Cluster-distribution network flexibility evaluation method and system considering communication attacks

Abstract

The present invention discloses a cluster-distribution network flexibility assessment method and system considering communication attacks, which belongs to the field of distribution network flexibility assessment technology and establishes a two-layer flexibility assessment architecture for clusters and distribution networks. At the cluster level, a false information injection attack modeling method based on robust optimization and a photovoltaic uncertainty modeling method based on distributed robust chance constraints are proposed, and a cluster robust-random flexibility assessment model is established. At the distribution network level, a distribution network random flexibility assessment model is established, including a distribution network flexibility range assessment model based on a simplified DistFlow equation and a distribution network baseline operating power and backup flexibility configuration model that introduces a demand ratio coefficient and a contribution coefficient. The present invention takes into account the impact of communication attacks and photovoltaic uncertainties on the flexibility of the distribution network, and further evaluates the upward and downward adjustment capabilities, baseline operating power and backup flexibility on the basis of the flexibility range, thereby achieving a comprehensive and integrated assessment of the flexibility of the distribution network.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network flexibility assessment, and in particular relates to a cluster-distribution network flexibility assessment method and system considering communication attacks. Background Art

[0002] The statements herein merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] How to explore and utilize the regulation potential of distributed resources in the distribution network to support the safe operation of the upper main network has been a problem that has attracted much attention in recent years. Due to the large number of distributed resources and their scattered layout, it is difficult for the upper distribution network to clearly understand the details of the operation and regulation of the underlying equipment. Existing studies mostly use aggregation methods to construct distributed resource clusters, and achieve flexibility aggregation and evaluation through a hierarchical partitioning architecture. However, malicious attacks in the power industry have been common in recent years, threatening the information security of the power system. Compared with the dedicated communication network used in large power grids, the existing communication network on the user side of distributed resource access has disadvantages such as poor stability and open information nodes without protection, making it more vulnerable to network attacks. In addition, there are many nodes in the distribution network that have highly random power injection, such as distributed photovoltaic power stations and rooftop photovoltaics scattered on the user side. These random disturbances are superimposed on possible malicious attack events, making the problem of distribution network flexibility evaluation very complicated. In this context, the inventors found that the current flexibility evaluation method mainly has the following two problems:

[0004] First, current research rarely considers the impact of communication attacks. However, communications attacks that maliciously tamper with key parameter information of distributed resources can affect the accuracy of flexibility assessments. For example, if the state of charge data of an energy storage device, which is originally close to the upper limit, is tampered with during communication and adjusted downward, the upper limit of the flexibility range will be incorrectly expanded. Grid dispatchers, receiving erroneous data, will overestimate the regulation potential and formulate clearly unrealistic dispatch control plans.

[0005] Secondly, in actual dispatch, distribution network operators must first determine the baseline operating power as the normal operating power of the distribution network; secondly, they must ensure that the ratio of upward and downward adjustment capabilities is as consistent as possible with power system requirements; finally, they must retain local reserve flexibility to address the safety risks brought about by photovoltaic uncertainty and clearly define the contribution of each distributed resource cluster to reserve flexibility to facilitate the subsequent decomposition and issuance of dispatch instructions. However, current research has mostly focused on assessing the range of flexibility without further refining the flexibility definition, resulting in insufficient granularity in flexibility assessment, making it difficult to meet actual dispatch requirements. Summary of the Invention

[0006] The purpose of the present invention is to overcome the deficiencies in the above-mentioned prior art and to provide a cluster-distribution network flexibility assessment method and system taking into account communication attacks. The method takes into account the impact of communication attacks and photovoltaic uncertainty on the flexibility of the distribution network, and further evaluates the upward and downward adjustment capabilities, baseline operating power and standby flexibility on the basis of the flexibility range, thereby achieving a comprehensive and integrated assessment of the flexibility of the distribution network.

[0007] In order to achieve the above object, the present invention is implemented through the following technical solutions:

[0008] On the one hand, the technical solution of the present invention provides a cluster-distribution network flexibility assessment method considering communication attacks, comprising:

[0009] Aggregate distributed resources into clusters, extract the initial state parameters of each distributed resource affected by communication, and establish a compact operational feasible domain for distributed resources that takes into account the impact of the communication process.

[0010] Model the false information injection attack and establish a robust box uncertainty set of the initial state parameters under the influence of communication attacks;

[0011] Considering the uncertainty of photovoltaic power generation, a distributed robust fuzzy set of photovoltaic power prediction deviation is constructed.

[0012] A cluster robust-stochastic flexibility evaluation model is established to solve the flexibility range reported to the distribution network, the local reserve flexibility reserved to cope with photovoltaic power uncertainty, and the local reserve flexibility reserved to cope with communication attacks.

[0013] Establish a distribution network flexibility range evaluation model to obtain the distribution network flexibility range under network constraints and the flexibility range that can be called upon by each cluster;

[0014] A distribution network baseline operating power and reserve flexibility configuration model is established, and the distribution network flexibility evaluation results are obtained by solving the problem.

[0015] In at least one embodiment, extracting initial state parameters of each distributed resource affected by communication and establishing a compact operational feasible domain of distributed resources considering the impact of the communication process specifically includes:

[0016] Construct the state equation of each distributed resource separately;

[0017] The initial state parameters affected by the communication process are extracted from the state equation as vectors, a compact representation of the distributed resource state equation is obtained, and a compact form of operation feasible domain of distributed resources considering the influence of the communication process is established.

[0018] In at least one embodiment, the modeling of the false information injection attack and the establishment of a robust box uncertainty set of initial state parameters under the influence of the communication attack specifically include:

[0019] By introducing false data parameters, false information injection attacks are modeled;

[0020] Based on the false information injection attack model, the initial state parameters received by the cluster and the amplitude of the false information injection attack are obtained, so as to establish a robust box uncertainty set of the initial state parameters under the influence of communication attack.

[0021] In at least one embodiment, the step of considering photovoltaic uncertainty and constructing a distributed robust fuzzy set of photovoltaic power prediction deviations specifically includes:

[0022] Considering the uncertainty of photovoltaic aggregate power, the difference between the actual photovoltaic power in historical data and the predicted photovoltaic power is calculated to obtain a sample set of prediction deviations containing multiple groups of data, and an empirical distribution of prediction deviations is constructed;

[0023] The distributed robust fuzzy set of photovoltaic power prediction deviation is established based on the empirical distribution of prediction deviation and Wasserstein distance.

[0024] In at least one embodiment, establishing a cluster robust-random flexibility evaluation model to solve for the flexibility range of the reported distribution network, the local reserve flexibility reserved to cope with photovoltaic power uncertainty, and the local reserve flexibility reserved to cope with communication attacks specifically includes:

[0025] By adding backup flexibility opportunity constraints, a cluster robust-stochastic flexibility evaluation model considering PV power forecast deviation and communication attacks is established.

[0026] For the distributed resource compact form operation feasible domain with uncertain parameters, it is transformed into a solvable form based on robust dual transformation;

[0027] Based on the distributed robust optimization method, the chance constraints are reconstructed into a probabilistic upper backup constraint set and a probabilistic lower backup constraint set.

[0028] The cluster robust-stochastic flexibility evaluation model is transformed into a convex optimization problem and solved to obtain the upper and lower limits of the cluster flexibility range as well as the upper and lower reserve flexibility reserved locally to cope with the uncertainty of photovoltaic power.

[0029] The difference between the upper and lower limits of the two sets of flexibility ranges obtained when the robustness parameter is 0 and the set value is calculated respectively, and the upper and lower backup flexibility retained locally to deal with communication attacks are obtained.

[0030] In at least one embodiment, establishing a distribution network flexibility range evaluation model to obtain the distribution network flexibility range and the flexibility range that can be called by each cluster under network constraints specifically includes:

[0031] The cluster flexibility range is converted into the flexibility range of the node where the cluster is located, and the distribution network flexibility range evaluation model is established in combination with the distribution network DistFlow power flow constraint set;

[0032] The distribution network flexibility range evaluation model is solved to obtain the upper and lower limits of the distribution network flexibility range under network constraints, and the active operating power values ​​of the clusters in the two optimization results are used as the upper and lower limits of the flexibility range that can be called by each cluster.

[0033] In at least one embodiment, the distribution network flexibility assessment results include the baseline operating power of the distribution network, the upward adjustment capability and downward adjustment capability reported to the dispatching center, the total reserve flexibility retained locally in the distribution network, and the baseline operating power and assumed reserve flexibility of each cluster.

[0034] In at least one embodiment, establishing a distribution network baseline operating power and standby flexibility configuration model and solving to obtain a distribution network flexibility assessment result specifically includes:

[0035] Constructing an objective function, the objective function including: minimizing the reserve flexibility of the distribution network for coping with power fluctuations of the distributed photovoltaic power station and minimizing the difference between the ratio of the upward adjustment capacity to the downward adjustment capacity in the evaluation result and the demand proportion coefficient;

[0036] Constructing constraint conditions, the constraint conditions including: integrating power forecast deviations of photovoltaic power plants at each node and assigning tasks for coping with the forecast deviations to each cluster using a contribution coefficient, and ensuring that the probability that the backup flexibility contributed by each cluster meets the forecast deviation is no less than a confidence parameter;

[0037] Based on the objective function and constraints, a distribution network baseline operating power and reserve flexibility configuration model is established and solved to obtain the total reserve flexibility of the distribution network to cope with the power deviation of the photovoltaic power station, as well as the reserve flexibility and contribution factor shared by each cluster.

[0038] In at least one embodiment, the value of the injected power at the root node in the optimization result of the distribution network baseline operating power and the backup flexibility configuration model is taken as the distribution network baseline operating power; the value of the cluster operating power in the optimization result of the distribution network baseline operating power and the backup flexibility configuration model is taken as the baseline operating power of each cluster.

[0039] On the other hand, the technical solution of the present invention further provides a cluster-distribution network flexibility assessment system considering communication attacks, comprising:

[0040] The first establishment module is configured to: aggregate distributed resources into clusters, extract initial state parameters of each distributed resource affected by communication, and establish a compact form operation feasible domain of distributed resources considering the influence of the communication process;

[0041] The second building module is configured to: model the false information injection attack and establish a robust box uncertainty set of initial state parameters under the influence of the communication attack;

[0042] The third building module is configured to: consider photovoltaic uncertainty and construct distributed robust fuzzy sets of photovoltaic power prediction deviation;

[0043] The cluster evaluation module is configured to: establish a cluster robust-stochastic flexibility evaluation model to solve for the flexibility range reported to the distribution network, the local reserve flexibility reserved to cope with photovoltaic power uncertainty, and the local reserve flexibility reserved to cope with communication attacks;

[0044] The first distribution network evaluation and establishment module is configured to: establish a distribution network flexibility range evaluation model to solve and obtain the distribution network flexibility range and the flexibility range that can be called by each cluster under network constraints;

[0045] The second distribution network evaluation module is configured to: establish a distribution network baseline operating power and standby flexibility configuration model, and solve to obtain a distribution network flexibility evaluation result.

[0046] The beneficial effects of the technical solution of the present invention are as follows:

[0047] The cluster-distribution network flexibility evaluation method considering communication attacks of the present invention aggregates decentralized distributed resources into clusters and establishes a two-layer flexibility evaluation architecture for clusters and distribution networks. At the cluster level, a false information injection attack modeling method based on robust optimization and a photovoltaic uncertainty modeling method based on distributed robust chance constraints are proposed, and a cluster robust-random flexibility evaluation model is established. At the distribution network level, a distribution network random flexibility evaluation model is established, including a distribution network flexibility range evaluation model based on a simplified DistFlow equation and a distribution network baseline operating power and backup flexibility configuration model that introduces a demand ratio coefficient and a contribution coefficient. The present invention takes into account the impact of communication attacks and photovoltaic uncertainties on the flexibility of the distribution network, and further evaluates the upward and downward adjustment capabilities, baseline operating power and backup flexibility on the basis of the flexibility range, thereby achieving a comprehensive and integrated evaluation of the flexibility of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0049] Figure 1 This is a flowchart of the steps of the cluster-distribution network flexibility assessment method considering communication attacks proposed in Example 1 of the present invention. DETAILED DESCRIPTION

[0050] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0051] As introduced in the background technology, the purpose of the present invention is to overcome the shortcomings of the above-mentioned existing technologies and provide a cluster-distribution network flexibility evaluation method and system considering communication attacks. It takes into account the impact of communication attacks and photovoltaic uncertainties on the flexibility of the distribution network, and further evaluates the upward and downward adjustment capabilities, baseline operating power and backup flexibility on the basis of the flexibility range, thereby realizing a comprehensive and integrated evaluation of the distribution network flexibility.

[0052] Example 1

[0053] In a typical embodiment of the present invention, Figure 1 As shown, this embodiment discloses a cluster-distribution network flexibility assessment method considering communication attacks, including the following steps:

[0054] S1. Aggregate distributed resources into clusters, extract the initial state parameters of each distributed resource affected by communication, and establish a feasible domain for the compact operation of distributed resources considering the influence of the communication process.

[0055] S2. Model the false information injection attack and establish a robust box uncertainty set of the initial state parameters under the influence of communication attacks;

[0056] S3. Considering the uncertainty of photovoltaic power generation, construct the distributed robust fuzzy set of photovoltaic power prediction deviation;

[0057] S4. Establish a cluster robust-random flexibility evaluation model to solve the flexibility range of the reported distribution network, the local reserve flexibility reserved to deal with photovoltaic power uncertainty, and the local reserve flexibility reserved to deal with communication attacks;

[0058] S5. Establish a distribution network flexibility range evaluation model to obtain the distribution network flexibility range and the flexibility range that can be called by each cluster under network constraints;

[0059] S6. Establish a distribution network baseline operating power and standby flexibility configuration model, and solve it to obtain the distribution network flexibility evaluation result.

[0060] The cluster-distribution network flexibility assessment method considering communication attacks is described in detail below in conjunction with specific implementation methods.

[0061] S1. Aggregate distributed resources into clusters and establish a two-tier flexibility assessment framework for clusters and distribution networks. Focus the research on three distributed resources: temperature-controlled loads, distributed energy storage, and electric vehicles. Construct state equations for each distributed resource and extract the initial state parameters affected by communication in the state equations separately. Establish a feasible domain for compact operation of distributed resources that considers the impact of the communication process. The specific process is as follows:

[0062] S101. Construct the state equations of temperature control load, distributed energy storage, and electric vehicles respectively.

[0063] In this step, the temperature control load state equation is constructed based on the equivalent thermal parameter model, which is specifically expressed as:

[0064] (1);

[0065] Where, denote the total number and width of evaluation periods, respectively. represents a set of integers; express t Time period n The operating power of each temperature-controlled load; express t Time period n The indoor temperature of the temperature-controlled load; Indicates the n The initial indoor temperature of the temperature-controlled load; express t Outdoor temperature during the period; Indicates auxiliary parameters; Respectively represent n Thermal resistance, heat capacity, and cooling / heating performance coefficient of the temperature control load; Respectively represent n The upper and lower limits of the indoor temperature of the temperature control load are affected by the user's suitable temperature and the allowable fluctuation range.

[0066] The distributed energy storage state equation is constructed based on the charge and discharge operation model, which is specifically expressed as:

[0067] (2);

[0068] Where: Respectively t Time period n The charging and discharging power of each energy storage device; express t Time period n The state of charge of each energy storage device; Indicates the n The initial state of charge of each energy storage device; Indicates then Self-discharge coefficient of each energy storage device; Indicates the n The charging and discharging efficiency coefficient of each energy storage device; Respectively represent n The upper and lower capacity limits of each energy storage device.

[0069] The state equation of the grid-connected electric vehicle is constructed based on the charge and discharge operation model, which is specifically expressed as follows:

[0070] (3);

[0071] Where: Respectively t Time period n Charging and discharging power of electric vehicles; express t Time period n The state of charge of the electric vehicle; Indicates the n The initial state of charge of an electric vehicle; Indicates the n The charging and discharging efficiency coefficient of electric vehicles; Respectively t Time period n The upper and lower limits of the state of charge of an electric vehicle are affected by the start time of charging, the end time of charging, the remaining power and the target power of the electric vehicle.

[0072] S102. Before each evaluation begins, the initial state parameters of each distributed resource need to be measured in real time by the user-side intelligent terminal and transmitted to the cluster via the user-side communication network. The initial state parameters affected by the communication process are extracted from the state equation as vectors, thereby obtaining a compact representation of the distributed resource state equation, specifically expressed as:

[0073] (4);

[0074] Where: represents the auxiliary matrix; represents the decision vector; represents the initial state vector affected by communication; They represent the upper limit vector and lower limit vector of distributed resource state parameters respectively.

[0075] For temperature control load, Abbreviated as , then the auxiliary matrix in formula (4) And each vector is expressed as:

[0076] (5);

[0077] For distributed energy storage, Abbreviated as , then the auxiliary matrix in formula (4) And each vector is expressed as:

[0078] (6);

[0079] For grid-connected electric vehicles, Abbreviated as , then the auxiliary matrix in formula (4) And each vector is expressed as:

[0080] (7);

[0081] S103. Based on the compact representation of the distributed resource state equation in S102, a distributed resource compact form operation feasible domain considering the influence of the communication process is established, which is specifically expressed as:

[0082] (8);

[0083] Where: It represents the feasible domain of operation; Represents the operating power constraint set of each distributed resource; Indicates the operating power of each distributed resource.

[0084] S2. Model the False Data Injection (FDI) attack and establish a robust box uncertainty set of the initial state parameters under the influence of communication attacks.

[0085] S201. By introducing false data parameters, a false information injection attack is modeled, which is specifically expressed as follows:

[0086] (9);

[0087] Where: Indicates the false data parameter, which is 0 when no false information injection attack occurs; Indicates the initial state parameters received by the cluster; represents the true initial state parameters; Indicates the magnitude of the false information injection attack.

[0088] S202. Based on the false information injection attack model constructed in S201, the initial state parameters received by the cluster and the amplitude of the false information injection attack are obtained, thereby establishing a robust box uncertainty set of the initial state parameters under the influence of the communication attack, which is specifically expressed as:

[0089] (10);

[0090] Where: represents the robust box uncertainty set; represent the upper and lower bounds of the uncertainty of the initial state parameters, respectively.

[0091] S3. Considering the uncertainty of photovoltaic power generation, a distributed robust fuzzy set of photovoltaic power prediction deviation is constructed.

[0092] S301. Considering that the aggregated photovoltaic power of the rooftops in the cluster is uncertain, based on the uncertainty of the aggregated photovoltaic power, the aggregated photovoltaic power is expressed as:

[0093] (11);

[0094] Where: represents the actual value of the aggregated photovoltaic power; represents the predicted value, Indicates the deviation between the predicted value and the actual value.

[0095] S302, by calculating the difference between the actual photovoltaic power in the historical data and the predicted photovoltaic power, obtain M Prediction bias sample set of group data ,in Indicates the m The empirical distribution of prediction deviations is constructed based on the prediction sample deviation set, which is specifically expressed as:

[0096] (12);

[0097] Where: represents the empirical distribution; express The Dirac measure of .

[0098] S303: Based on the empirical distribution of the prediction deviation and the Wasserstein distance, a distributed robust fuzzy set of the photovoltaic power prediction deviation is established, which is specifically expressed as follows:

[0099] (13);

[0100] Where: represents distributed blue stick fuzzy set; represents the probability distribution of the forecast deviation; represents the Wasserstein distance; represents the radius of the Wasserstein sphere; represents the set of all distributions that satisfy the support set of the polyhedron; represents the support set of the polyhedron, where represents the bias vector, denote the support matrix and support vector respectively.

[0101] S4. Establish a cluster robust-random flexibility evaluation model to solve the flexibility range of the reported distribution network, the local reserve flexibility reserved to deal with photovoltaic power uncertainty, and the local reserve flexibility reserved to deal with communication attacks.

[0102] S401. By adding backup flexibility opportunity constraints, a cluster robust-random flexibility evaluation model considering photovoltaic power forecast deviation and communication attacks is established. The specific expression is:

[0103] (14);

[0104] (15);

[0105] Where: They represent the upper and lower limits of the cluster flexibility range respectively; N represents the total number of distributed resources participating in the evaluation within the cluster; Represents a collection of resources; represents the aggregate power of nodes in period t; represents the confidence parameter; They represent the upper and lower spare flexibility of preserving local response to randomness injection; Indicates other fixed load power within the node aggregator.

[0106] S402: For the distributed resource compact form operation feasible domain with uncertain parameters, it is converted into a solvable form based on the robust dual transformation, which is specifically expressed as:

[0107] (16);

[0108] Where: Represents the feasible domain of robust operation of distributed resources; represents the initial state parameter vector received by the cluster; The magnitude vector representing the false information injection attack; and represents the auxiliary decision vector, represents the set of real numbers; represents auxiliary decision variables; Represents the robustness parameter, the higher the value, the more conservative it is.

[0109] S403. Based on the distributional robust optimization method, the chance constraints are reconstructed into a probabilistic upper backup constraint set and a probabilistic lower backup constraint set.

[0110] Taking the above backup flexibility as an example, the probabilistic backup constraint set is specifically expressed as:

[0111] (17);

[0112] Where: represents the probabilistic upper spare constraint set; and indivual represents the distributed robust decision vector; represents the distributed robust decision matrix, represents an element in the decision matrix; Represents an auxiliary vector, except for the tth element which is 1, the rest of the elements are 0.

[0113] Similarly, the probability of calculating the backup constraint set is .

[0114] S404. Convert the cluster robustness-random flexibility evaluation model into a convex optimization problem, which is specifically expressed as:

[0115] (18);

[0116] (19);

[0117] Solving the convex optimization problem can obtain the upper limit of the cluster flexibility range and lower limit , and local reserve flexibility to cope with PV power uncertainty and lower standby flexibility . is used in the next evaluation process.

[0118] Locally retain the flexibility of up and down backup to deal with communication attacks Taking the above standby flexibility as an example, it can be calculated by taking the robustness parameter as 0 (indicating ignoring communication attacks) and taking the set value The difference between the two sets of flexible range upper limits obtained when is:

[0119] (20);

[0120] Where: Respectively represent the robustness parameter taking 0 and the set value Similarly, the calculation of robustness parameters takes 0 (indicates ignoring communication attacks) and takes the set value The difference between the lower limits of the two sets of flexible ranges obtained at this time is the lower spare flexibility reserved locally to deal with communication attacks.

[0121] S5. Establish a distribution network flexibility range assessment model to obtain the distribution network flexibility range under network constraints and the flexibility range that can be called upon by each cluster. This serves as the first part of the distribution network random flexibility assessment model. The specific process is as follows:

[0122] S501. Establish a distribution network flow constraint set based on the simplified DistFlow model, which is specifically expressed as follows:

[0123] (twenty one);

[0124] Where: Represents the distribution network DistFlow flow constraint set; express t Time period slave node i Flow Node j Active and reactive power; express t Time period node j Fixed active and reactive load power; express t Time period cluster j The active and reactive operating power, power factor is 0.85; express t Time period node j The predicted active power of the distributed photovoltaic power station is adjusted to 0; Respectively represent branches ij resistance and reactance; Respectively represent branches ij Active power lower limit, active power upper limit, reactive power lower limit and reactive power upper limit; express t Time period node j The square of the voltage; Indicates the allowable voltage deviation, which is 0.05.

[0125] S502: Based on the upper limit of the cluster flexibility range obtained in S404 and lower limit After reporting to the distribution network operator, it is converted into the upper limit of the flexibility range of the node where the cluster is located. and lower limit On this basis, the distribution network DistFlow power flow constraint set obtained in S501 is combined to establish a distribution network flexibility range evaluation model, which is specifically expressed as follows:

[0126] (twenty two);

[0127] (twenty three);

[0128] Where: They represent the upper and lower limits of the distribution network flexibility range under network constraints; express t The injected power at the root node of the time period; express t Time period cluster j Upper and lower limits of upload flexibility; Represents the set of grid-connected nodes of a cluster.

[0129] By solving the above distribution network flexibility range evaluation model, the upper limit of the distribution network flexibility range under network constraints is obtained. and lower limit and the active operating power of the cluster in the two optimization results The value of is used as the upper limit of the flexibility range that each cluster can call and lower limit Among them, the first optimization result is the optimal value corresponding to each decision variable obtained by solving model (22), and the second optimization result is the optimal value corresponding to each decision variable obtained by solving model (23).

[0130] S6. As the second part of the distribution network random flexibility assessment model, a distribution network baseline operating power and reserve flexibility configuration model is established to obtain the distribution network flexibility assessment results, including the distribution network baseline operating power, the upward and downward adjustment capabilities reported to the dispatch center, the total reserve flexibility retained in the distribution network, and the baseline operating power and assumed reserve flexibility of each cluster. The specific process is as follows:

[0131] S601. Construct an objective function. The optimization objectives of the distribution network operator include two parts. The first part minimizes the local reserve flexibility of the distribution network to cope with the power fluctuations of distributed photovoltaic power plants, ensuring that the flexibility reported to the dispatch center is maximized while meeting local reserve requirements. The second part minimizes the difference between the ratio of the upward adjustment capacity to the downward adjustment capacity in the evaluation results and the demand ratio coefficient, so that the evaluation results can best meet the needs of the dispatch center. The mathematical form of the objective function is as follows:

[0132] (twenty four);

[0133] Where: denote the upper reserve flexibility and lower reserve flexibility contributed by the jth cluster in period t, respectively; They represent the upward adjustment capability vector and downward adjustment capability vector reported to the dispatch center in time period t respectively; Indicates the flexibility demand ratio coefficient issued by the dispatch center.

[0134] S602, build constraints. On the one hand, the distribution network operator needs to integrate the power forecast deviations of the photovoltaic power stations at each node and assign the task of dealing with the forecast deviations to each cluster through the contribution coefficient; on the other hand, it needs to ensure that the probability of the backup flexibility contributed by each cluster meeting the forecast deviation is not less than the confidence parameter through opportunity constraints. The mathematical form of the constraints is as follows:

[0135] (25);

[0136] Where: They represent the upward adjustment capability and downward adjustment capability reported by the distribution network operator to the dispatching center in period t respectively; They represent the upper reserve flexibility and lower reserve flexibility retained by the distribution network operator in period t; Represents the set of grid-connected nodes of distributed photovoltaic power stations; They represent the upper and lower limits of the distribution network flexibility range under the network constraints obtained in step 5 respectively; They represent the upper and lower limits of the flexibility range of each cluster obtained in step 5; represents the power prediction deviation of the jth distributed photovoltaic power station in period t; represents the total forecast deviation; It represents the backup contribution coefficient of the jth cluster in the tth period.

[0137] S603: Based on the above two objective functions and two constraints, a distribution network baseline operating power and backup flexibility configuration model is established and solved.

[0138] Specifically, in this step, the complex opportunity constraint expressed by Equation (25) can be transformed into a convex constraint form similar to Equation (17), and then the distribution network baseline operating power and reserve flexibility configuration model is transformed into a convex optimization problem that can be directly solved. By solving the model, the total reserve flexibility of the distribution network to cope with the power deviation of the photovoltaic power station is obtained. , as well as the backup flexibility and contribution factors shared by each cluster Furthermore, the distribution network baseline operating power and the baseline operating power of each cluster Inject power into the root node of the optimization result of the distribution network baseline operating power and reserve flexibility configuration model and cluster operating power The value of .

[0139] Example 2

[0140] In a typical embodiment of the present invention, this embodiment discloses a cluster-distribution network flexibility assessment system considering communication attacks, including:

[0141] The first establishment module is configured to: aggregate distributed resources into clusters, extract initial state parameters of each distributed resource affected by communication, and establish a compact form operation feasible domain of distributed resources considering the influence of the communication process;

[0142] The second building module is configured to: model the false information injection attack and establish a robust box uncertainty set of initial state parameters under the influence of the communication attack;

[0143] The third building module is configured to: consider photovoltaic uncertainty and construct distributed robust fuzzy sets of photovoltaic power prediction deviation;

[0144] The cluster evaluation module is configured to: establish a cluster robust-random flexibility evaluation model and solve it to obtain the cluster flexibility range;

[0145] The first distribution network evaluation and establishment module is configured to: establish a distribution network flexibility range evaluation model to solve and obtain the distribution network flexibility range and the flexibility range that can be called by each cluster under network constraints;

[0146] The second distribution network assessment module is configured to: establish a distribution network baseline operating power and standby flexibility configuration model, and solve to obtain the distribution network flexibility assessment result

[0147] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A cluster-distribution network flexibility assessment method considering communication attacks, characterized by: include: Aggregate distributed resources into clusters, extract the initial state parameters of each distributed resource affected by communication, and establish a compact operational feasible domain for distributed resources that takes into account the impact of the communication process. Model the false information injection attack and establish a robust box uncertainty set of the initial state parameters under the influence of communication attacks; Considering the uncertainty of photovoltaic power generation, a distributed robust fuzzy set of photovoltaic power prediction deviation is constructed. The method of considering photovoltaic uncertainty and constructing a distributed robust fuzzy set of photovoltaic power prediction deviation specifically includes: Considering the uncertainty of photovoltaic aggregate power, the difference between the actual photovoltaic power in historical data and the predicted photovoltaic power is calculated to obtain a sample set of prediction deviations containing multiple groups of data, and an empirical distribution of prediction deviations is constructed; Based on the empirical distribution of prediction deviation and Wasserstein distance, a distributed robust fuzzy set of photovoltaic power prediction deviation is established. A cluster robust-stochastic flexibility evaluation model is established to solve the flexibility range reported to the distribution network, the local reserve flexibility reserved to cope with photovoltaic power uncertainty, and the local reserve flexibility reserved to cope with communication attacks. Establish a distribution network flexibility range evaluation model to obtain the distribution network flexibility range under network constraints and the flexibility range that can be called upon by each cluster; A distribution network baseline operating power and reserve flexibility configuration model is established, and the distribution network flexibility evaluation results are obtained by solving the problem.

2. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 1, characterized in that: The extraction of initial state parameters of each distributed resource affected by communication and the establishment of a compact operational feasible domain of distributed resources considering the influence of the communication process specifically include: Construct the state equation of each distributed resource separately; The initial state parameters affected by the communication process are extracted from the state equation as vectors, a compact representation of the distributed resource state equation is obtained, and a compact form of operation feasible domain of distributed resources considering the influence of the communication process is established.

3. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 1, characterized in that: The false information injection attack model is constructed to establish a robust box uncertainty set of initial state parameters under the influence of communication attacks, specifically including: By introducing false data parameters, false information injection attacks are modeled; Based on the false information injection attack model, the initial state parameters received by the cluster and the amplitude of the false information injection attack are obtained, so as to establish a robust box uncertainty set of the initial state parameters under the influence of communication attack.

4. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 1, characterized in that: The cluster robust-random flexibility evaluation model is established to solve the flexibility range of the reported distribution network, the local reserve flexibility to cope with photovoltaic power uncertainty, and the local reserve flexibility to cope with communication attacks, specifically including: By adding backup flexibility opportunity constraints, a cluster robust-stochastic flexibility evaluation model considering PV power forecast deviation and communication attacks is established. For the distributed resource compact form operation feasible domain with uncertain parameters, it is transformed into a solvable form based on robust dual transformation; Based on the distributed robust optimization method, the chance constraints are reconstructed into a probabilistic upper backup constraint set and a probabilistic lower backup constraint set. The cluster robust-stochastic flexibility evaluation model is transformed into a convex optimization problem and solved to obtain the upper and lower limits of the cluster flexibility range as well as the upper and lower reserve flexibility reserved locally to cope with the uncertainty of photovoltaic power. The difference between the upper and lower limits of the two sets of flexibility ranges obtained when the robustness parameter is 0 and the set value is calculated respectively, and the upper and lower backup flexibility retained locally to deal with communication attacks are obtained.

5. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 1, characterized in that: The distribution network flexibility range evaluation model is established to solve the distribution network flexibility range under network constraints and the flexibility range that can be called by each cluster, specifically including: The cluster flexibility range is converted into the flexibility range of the node where the cluster is located, and the distribution network flexibility range evaluation model is established in combination with the distribution network DistFlow power flow constraint set; The distribution network flexibility range evaluation model is solved to obtain the upper and lower limits of the distribution network flexibility range under network constraints, and the active operating power values ​​of the clusters in the two optimization results are used as the upper and lower limits of the flexibility range that can be called by each cluster.

6. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 1, characterized in that: The distribution network flexibility assessment results include the baseline operating power of the distribution network, the upward adjustment capability and downward adjustment capability reported to the dispatching center, the total reserve flexibility retained in the distribution network, the baseline operating power of each cluster and the reserve flexibility assumed.

7. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 1, characterized in that: The establishment of the distribution network baseline operating power and standby flexibility configuration model and the solution to obtain the distribution network flexibility evaluation result specifically include: Constructing an objective function, the objective function including: minimizing the reserve flexibility of the distribution network for coping with power fluctuations of the distributed photovoltaic power station and minimizing the difference between the ratio of the upward adjustment capacity to the downward adjustment capacity in the evaluation result and the demand proportion coefficient; Constructing constraint conditions, the constraint conditions including: integrating power forecast deviations of photovoltaic power plants at each node and assigning tasks for coping with the forecast deviations to each cluster using a contribution coefficient, and ensuring that the probability that the backup flexibility contributed by each cluster meets the forecast deviation is no less than a confidence parameter; Based on the objective function and constraints, a distribution network baseline operating power and reserve flexibility configuration model is established and solved to obtain the total reserve flexibility of the distribution network to cope with the power deviation of the photovoltaic power station, as well as the reserve flexibility and contribution factor shared by each cluster.

8. The cluster-distribution network flexibility assessment method considering communication attacks according to claim 7, characterized in that: The value of the injected power at the root node in the optimization result of the distribution network baseline operating power and the backup flexibility configuration model is taken as the distribution network baseline operating power; the value of the cluster operating power in the optimization result of the distribution network baseline operating power and the backup flexibility configuration model is taken as the baseline operating power of each cluster.

9. A cluster-distribution network flexibility evaluation system considering communication attacks, characterized in that: include: The first establishment module is configured to: aggregate distributed resources into clusters, extract initial state parameters of each distributed resource affected by communication, and establish a compact form operation feasible domain of distributed resources considering the influence of the communication process; The second building module is configured to: model the false information injection attack and establish a robust box uncertainty set of initial state parameters under the influence of the communication attack; The third building module is configured to: consider photovoltaic uncertainty and construct distributed robust fuzzy sets of photovoltaic power prediction deviation; The method of considering photovoltaic uncertainty and constructing a distributed robust fuzzy set of photovoltaic power prediction deviation specifically includes: Considering the uncertainty of photovoltaic aggregate power, the difference between the actual photovoltaic power in historical data and the predicted photovoltaic power is calculated to obtain a sample set of prediction deviations containing multiple groups of data, and an empirical distribution of prediction deviations is constructed; Based on the empirical distribution of prediction deviation and Wasserstein distance, a distributed robust fuzzy set of photovoltaic power prediction deviation is established. The cluster evaluation module is configured to: solve and obtain the flexibility range of the reported distribution network, the local reserve flexibility reserved to deal with photovoltaic power uncertainty, and the local reserve flexibility reserved to deal with communication attacks; The first distribution network evaluation and establishment module is configured to: establish a distribution network flexibility range evaluation model to solve and obtain the distribution network flexibility range and the flexibility range that can be called by each cluster under network constraints; The second distribution network evaluation module is configured to: establish a distribution network baseline operating power and standby flexibility configuration model, and solve to obtain a distribution network flexibility evaluation result.

Images