of generation units and load units in the target distribution network are obtained. Generation units include wind power generation units and photovoltaic power generation units. Based on the historical operating data, a multidimensional joint probability distribution model of the target distribution network is constructed. The multidimensional joint probability distribution model characterizes the correlation between generation units and load units. According to the multidimensional joint probability distribution model, a typical scenario set of the target distribution network is generated. Based on the typical scenario set, a two-stage optimization model of the target distribution network is constructed. The two-stage optimization model includes first-stage decision variables and second-stage decision variables. The first-stage decision variables are used to determine the topology and demand response capacity of the target distribution network. The second-stage decision variables are used to determine the control information of generation units and load units under each typical scenario in the typical scenario set. Based on the two-stage optimization model, the reconfiguration scheme of the target distribution network is determined. This scheme acquires historical operational data of wind power generation units, photovoltaic power generation units, and load units in the target distribution network and constructs a multidimensional joint probability distribution model. This accurately characterizes the relationships between these units, facilitating a realistic portrayal of the spatiotemporal coupling characteristics of various uncertainties in the target distribution network. This avoids the distortion of uncertainty description caused by treating each uncertainty independently. By generating a set of typical scenarios based on the multidimensional joint probability distribution model, the scheme effectively covers the main characteristics of the uncertainty space with a limited number of typical scenarios, allowing for a reasonable reflection of the uncertainties of wind power generation units, photovoltaic power generation units, and load units in the optimization model. Furthermore, by constructing a two-stage optimization model and using topology and demand response capacity as the first-stage decision variables, and power generation unit control information and load unit control information under each typical scenario as the second-stage decision variables, the scheme achieves coordinated optimization of day-ahead topology reconfiguration decisions and real-time operation control decisions. This facilitates a comprehensive consideration of distribution network topology adjustments and demand response resource allocation under the influence of uncertainties, thereby improving the accuracy of determining the distribution network reconfiguration scheme.

[0081] In one exemplary embodiment, reference is made to Figure 2 Based on historical operational data, a multidimensional joint probability distribution model of the target distribution network is constructed, including:

[0082] Step S201: Perform edge probability distribution construction processing on the power generation unit data and load unit data in the historical operation data to obtain the edge probability distribution information corresponding to each unit;

[0083] Step S202: Construct a multidimensional joint probability distribution model based on the marginal probability distribution information.

[0084] Among them, the power generation unit data can be the historical output data of wind power generation units and the historical output data of photovoltaic power generation units from the historical operation data.

[0085] Among them, the load unit data can be the historical active load data and historical reactive load data of the load unit in the historical operation data.

[0086] The marginal probability distribution construction process can be a process of fitting the probability density function of historical running data using the kernel density estimation method. By setting parameters such as kernel function, bandwidth parameter and number of historical samples, the marginal probability distribution of each variable is modeled.

[0087] Among them, the edge probability distribution information can be the edge probability distribution function of the wind power generation unit, the edge probability distribution function of the photovoltaic power generation unit, and the edge probability distribution function of the load unit, which are obtained by edge probability distribution construction and processing.

[0088] Optionally, the terminal performs edge probability distribution construction processing on the historical output data of wind power generation units, historical output data of photovoltaic power generation units, and historical load data of load units in the historical operation data. The kernel density estimation method is used to establish the edge probability distribution functions of wind power generation units, photovoltaic power generation units, and load units respectively, so as to obtain the edge probability distribution information of wind power generation units, photovoltaic power generation units, and load units. Based on the edge probability distribution information of wind power generation units, photovoltaic power generation units, and load units, a multidimensional joint probability distribution model is constructed using the vine structure dependency function theory. This multidimensional joint probability distribution model describes the multidimensional joint probability distribution relationship between wind power generation units, photovoltaic power generation units, and load units by combining the edge probability distribution information of each unit.

[0089] The technical solution provided in this embodiment, by constructing edge probability distributions of power generation unit data and load unit data in historical operation data, can accurately fit the probability distribution characteristics of wind power generation units, photovoltaic power generation units, and load units, which is beneficial for providing accurate edge probability distribution information for constructing a multidimensional joint probability distribution model. By constructing a multidimensional joint probability distribution model based on edge probability distribution information, the edge probability distribution information of wind power generation units, photovoltaic power generation units, and load units can be effectively combined, which is beneficial for accurately depicting the correlation between power generation units and load units.

[0090] In one exemplary embodiment, reference is made to Figure 3 Based on marginal probability distribution information, a multidimensional joint probability distribution model is constructed, including:

[0091] Step S301: Construct a tree structure model based on the edge probability distribution information; the tree structure model consists of multiple tree structures.

[0092] Step S302: For each edge in the tree structure model, determine the corresponding coupling information; the coupling information is used to characterize the conditional dependencies.

[0093] Step S303: Determine the multidimensional joint probability distribution model based on the tree structure model and coupling information.

[0094] Among them, the tree structure model can be a model composed of a series of tree structures built based on the vine structure dependency function theory, used to describe the multidimensional relationship structure between wind power generation units, photovoltaic power generation units and load units.

[0095] Among them, the tree structure can be the tree-like connection structure at each level in the tree structure model. The tree structure model represents the dependency relationship between multiple variables through the combination of multiple tree structures.

[0096] The coupling information can be a binary dependency function selected for each edge in the tree structure model, which describes the conditional dependency structure between two variables under given conditions.

[0097] Among them, conditional dependency can be the correlation between any two variables in the wind power generation unit, photovoltaic power generation unit, and load unit given a set of conditional variables.

[0098] Optionally, the terminal constructs a tree structure model composed of a series of tree structures using a heuristic algorithm based on the edge probability distribution information of wind power generation units, photovoltaic power generation units, and load units. The number of tree structures in the tree structure model is determined by the number of variables in the wind power generation units, photovoltaic power generation units, and load units. For each edge in the tree structure model, a suitable binary dependency function is selected as the corresponding coupling information. This coupling information is used to describe the conditional dependency relationship between two variables under a given set of conditional variables. Based on the tree structure model and the coupling information corresponding to each edge in the tree structure model, the edge probability distribution information of wind power generation units, photovoltaic power generation units, and load units are combined through coupling information to form a multidimensional joint probability distribution model.

[0099] The technical solution provided in this embodiment can establish a tree-like connection structure that represents the dependency relationship between multiple variables by constructing a tree structure model based on the edge probability distribution information. This is beneficial for systematically describing the association pattern between wind power generation units, photovoltaic power generation units and load units. By determining the corresponding coupling information for each edge in the tree structure model, the conditional dependency relationship between each variable can be accurately quantified. This is beneficial for effectively combining the edge probability distribution information to form a multidimensional joint probability distribution model.

[0100] In an exemplary embodiment, generating a typical scenario set of the target distribution network according to a multidimensional joint probability distribution model includes: generating an initial scenario set of the target distribution network according to the multidimensional joint probability distribution model; and performing clustering processing on the initial scenario set to obtain a typical scenario set.

[0101] The initial scenario set can be a collection of a large number of correlated wind power generation unit output scenarios, photovoltaic power generation unit output scenarios, and load unit power scenarios generated by Monte Carlo sampling technique based on a multidimensional joint probability distribution model.

[0102] The clustering process can be a process of reducing the number of scenes in the initial scene set by using the mean clustering algorithm. This involves setting a target number of scenes to be reduced, selecting initial cluster center scenes, calculating the distance from each scene to the cluster center scene, and iterating repeatedly until the changes in the cluster center scenes meet the convergence condition, thereby obtaining a scene set of appropriate size.

[0103] Optionally, the terminal generates a large number of correlated wind power generation unit output scenarios, photovoltaic power generation unit output scenarios, and load unit power scenarios as the initial scenario set of the target distribution network based on the multidimensional joint probability distribution model and Monte Carlo sampling technology. This initial scenario set can reflect the spatiotemporal coupling relationship between wind power generation units, photovoltaic power generation units, and load units. The initial scenario set is then subjected to mean clustering. This process involves setting the target number of clusters, selecting the initial cluster center scenario, calculating the distance from each scenario to the initial center scenario, clustering according to the nearest distance principle, obtaining the center of each cluster as the new cluster center scenario, and iterating repeatedly until convergence. This yields a typical scenario set with a moderate number of scenarios and each scenario having a probability weight.

[0104] The technical solution provided in this embodiment generates an initial scenario set based on a multidimensional joint probability distribution model, which can produce a large number of scenario samples that retain the correlation between wind power generation units, photovoltaic power generation units and load units, thus facilitating a comprehensive characterization of the uncertainty space of the target distribution network. By performing clustering processing on the initial scenario set, the number of scenarios can be effectively reduced while retaining the main characteristics of uncertainty, which is beneficial to improving the solution efficiency of the subsequent optimization model.

[0105] In an exemplary embodiment, determining the reconfiguration scheme of the target distribution network based on a two-stage optimization model includes: performing model fusion processing on the two-stage optimization model to obtain a fused model of the target distribution network; and determining the reconfiguration scheme of the target distribution network based on the fused model.

[0106] Among them, model fusion processing can be used to transform the two-stage optimization model into a mixed-integer second-order cone programming model through second-order cone relaxation techniques, and introduce the Big M method to handle the transformation process of discrete variables.

[0107] The fusion model can be a mixed integer second-order cone programming model obtained through model fusion processing. This model can be directly solved by the solver to obtain the reconfiguration scheme of the target distribution network.

[0108] The reconfiguration scheme can be a comprehensive optimization scheme that includes the topology of the target distribution network, demand response capacity protocol, generation unit control information, and load unit control information.

[0109] Optionally, the terminal performs model fusion processing on the two-stage optimization model, using the second-order cone relaxation technique to transform the nonlinear constraints in the two-stage optimization model into the form of second-order cone constraints, and introducing the Big M method to process discrete variables such as switch states in the two-stage optimization model, to obtain a fused model of the target distribution network. This fused model is a mixed-integer second-order cone programming model. Based on the fused model, the mixed-integer second-order cone programming solver is called to solve the fused model, obtaining a reconstruction scheme that includes the topology of the target distribution network, demand response capacity protocol, generation unit control information, and load unit control information.

[0110] The technical solution provided in this embodiment can transform the two-stage stochastic optimization model, which is originally difficult to solve directly, into a standard mixed integer second-order cone programming model by performing model fusion processing on the two-stage optimization model, which is conducive to efficient solution using a solver; by determining the reconfiguration scheme based on the fusion model, a comprehensive optimization decision including topology and demand response capacity can be obtained, which is conducive to achieving efficient operation of the target distribution network.

[0111] In an exemplary embodiment, after determining the reconfiguration scheme of the target distribution network according to the two-stage optimization model, the method further includes: performing prediction processing on the reconfiguration scheme to obtain the prediction result of the reconfiguration scheme; wherein the prediction result includes at least one of operating cost, load shedding rate and renewable energy curtailment rate.

[0112] Among them, the predictive processing can be the process of evaluating and analyzing the operational effect of the target distribution network after the reconfiguration scheme is implemented. The implementation effect of the reconfiguration scheme is predicted by calculating various operational indicators of the target distribution network after the reconfiguration scheme is implemented.

[0113] The prediction results can be evaluation index information reflecting the operating effect of the target distribution network obtained through prediction processing, including at least one of operating costs, load shedding rate and renewable energy curtailment rate.

[0114] Among them, the operating cost can be the total system operating cost of the target distribution network after the implementation of the reconfiguration scheme, including the demand response capacity agreement cost and the weighted sum of the operating costs under each typical scenario.

[0115] The load shedding rate can be the weighted sum of the ratio of the load shedding amount of each load unit to the total load of each load unit under each typical scenario after the implementation of the reconfiguration scheme, and is used to measure the power supply reliability of the target distribution network.

[0116] Among them, the renewable energy curtailment rate can be the weighted sum of the ratio of renewable energy curtailment to the predicted renewable energy generation in each typical scenario after the implementation of the relocation plan, which is used to measure the renewable energy absorption capacity of the target distribution network.

[0117] Optionally, the terminal performs predictive processing on the reconfiguration scheme, calculates the total system operating cost of the target distribution network after implementing the reconfiguration scheme as the operating cost, calculates the load shedding rate based on the load shedding amount and total load under each typical scenario in the reconfiguration scheme, and calculates the renewable energy curtailment rate based on the renewable energy curtailment amount and the predicted renewable energy generation under each typical scenario in the reconfiguration scheme, and obtains the prediction results including operating cost, load shedding rate and renewable energy curtailment rate. The prediction results can comprehensively evaluate the implementation effect of the reconfiguration scheme from multiple dimensions such as economy, reliability and renewable energy absorption capacity.

[0118] The technical solution provided in this embodiment can calculate multiple evaluation indicators such as operating cost, load shedding rate and renewable energy curtailment rate after the reconfiguration scheme is implemented by predicting the reconfiguration scheme. This is conducive to quantitatively evaluating the implementation effect of the reconfiguration scheme from multiple dimensions such as economy, power supply reliability and renewable energy absorption capacity, and thus facilitates comprehensive analysis of the actual performance of the reconfiguration scheme in the target distribution network operation.

[0119] The following example illustrates the method for determining an active distribution network reconfiguration scheme considering demand response under uncertainty provided in this application. This example demonstrates the application of this method to a terminal.

[0120] Driven by the strategic goals of "carbon peaking and carbon neutrality," the penetration rate of renewable energy, represented by wind power and photovoltaics, in distribution networks is continuously increasing. However, both wind and solar power output and load demand exhibit strong uncertainty and volatility, making the operating conditions of distribution networks increasingly complex and posing significant challenges to traditional deterministic planning and operation methods. Existing research on distribution network reconfiguration is mostly based on deterministic scenarios, or while considering uncertainty, it fails to achieve unified and coordinated optimization of network reconfiguration and demand response. This results in insufficient system operating economics, limited renewable energy absorption capacity, and difficulty in effectively addressing the security risks arising from the coupling of multiple uncertainties. Therefore, there is an urgent need for an optimization method that can comprehensively consider source-load uncertainties and achieve coordinated decision-making between reconfiguration and demand response to improve the robustness, economy, and power supply reliability of active distribution networks.

[0121] To address the aforementioned issues, this embodiment provides an active distribution network reconfiguration optimization method considering demand response under uncertainty. First, based on R-Vine Copula theory, high-dimensional correlation modeling is performed on historical data of wind power, photovoltaic output, and load. A massive number of coupled scenarios are generated through Monte Carlo sampling, and K-means clustering (an unsupervised clustering algorithm) is used to reduce the number of scenarios, resulting in a set of typical scenarios. Then, a two-stage stochastic optimization model is established. The first stage determines the distribution network topology and day-ahead demand response capacity agreement; the second stage optimizes generator output and load control strategies under each typical scenario. Subsequently, the model is transformed into a mixed-integer second-order cone programming model. DistFlow (a mathematical model describing the power flow of distribution networks with distributed generation) power flow equations are used, and the Big-M method is introduced to handle discrete variables, solving for the optimal reconfiguration strategy and operating scheme. Finally, through multiple evaluation indicators such as system operating cost, expected load shedding rate, and expected renewable energy curtailment rate, the performance of the reconfiguration scheme in terms of economy, reliability, and power quality is comprehensively analyzed, thereby improving the operating performance of the active distribution network.

[0122] The technical solution is as follows:

[0123] Includes the following steps:

[0124] Step 1: Input historical data on wind power, solar power output and load, establish the marginal probability distribution of each variable using the kernel density estimation method, and construct the multidimensional joint probability distribution between wind power, solar power and load based on R-Vine Copula theory;

[0125] Step 2: Based on the multidimensional joint probability distribution of wind and solar load obtained in Step 1, a large number of coupled scenes are generated by Monte Carlo sampling, and the K-means clustering algorithm is used to reduce the number of scenes to obtain a typical scene set.

[0126] Step 3: Based on the typical scenario set obtained in Step 2, establish a two-stage stochastic optimization model. The first stage is to decide the distribution network topology and day-ahead demand response capacity agreement. The second stage is to optimize generator output and load control strategies under each typical scenario.

[0127] Step 4: Transform the two-stage stochastic optimization model established in Step 3 into a mixed-integer second-order cone programming model, and introduce the Big-M method to process discrete variables to obtain the optimal reconstruction strategy and running scheme;

[0128] Step 5: Based on the reconfiguration strategy and operation plan obtained in Step 4, establish load shedding rate and renewable energy curtailment rate indicators, construct a comprehensive benefit evaluation model, and quantitatively analyze the performance of the reconfiguration plan in terms of economy, reliability and power quality.

[0129] Step 1 is as follows:

[0130] 1) Establishing marginal probability distributions: Using a nonparametric kernel density estimation method, wind power output is fitted based on historical data. Photovoltaic power output and load The marginal probability distribution for any random variable. (represent , or Its probability density function The estimation formula is:

[0131] ,

[0132] in, For the number of historical samples, For kernel function, For bandwidth parameters, Historical observation data for variables.

[0133] 2) Constructing the R-Vine Copula Joint Distribution: To characterize the complex spatiotemporal coupling relationship between wind power, photovoltaics, and loads, an R-Vine Copula joint distribution model is constructed based on the marginal probability distribution obtained above. A heuristic algorithm is used to select and construct a model consisting of a series of tree nodes. The resulting R-Vine tree structure is formed, where d is the number of variables; for each edge e in the above R-Vine tree structure, a suitable binary Copula function is selected. Used to describe a given set of condition variables Below, variables and The conditional dependency structure between them; ultimately, wind power output Photovoltaic power output and load joint distribution function It is given by the following formula:

[0134] ,

[0135] Where c is a multivariate Copula function determined by the aforementioned R-Vine tree structure. , , These are the marginal cumulative distribution functions for wind power, photovoltaic, and load variables, respectively.

[0136] Step 2 is as follows:

[0137] Based on the established R-Vine Copula joint distribution, a large number of correlated wind power-solar power-load multidimensional initial scenarios were generated using Monte Carlo sampling. Subsequently, the K-means clustering algorithm was used to reduce this initial scenario set, resulting in a set of typical scenarios of moderate size, each with its own probability weight, used to efficiently represent the space of uncertainty in the optimization model. The K-means clustering steps are as follows:

[0138] 1) Set the target number of scenarios to be reduced, k, as the target number of K-means clusters.

[0139] 2) Select k output scenarios As the initial cluster center point (initial cluster center scenario), calculate the distance from each scenario to the initial center scenario:

[0140] .

[0141] In the formula, For the scene To the central scene The distance, T is the data dimension of the scene.

[0142] 3) After calculating the distance from each scene to the central scene, perform the first clustering according to the nearest central scene to obtain the initial clustering results.

[0143] 4) Based on the clustering results, the center of each cluster is obtained as the new cluster center scenario.

[0144] 5) Iterate repeatedly until the changes in the cluster center scenario meet the convergence condition, and obtain the final cluster center scenario as the typical power output scenario of wind and solar power and the typical load scenario of the distribution network.

[0145] Step 3 specifically involves:

[0146] The two-stage stochastic optimization model is as follows:

[0147] Objective function:

[0148] ,

[0149] ,

[0150] ,

[0151] ,

[0152] ,

[0153] ,

[0154] ,

[0155] in, Cost of demand response capacity agreement, Let be the probability of the s-th scenario. For a set of nodes, For scene collection, For new energy power generation collection, For distributed generation collection, It represents the capacity of the demand response agreement signed by node i during time period t. Let be the unit demand response cost of the i-th node. The system operating cost in scenario s, The actual execution cost of demand response under scenario s. Let i be the unit demand response execution cost. The cost of curtailment penalties for renewable energy in scenario s. The cost of load shedding in scenario s. The voltage deviation penalty cost in scenario s, The cost of distributed power generation in scenario s. The cost of purchasing electricity for a substation in scenario s. The time interval for execution. Let m be the actual output of the m-th renewable energy power generation in scenario s during time period t. For the predicted active power output of the m-th new energy source in time period t under scenario s, The cost of penalties for abandoning renewable energy per unit of electricity. Let be the load shedding amount of the i-th node in scenario s during time period t. The unit load shedding cost for node i, Let g be the active power output of the g-th distributed generator in scenario s during time period t. Let g be the power generation cost of the g-th distributed generation unit in scenario s. This refers to the electricity purchased by the substation in scenario s during time period t. The unit electricity purchase cost of the substation during time period t.

[0156] The constraints are as follows:

[0157] 1) Topological and radial constraints:

[0158] ,

[0159] ,

[0160] in, For the virtual load of node i, For the number of nodes, The virtual flow passing through branch ij. This indicates that branch ij belongs to the subtree rooted at node i. The switch status of branch ij is 0 if the switch is open, and 1 if the switch is closed. The number of root nodes, A collection of branches in a power distribution network;

[0161] The purpose of virtual loads is to ensure that all load nodes satisfy connectivity constraints during the reconfiguration process. If a load node is isolated, its virtual power balance constraint is not satisfied. To satisfy the node virtual power balance constraint, all load nodes must remain connected, and the number of connected network lines in the distribution network must equal... .

[0162] 2) Demand response capacity constraints:

[0163] ,

[0164] ,

[0165] in, The actual required response execution power of node i in scenario s during time period t. It represents the capacity of the demand response agreement signed by node i during time period t. It represents the maximum capacity of demand response protocols that node i can sign during time period t.

[0166] 3) Current constraints:

[0167] ,

[0168] ,

[0169] ,

[0170] ,

[0171] ,

[0172] ,

[0173] ,

[0174] in, and These are subsets of distributed generation and new energy generation connected to node j, respectively. This indicates that branch jk belongs to the subtree rooted at node j. and For the net outflow of active and reactive power at node j in time period t under scenario s, and This represents the active and reactive power flow of branch ij in scenario s during time period t. and Let these represent the resistance and reactance of branch ij, respectively. Let M be the voltage amplitude of node i in scenario s during time period t, where M is a very large number. It is an intermediate variable.

[0175] 4) Load shedding constraint:

[0176] ,

[0177] ,

[0178] ,

[0179] in, Let the reactive power of node j in time period t be the switching power in scenario s. Let be the reactive power of node j in time period t under scenario s.

[0180] 5) Voltage constraint:

[0181] ,

[0182] in, and These are the lower and upper limits of the allowed node i voltage in the system.

[0183] 6) Distributed power sources:

[0184] ,

[0185] ,

[0186] in, and These are the lower and upper limits of the active power output of the g-th distributed power source, respectively. and These are the lower and upper limits of the reactive power output of the g-th distributed power source, respectively.

[0187] 7) Power output from new energy sources:

[0188] .

[0189] 8) Branch capacity constraints:

[0190] .

[0191] in, This represents the maximum capacity of the line.

[0192] Step 4 specifically involves:

[0193] 1) Order ,

[0194] in, Let be the current amplitude of branch ij during time period t in scenario s.

[0195] The voltage drop formula can then be transformed into:

[0196] .

[0197] 2) Second-order cone relaxation constraint:

[0198] .

[0199] Step 5 specifically involves:

[0200] The specific indicators for load shedding rate and renewable energy curtailment rate are as follows:

[0201] ,

[0202] ,

[0203] in, For load shedding rate, This refers to the curtailment rate of renewable energy.

[0204] The technical solution provided in this embodiment adopts a multi-dimensional uncertainty scenario generation method based on R-Vine Copula theory, which can accurately characterize the complex spatiotemporal coupling relationship between wind power, photovoltaic power output and load, and provide a more realistic uncertainty representation for optimization decision-making. It proposes a two-stage stochastic optimization framework that is highly compatible with the day-ahead demand response mechanism, organically separating topology decision-making from demand response capacity agreement (first stage) and multi-scenario operation optimization (second stage), realizing collaborative decision-making for reconfiguration and demand response under uncertain conditions.

[0205] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0206] Based on the same inventive concept, this application also provides an apparatus for determining an active distribution network reconfiguration scheme considering demand response under uncertainty, which is used to implement the aforementioned method for determining an active distribution network reconfiguration scheme considering demand response under uncertainty. The solution provided by this apparatus is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the apparatus for determining an active distribution network reconfiguration scheme considering demand response under uncertainty provided below can be found in the limitations of the method for determining an active distribution network reconfiguration scheme considering demand response under uncertainty described above, and will not be repeated here.

[0207] In one exemplary embodiment, such as Figure 4 As shown, an active distribution network reconfiguration scheme determination device considering demand response under uncertainty is provided. This active distribution network reconfiguration scheme determination device 400 may include:

[0208] Data acquisition module 401 is used to acquire historical operating data of power generation units and load units in the target distribution network; power generation units include wind power generation units and photovoltaic power generation units;

[0209] The first construction module 402 is used to construct a multidimensional joint probability distribution model of the target distribution network based on historical operating data; the multidimensional joint probability distribution model characterizes the relationship between generation units and load units;

[0210] The scenario generation module 403 is used to generate a typical scenario set of the target distribution network based on the multidimensional joint probability distribution model.

[0211] The second construction module 404 is used to construct a two-stage optimization model of the target distribution network based on a set of typical scenarios. The two-stage optimization model includes first-stage decision variables and second-stage decision variables. The first-stage decision variables are used to determine the topology and demand response capacity of the target distribution network. The second-stage decision variables are used to determine the control information of the generation unit and the control information of the load unit under each typical scenario in the set of typical scenarios.

[0212] The scheme determination module 405 is used to determine the reconfiguration scheme of the target distribution network based on the two-stage optimization model.

[0213] In an exemplary embodiment, the first construction module 402 is further configured to perform edge probability distribution construction processing on the power generation unit data and load unit data in the historical operation data to obtain the edge probability distribution information corresponding to each unit; and construct a multidimensional joint probability distribution model based on the edge probability distribution information.

[0214] In an exemplary embodiment, the first construction module 402 is further configured to construct a tree structure model based on edge probability distribution information; the tree structure model consists of multiple tree structures; for each edge in the tree structure model, the corresponding coupling information is determined; the coupling information is used to characterize conditional dependencies; and a multidimensional joint probability distribution model is determined based on the tree structure model and the coupling information.

[0215] In an exemplary embodiment, the scene generation module 403 is further configured to generate an initial scene set of the target distribution network based on a multidimensional joint probability distribution model; and to perform clustering processing on the initial scene set to obtain a typical scene set.

[0216] In an exemplary embodiment, the scheme determination module 405 is further configured to perform model fusion processing on the two-stage optimization model to obtain a fused model of the target distribution network; and determine the reconfiguration scheme of the target distribution network based on the fused model.

[0217] In an exemplary embodiment, the device 400 further includes: a scheme prediction module, used to perform prediction processing on the reconfiguration scheme to obtain the prediction result of the reconfiguration scheme; wherein the prediction result includes at least one of operating cost, load shedding rate and renewable energy curtailment rate.

[0218] The various modules in the active distribution network reconfiguration scheme determination device considering demand response under the aforementioned uncertainties can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0219] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 5 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When executed by the processor, the computer program implements a method for determining an active power distribution network reconfiguration scheme considering demand response under uncertainty. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0220] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0221] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0222] In one exemplary embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above-described method embodiments.

[0223] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0224] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0225] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0226] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for determining an active distribution network reconfiguration scheme considering demand response under uncertainty, characterized in that, The method includes: Acquire historical operating data of power generation units and load units in the target distribution network; the power generation units include wind power generation units and photovoltaic power generation units. Based on the historical operating data, a multidimensional joint probability distribution model of the target distribution network is constructed; the multidimensional joint probability distribution model characterizes the correlation between the power generation unit and the load unit. Based on the multidimensional joint probability distribution model, a typical scenario set for the target distribution network is generated; Based on the typical scenario set, a two-stage optimization model for the target distribution network is constructed. The two-stage optimization model includes first-stage decision variables and second-stage decision variables. The first-stage decision variables are used to determine the topology and demand response capacity of the target distribution network. The second-stage decision variables are used to determine the control information of the generation unit and the control information of the load unit under each typical scenario in the typical scenario set. Based on the two-stage optimization model, the reconfiguration scheme of the target distribution network is determined.

2. The method according to claim 1, characterized in that, The construction of a multidimensional joint probability distribution model for the target distribution network based on the historical operating data includes: The edge probability distribution construction process is performed on the power generation unit data and load unit data in the historical operation data to obtain the edge probability distribution information corresponding to each unit; Based on the marginal probability distribution information, the multidimensional joint probability distribution model is constructed.

3. The method according to claim 2, characterized in that, The construction of the multidimensional joint probability distribution model based on the marginal probability distribution information includes: Based on the edge probability distribution information, a tree structure model is constructed; the tree structure model consists of multiple tree structures. For each edge in the tree structure model, the corresponding coupling information is determined; the coupling information is used to characterize conditional dependencies. Based on the tree structure model and the coupling information, the multidimensional joint probability distribution model is determined.

4. The method according to claim 1, characterized in that, The step of generating a typical scenario set for the target distribution network based on the multidimensional joint probability distribution model includes: Based on the multidimensional joint probability distribution model, an initial scenario set for the target distribution network is generated; The initial scene set is clustered to obtain the typical scene set.

5. The method according to claim 1, characterized in that, The step of determining the reconfiguration scheme of the target distribution network based on the two-stage optimization model includes: The two-stage optimization model is fused to obtain the fused model of the target distribution network. Based on the fusion model, the reconfiguration scheme of the target distribution network is determined.

6. The method according to any one of claims 1 to 5, characterized in that, After determining the reconfiguration scheme of the target distribution network based on the two-stage optimization model, the process further includes: The reconstruction scheme is subjected to prediction processing to obtain the prediction result of the reconstruction scheme; The prediction results include at least one of operating costs, load shedding rate, and renewable energy curtailment rate.

7. A device for determining an active distribution network reconfiguration scheme considering demand response under uncertainty, characterized in that, The device includes: The data acquisition module is used to acquire historical operating data of power generation units and load units in the target distribution network; the power generation units include wind power generation units and photovoltaic power generation units; The first construction module is used to construct a multidimensional joint probability distribution model of the target distribution network based on the historical operating data; the multidimensional joint probability distribution model characterizes the correlation between the power generation unit and the load unit; The scenario generation module is used to generate a typical scenario set of the target distribution network based on the multidimensional joint probability distribution model. The second construction module is used to construct a two-stage optimization model of the target distribution network based on the typical scenario set. The two-stage optimization model includes a first-stage decision variable and a second-stage decision variable. The first-stage decision variable is used to determine the topology and demand response capacity of the target distribution network. The second-stage decision variable is used to determine the control information of the generation unit and the control information of the load unit under each typical scenario in the typical scenario set. The scheme determination module is used to determine the reconfiguration scheme of the target distribution network based on the two-stage optimization model.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images