Micro-grid group information interaction and control system and method

Abstract

The application provides a micro-grid group information calculation and interaction system and method, wherein the system comprises: a physical layer for describing the interconnected topological structure of the micro-grid group; a network layer for abstracting corresponding information interaction objects into agent nodes based on the interconnected topological structure of the micro-grid group, obtaining local measurement data of the micro-grid group through the agent nodes, and performing data interaction through private data negotiation and abnormality identification through consensus anomaly sensing; meanwhile, the network layer generates a control strategy for energy scheduling of the micro-grid group based on the local measurement data; and a service layer for realizing information interaction between the network layer and obtaining required abnormality identification information, state estimation information and control strategy information. The application only exchanges and stores boundary values while protecting privacy, reduces transmission and storage requirements, effectively improves the flexibility and economy of micro-grid operation control, develops the potential of the micro-grid, and lays a foundation for subsequent energy management and scheduling of the micro-grid group.

Description

Technical Field

[0001] This invention relates to the field of microgrid technology, specifically to a microgrid group information interaction and control system and method, and also provides a corresponding terminal and medium. Background Technology

[0002] A microgrid is a small-scale power generation and distribution system composed of distributed power sources, energy storage devices, energy conversion devices, loads, monitoring and protection devices, etc. The microgrid was proposed to enable the flexible and efficient application of distributed power sources and solve the grid connection problem for a large number of diverse distributed power sources. Developing and extending microgrids can fully promote the large-scale integration of distributed power sources and renewable energy, achieving highly reliable supply of multiple energy forms to loads. It is an effective way to realize an active distribution network, enabling the transition from traditional power grids to smart grids.

[0003] Compared to a single microgrid, a microgrid cluster requires further consideration of its topology, resulting in diverse physical structures. To improve the stability of a microgrid cluster system, frequent power flows and information exchanges between microgrids are necessary. Therefore, the design of the information exchange and control system for a microgrid cluster is fundamental to subsequent energy management and scheduling. Each sub-microgrid in a microgrid cluster only communicates with its neighboring sub-microgrids. Using a central control system for information exchange would not only increase the construction and maintenance costs of communication but also compromise information privacy. Existing technologies often assume perfectly observable measurement data and fail to address the issue of protecting microgrid data privacy and achieving distributed data protection. Furthermore, network anomalies such as topology changes and bad data can also interfere with energy management and scheduling within the microgrid cluster.

[0004] Currently, no descriptions or reports of technologies similar to this invention have been found, and no similar information has been collected domestically or internationally. Summary of the Invention

[0005] To address the aforementioned shortcomings in the prior art, this invention provides a microgrid group information interaction and control system and method, along with a corresponding terminal and medium.

[0006] According to one aspect of the present invention, a microgrid group information interaction and control system is provided, comprising: a physical layer, a network layer, and a service layer; wherein:

[0007] The physical layer is used to describe the interconnection topology that makes up the microgrid group;

[0008] Based on the interconnected topology of the microgrid cluster, the network layer abstracts the corresponding information interaction objects into proxy nodes, obtains local measurement data of the microgrid cluster through the proxy nodes, performs data interaction through private data negotiation, and identifies anomalies through consensus anomaly detection; at the same time, based on the local measurement data, it generates a control strategy for energy scheduling of the microgrid cluster.

[0009] The service layer is used to realize information interaction with the network layer and obtain the required anomaly identification information, state estimation information and control strategy information.

[0010] Optionally, the microgrid group includes: a centralized physical topology microgrid group, a distributed physical topology microgrid group, and a hybrid physical topology microgrid group.

[0011] Optionally, the network layer includes: a microgrid layer, a tie-line layer, and a distribution network layer; wherein:

[0012] The microgrid layer includes: internal nodes and boundary nodes; the internal nodes correspond to the structural units of the physical layer and are used to acquire the measurement data of the microgrid and perform state estimation calculations at the regional grid level; the boundary nodes are used to transmit the calculation results of the state estimation to the tie-line layer.

[0013] The tie-line level receives its own measurement data at the induction tie-line, and at the same time receives voltage estimates from the boundary nodes of the microgrid level. It exchanges voltage estimate data between boundary nodes through private data negotiation and identifies anomalies through consensus anomaly detection.

[0014] The distribution network level collects voltage distribution data from each tie-line level to obtain the voltage distribution of the entire network. Based on the voltage distribution, it provides control optimization targets for the network layer to implement control strategies for energy dispatching of the microgrid group.

[0015] Optionally, in the microgrid level, acquiring the microgrid's measurement data and performing state estimation calculations at the regional grid level includes:

[0016] We employ a weighted least squares algorithm as the parallel local state estimation algorithm for each region, considering the following measurement equation:

[0017] z = h(x) + e (18)

[0018] In the formula: z is the measurement vector in the sub-microgrid area; x is the state variable vector in the sub-microgrid area, with a dimension less than z; e is the measurement error vector, and its covariance matrix is ​​R. The covariance matrix R is set as a diagonal matrix, and the diagonal elements are the variances σ of each measurement data. 2h(x) represents the functional relationship between the state variable vector and the measurement vector.

[0019] The objective of the weighted least squares algorithm is to find the optimal estimate that minimizes the following objective function:

[0020]

[0021] In the formula, J(x) is the objective function for the desired state estimation, and z i For the i-th measurement data, h i (x) is the i-th measurement function, R ii is the i-th diagonal element of the covariance matrix, and m is the number of measurement data;

[0022] The optimal estimate is obtained by solving the following equation iteratively:

[0023] G (p) (x (p) )Δx (p) =HT(x (p) )R -1 [zh(x (p (20)

[0024] In the formula, p represents the p-th iteration; Δx (p) =x (p+1) -x (p) ; The Jacobian matrix representing the measurement function h(x); G (p) =H T (x (p) )R -1 H(x (p) ), where is the gain matrix.

[0025] Optionally, in the tie-line layer, voltage estimation data exchange at boundary nodes is performed through private data negotiation, and anomaly identification is performed through consensus anomaly detection, including:

[0026] The private data negotiation employs a parallel computing approach across microgrid agents to exchange voltage estimation data at boundary nodes in encrypted form, including:

[0027] When the measurement data of microgrid k is updated, the microgrid level m k To its adjacent tie-line level l m Send first encrypted information, which includes: measurement data related to the boundary node and the extended node, topology information of the extended node, micronet proxy address as the sending micronet layer and tie-line proxy address as the receiving tie-line layer, and the current time T. start Among them, extended nodes are internal nodes that are willing to share with the tie-line level;

[0028] The topology of the communication network of the microgrid cluster is represented as an undirected graph, where each sub-microgrid is considered a node in the graph, and the branches in the graph represent communication connections between sub-microgrids. Under normal circumstances, the undirected graph is connected, and its Laplace matrix L is obtained based on its topology. Let N... i Let |N| be the set of nodes that have a communicative connection with node i in an undirected graph. i | represents the number of nodes in the set, then the elements L of the Laplace matrix L are... ij Defined as:

[0029]

[0030] According to the finite-time consistency theory, for a Laplace matrix L of a connected graph, if there exist K independent non-zero eigenvalues ​​λ2≠λ3≠...≠λ K+1 If ≠0, then the values ​​at each node in the graph will reach the average value after K iterations, that is:

[0031]

[0032] The finite-time consensus algorithm iteration rule for each node in the graph is as follows:

[0033]

[0034] In the formula: v*(s) is the updated value of node * after the s-th iteration; ω ii (s) and ω ij (s) is the weight factor at the s-th iteration;

[0035] The weight coefficients are updated according to the following formula during each iteration to ensure convergence after the Kth iteration:

[0036]

[0037] In the formula, λ s+1 Let be the (s+1)th independent non-zero eigenvalue of the Laplacian matrix L;

[0038] After receiving the first information, the tie-line layer triggers the same parallel local estimation calculation as the microgrid layer. The parallel local estimation calculation at the microgrid layer uses weighted least squares and includes:

[0039] Consider the following measurement equation:

[0040] z = h(x) + e (25)

[0041] In the formula: z is the measurement vector in the sub-microgrid area; x is the state variable vector in the sub-microgrid area, with a dimension less than z; e is the measurement error vector, whose covariance matrix is ​​R, wherein the covariance matrix R is a diagonal matrix, and the diagonal elements in the matrix are the variances σ of each measurement data. 2 h(x) represents the functional relationship between the state variable vector and the measurement vector.

[0042] Calculate the optimal estimate that minimizes the following objective function.

[0043]

[0044] In the formula, J(x) is the objective function for the desired state estimation, and z i For the i-th measurement data, h i (x) is the i-th measurement function, R ii is the i-th diagonal element of the covariance matrix, and m is the number of measurement data;

[0045] Optimal estimate The solution is obtained through the following loop:

[0046] G (p) (x (p) )Δx (p) =H T (x (p) W (p) [z (p) -h(x (p) (27)

[0047] In the formula: p represents the p-th iteration; Δx (p) =x (p+1) -x (p) ; The Jacobian matrix representing the measurement function h(x); G (p) =H T (x (p) )R -1 H(x (p) Let be the gain matrix. Inverting the gain matrix yields the estimation error covariance matrix of the state variables.

[0048]

[0049] After parallel computation is completed, the tie-line layer sends a second encrypted message to the adjacent microgrid layer. The second encrypted message includes: the estimated value of the boundary node x = [v; θ], the tie-line proxy address of the sending tie-line layer and the microgrid proxy address of the receiving microgrid layer, and the current time T. end Where v is the node voltage magnitude vector and θ is the node voltage phase angle vector;

[0050] The consensus anomaly detection includes:

[0051] When the microgrid layer receives the second encrypted information, it triggers consensus anomaly detection, performing data comparison and anomaly identification; wherein:

[0052] After the microgrid layer receives the second encrypted information, it performs the following data consistency check to determine whether the consistency check is acceptable:

[0053]

[0054] In the formula: and These are micro-network agents m s and contact line agent l s The estimated value for the same boundary node i; For micro-network agent m s The local estimate of G is the i-th diagonal element; c is the proportionality coefficient. When c = 1, 2, 3, it indicates that the intervals falling into the standard Gaussian distribution are 68.2%, 95.4%, and 99.7% of the total interval, respectively.

[0055] After the data consistency check is completed, the micronet agent will draft the data consistency check requirements, including: the micronet agent and tie-line agent addresses, the micronet agent's estimate of the boundary nodes, the tie-line agent's estimate of the boundary nodes, the consistency status, and the time T when the first encrypted message was generated. start and the time T when the second encrypted information is generated end The consistency states include: acceptance and rejection; if the rejection in the consistency state comes from multiple adjacent micronet agents, the common contact line agent may be abnormal; if the rejection in the consistency state is from a single micronet agent, that micronet agent may be abnormal.

[0056] For a microgrid agent with a consistency rejection status, query the absolute deviation Δx of the boundary node estimate between historical time t-1 and current time t. t,t-1 If Δx t,t-1 Exceeding the set threshold Normal max If Δx t,t-1 Exceeding the set threshold FC max If the value is less than 1, it is considered to be negative data; among which, the threshold Normal max and FC max The value of is obtained from the error of the largest estimate in the historical data without anomalies, where:

[0057]

[0058]

[0059] According to equations (61) and (62), we get:

[0060]

[0061] In the formula: These are estimates for scenarios with bad data, environmental noise, data latency, and topological changes, respectively. is the true value; g(·) is the functional relationship between the measured value and the state estimate; z1 is the measured value under normal conditions of sub-microgrid area 1. These represent the measurement deviation values ​​for bad data, environmental noise, data latency, and topology changes, respectively. Y1, Y2, and ΔY1 are the topology matrices of the first sub-microgrid, the second sub-microgrid, and the change in the topology matrix when the first sub-microgrid undergoes a topology change, respectively.

[0062] Warnings about specific anomalies and their locations will be sent to the relevant microgrid area, and error data analysis or topology checks will be initiated.

[0063] Once the abnormal data comparison is complete, the abnormal identification information will be communicated to each layer of the network layer and broadcast to all nodes.

[0064] Optionally, the network layer employs a distribution network-level coordinated control strategy, an inter-regional control strategy, and a microgrid-based control strategy to achieve energy dispatching for the microgrid group; wherein:

[0065] The aforementioned distribution network hierarchical coordination control strategy collects voltage estimation data for each tie line to obtain the overall operation information of the distribution network. It then uses a global optimization algorithm to perform global-level optimization control on the distribution network and microgrid clusters. For the lower-level regions, it sets regional optimization targets, namely the exchange power of the sub-microgrids, to guide all controllable distributed generation units, flexible load units, and tie switches on feeders throughout the entire distribution network. Furthermore, it sets regional optimization targets within the microgrid clusters, namely the target value of the tie line output power and the tie line coordination coefficient k. i ;

[0066] The inter-regional control strategy coordinates all controllable distributed power sources and energy storage units within an autonomous region on a feeder through a tie-line hierarchy.

[0067] The microgrid control strategy divides the control structure of the sub-microgrid into two layers. The first layer adopts a droop control strategy based on equal-cost micro-increment to realize the proportional distribution of active power within the microgrid. The second layer adopts a microgrid DG control strategy based on the consistency principle to control the information interaction between distributed power sources on the same distribution node, thereby realizing secondary adjustment of the microgrid frequency, voltage, and reactive power, achieving voltage stability and power balance of the microgrid group, and completing the energy dispatch of the microgrid group.

[0068] Optionally, the distribution network-level coordination control strategy includes:

[0069] The distribution network-level coordinated control is achieved through the distribution network level, which monitors the regional exchange power index of the distribution network in real time and evaluates the status of each microgrid region in real time.

[0070] P i microgrid (t)=P i microgrid-in (t)-P i microgrid-out (t) (33)

[0071] In the formula, P i microgrid (t) represents the power exchanged by sub-microgrid i at time t; P i microgrid-in (t) represents the power flowing into sub-microgrid i at time t; P i microgrid-out (t) represents the power outflow value of sub-microgrid i at time t;

[0072] At fixed intervals, the global optimization target value is calculated based on the current state of the distribution network and the state assessment results. After being decomposed, the regional optimization target value is issued to the control areas of each level. The control areas respond to external disturbances in real time according to their corresponding control strategies, and coordinate the distributed energy within the control areas to minimize the gap between their respective control areas and the global target. As a result, the operation of the entire distribution network is closer to the global target optimization value.

[0073] Optionally, the inter-regional control strategy includes:

[0074] The inter-regional control is achieved through a tie-line hierarchy, which coordinates all controllable distributed generation and energy storage units within an autonomous region on a control feeder, forming the objective function for controlling sub-microgrid i:

[0075]

[0076] In the formula: P FCEi For the tie-line control error of sub-microgrid i; k i ΔP is the coordination factor of the tie line where sub-microgrid i is located; F P is the difference between the actual tie-line output power and its target value. F This represents the current output power of the tie line, where the power flowing from the busbar into the feeder is positive. The target value for the tie-line output power; ΔP microgrid-i P is the difference between the actual power exchange value and its optimized target value of sub-microgrid i;microgrid-i This represents the actual value of the exchange power of sub-microgrid i at the current moment; Let k be the target value of the switching power of sub-microgrid i; where k i , These are the global optimization target values ​​calculated and issued by the distribution network level at fixed intervals in the distribution network level coordination control, respectively. The tie line level updates simultaneously after receiving the three global optimization target values.

[0077] Optionally, the microgrid control strategy includes:

[0078] The microgrid control is implemented through a microgrid hierarchy, and the controlled objects are the active and reactive power outputs of each controllable distributed power source. To achieve the target of the microgrid control, the control structure of the sub-microgrid is divided into two layers. The primary control adopts a droop control strategy based on equal cost micro-increments to realize the proportional distribution of active power within the microgrid. The secondary control is based on a microgrid distributed power source control strategy based on the consistency principle, which realizes secondary adjustment of the microgrid frequency, voltage, and reactive power through information interaction between distributed power sources.

[0079] According to another aspect of the present invention, a method for information interaction and control of a microgrid group is provided, comprising:

[0080] Describe the interconnection topology that makes up a microgrid group;

[0081] Based on the interconnected topology of the microgrid cluster, the corresponding information interaction objects are abstracted as proxy nodes. The local measurement data of the microgrid cluster is obtained through the proxy nodes, and data interaction is carried out through private data negotiation and anomaly identification is carried out through consensus anomaly perception. At the same time, a control strategy for energy scheduling of the microgrid cluster is generated.

[0082] It obtains the necessary anomaly identification information, state estimation information, and control strategy information to achieve information exchange among microgrid groups.

[0083] According to a third aspect of the present invention, a terminal is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can be used to run the system described in any of the preceding claims, or to perform the methods described above.

[0084] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, can be used to run the system described in any of the preceding claims, or to perform the methods described above.

[0085] By adopting the above technical solution, the present invention has at least one of the following beneficial effects compared with the prior art:

[0086] This invention proposes a microgrid group information interaction mechanism that takes into account microgrid data privacy. It makes full use of user data within the microgrid (i.e., local measurement data of the microgrid group), resolves the contradiction between the observability of the microgrid group and the privacy of user data, and reduces transmission and storage requirements by exchanging and storing only boundary values ​​while protecting privacy.

[0087] This invention employs a decentralized framework and a parallel information exchange mechanism between the distribution network level, tie-line level, and microgrid level. It allows for flexible partitioning and parallel computation across microgrids, effectively alleviating computational pressure. Simultaneously, it is compatible with communication devices using different measurement sampling rates, triggering private data negotiation and consensus anomaly detection based on measurement data, thus achieving more flexible microgrid group information exchange.

[0088] This invention is scalable, and the interaction mechanism can be applied to large-scale microgrid clusters. Through distributed state estimation, it effectively improves the flexibility and economy of microgrid operation and control, and realizes the potential of microgrids.

[0089] This invention enables each sub-microgrid to achieve reliable information exchange while ensuring information privacy, effectively respond to network anomalies, improve the ability to observe status data, and lay the foundation for subsequent energy management and scheduling of microgrid groups. Attached Figure Description

[0090] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0091] Figure 1 This is a schematic diagram of the system architecture of the microgrid group information interaction and control system in a preferred embodiment of the present invention.

[0092] Figure 2 This is a schematic diagram of the network layer structure in a preferred embodiment of the present invention.

[0093] Figure 3 This is a flowchart of a microgrid group information interaction and control method in a preferred embodiment of the present invention.

[0094] Figure 4 This is a flowchart of the algorithm for private data negotiation in a preferred embodiment of the present invention.

[0095] Figure 5 This is a flowchart of the consensus anomaly detection algorithm in a preferred embodiment of the present invention.

[0096] Figure 6 This is a schematic diagram illustrating the operation of control information interaction in a preferred embodiment of the present invention. Detailed Implementation

[0097] The embodiments of the present invention are described in detail below: These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. It should be noted that those skilled in the art can make several modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.

[0098] One embodiment of the present invention provides a microgrid group information interaction and control system, comprising: a physical layer at the bottom, a network layer in the middle, and a service layer at the top; wherein:

[0099] The physical layer is used to describe the interconnection topology that makes up a microgrid group;

[0100] Based on the interconnected topology of the microgrid cluster, the network layer abstracts the corresponding information interaction objects into proxy nodes. It obtains local measurement data of the microgrid cluster through proxy nodes, conducts data interaction through private data negotiation, and identifies anomalies through consensus anomaly perception. At the same time, based on the local measurement data, it generates control strategies for energy scheduling of the microgrid cluster.

[0101] The service layer is used to realize information interaction with the network layer and obtain the required anomaly identification information, state estimation information (including node voltage data and network power distribution data), and control strategy information.

[0102] In a preferred embodiment, the microgrid group includes: a centralized physical topology microgrid group, a distributed physical topology microgrid group, and a hybrid physical topology microgrid group.

[0103] In a preferred embodiment, the network layer includes: a microgrid layer, a tie-line layer, and a distribution network layer; wherein:

[0104] The microgrid level includes internal nodes and boundary nodes. Internal nodes correspond to the structural units of the physical layer and are used to acquire the measurement data of the microgrid and perform state estimation calculations at the regional grid level (the results of the state estimation calculations are node voltage data and network power distribution data; therefore, voltage estimation and power estimation are both objectives of state estimation). Boundary nodes are used to pass the state estimation calculation results to the tie-line level.

[0105] At the tie-line level, while sensing the tie-line and receiving its own measurement data, it also receives voltage estimates from the boundary nodes at the microgrid level. It exchanges voltage estimate data between boundary nodes through private data negotiation and identifies anomalies through consensus anomaly detection.

[0106] At the distribution network level, voltage distribution data of each tie line is collected to obtain the voltage distribution of the entire network. This allows us to understand the operation of the entire distribution network and, based on the voltage distribution, provide control optimization targets for the network layer to implement energy dispatch control strategies for microgrid groups.

[0107] In a preferred embodiment, at the microgrid level, measurement data of the microgrid is acquired, and state estimation calculations are performed at the regional grid level, including:

[0108] We employ the Weighted Least Squares (WLS) algorithm as a parallel local state estimation algorithm for each region, considering the following measurement equations:

[0109] z = h(x) + e (35)

[0110] In the formula: z is the measurement vector in the sub-microgrid; x is the state variable vector in the sub-microgrid, with a dimension less than z; e is the measurement error vector, whose covariance matrix is ​​R, which is usually assumed to be a diagonal matrix, with the diagonal elements being the variance σ of each measurement data. 2 h(x) represents the functional relationship between the state variable vector and the measurement vector.

[0111] The objective of WLS is to find the optimal estimate that minimizes the following objective function.

[0112]

[0113] In the formula, J(x) is the objective function for the desired state estimation, and z i For the i-th measurement data, h i (x) is the i-th measurement function, R ii is the i-th diagonal element of the covariance matrix, and m is the number of measurement data;

[0114] Optimal estimate The solution can be obtained through the following loop:

[0115] G (p) (x (p) )Δx (p) =H T (x (p) W (p) [z (p) -h(x (p) (37)

[0116] In the formula, p represents the p-th iteration; Δx (p) =x (p+1) -x (p) ; The Jacobian matrix representing the measurement function h(x); G(p) =H T (x (p) )R -1 H(x (p) ), where is the gain matrix.

[0117] In a preferred embodiment, at the tie-line level, voltage estimation data exchange between boundary nodes is performed through private data negotiation, and anomaly identification is performed through consensus anomaly detection, including:

[0118] Private data negotiation employs a parallel computing approach across microgrid agents to exchange voltage estimation data at boundary nodes in encrypted form, including:

[0119] When the measurement data of microgrid k is updated, the microgrid level m k To its adjacent tie-line level l m Send the first encrypted information, which includes: measurement data related to the boundary nodes and extended nodes (internal nodes willing to share with the tie-line layer), topology information of the extended nodes, the micronet proxy address as the sending micronet layer and the tie-line proxy address as the receiving tie-line layer, and the current time T. start ;

[0120] The topology of the communication network of a microgrid cluster is represented as an undirected graph, where each sub-microgrid is considered a node in the graph, and the branches in the graph represent communication connections between sub-microgrids. Under normal circumstances, the undirected graph is connected, and its Laplace matrix L is obtained based on its topology. Let N... i Let |N| be the set of nodes that have a communicative connection with node i in an undirected graph. i | represents the number of nodes in the set, then the elements L of the Laplace matrix L are... ij Defined as:

[0121]

[0122] According to the finite-time consistency theory, for a Laplace matrix L of a connected graph, if there exist K independent non-zero eigenvalues ​​λ2≠λ3≠...≠λ K+1 If ≠0, then the values ​​at each node in the graph will reach the average value after K iterations, that is:

[0123]

[0124] The finite-time consensus algorithm iteration rule for each node in the graph is as follows:

[0125]

[0126] In the formula: v*(s) is the updated value of node * after the s-th iteration; ωii (s) and ω ij (s) is the weight factor at the s-th iteration;

[0127] The weight coefficients are updated according to the following formula during each iteration to ensure convergence after the Kth iteration:

[0128]

[0129] In the formula, λ s+1 Let be the (s+1)th independent non-zero eigenvalue of the Laplacian matrix L;

[0130] After receiving the first message, the tie-line level triggers the same parallel local estimation calculation as the microgrid level. The parallel local estimation calculation at the microgrid level uses weighted least squares, including:

[0131] Consider the following measurement equation:

[0132] z = h(x) + e (42)

[0133] In the formula: z is the measurement vector in the sub-microgrid area; x is the state variable vector in the sub-microgrid area, with a dimension less than z; e is the measurement error vector, whose covariance matrix is ​​R, which is a diagonal matrix, and the diagonal elements of the matrix are the variances σ of each measurement data. 2 h(x) represents the functional relationship between the state variable vector and the measurement vector.

[0134] Calculate the optimal estimate that minimizes the following objective function.

[0135]

[0136] In the formula, J(x) is the objective function for the desired state estimation, and z i For the i-th measurement data, h i (x) is the i-th measurement function, R ii is the i-th diagonal element of the covariance matrix, and m is the number of measurement data;

[0137] Optimal estimate The solution is obtained through the following loop:

[0138] G (p) (x (p) )Δx (p) =H T (x (p) W (p) [z (p) -h(x (p) (44)

[0139] In the formula: p represents the p-th iteration; Δx (p)=x (p+1) -x (p) ; The Jacobian matrix representing the measurement function h(x); G (p) =H T (x (p) )R -1 H(x (p) Let be the gain matrix. Inverting the gain matrix yields the estimation error covariance matrix of the state variables.

[0140]

[0141] After parallel computation is completed, the tie-line level sends a second encrypted message to the adjacent microgrid level. The second encrypted message includes: the estimated value of the boundary node (x = [v; θ], where v is the node voltage magnitude vector and θ is the node voltage phase angle vector), the tie-line proxy address as the sending tie-line level, the microgrid proxy address as the receiving microgrid level, and the current time T. end ;

[0142] Consensus anomaly detection includes:

[0143] When the microgrid layer receives the second encrypted information, it triggers consensus anomaly detection, performing data comparison and anomaly identification; among which:

[0144] After the microgrid layer receives the second encrypted information, it performs the following data consistency check to determine whether the consistency check is acceptable:

[0145]

[0146] In the formula: and These are micro-network agents m s and contact line agent l s The estimated value for the same boundary node i; For micro-network agent m s The local estimate of G is the i-th diagonal element; c is the proportionality coefficient. When c = 1, 2, 3, it indicates that the intervals falling into the standard Gaussian distribution are 68.2%, 95.4%, and 99.7% of the total interval, respectively.

[0147] After the data consistency check is completed, the micronet agent will draft the data consistency check requirements, including: the micronet agent and tie-line agent addresses, the micronet agent's estimate of the boundary nodes, the tie-line agent's estimate of the boundary nodes, the consistency status, and the time T when the first encrypted message was generated. start and the time T when the second encrypted information is generated endConsistency states include: Acceptance and Rejection; if a rejection in a consistency state comes from multiple neighboring micronet agents, the common contact line agent may be abnormal; if a rejection in a consistency state comes from a single micronet agent, that micronet agent may be abnormal.

[0148] For a microgrid agent with a consistency rejection status, query the absolute deviation Δx of the boundary node estimate between historical time t-1 and current time t. t,t-1 If Δx t,t-1 Exceeding the set threshold Normal max If Δx t,t-1 Exceeding the set threshold FC max If the value is less than 1, it is considered to be negative data; among which, the threshold Normal max and FC max The value of is obtained from the error of the largest estimate in the historical data without anomalies, where:

[0149]

[0150]

[0151] According to equations (61) and (62), we get:

[0152]

[0153] In the formula: These are estimates for scenarios with bad data, environmental noise, data latency, and topological changes, respectively. is the true value; g(·) is the functional relationship between the measured value and the state estimate; z1 is the measured value under normal conditions of sub-microgrid area 1. Y1, Y2, and ΔY1 represent the measurement deviation values ​​for bad data, environmental noise, data latency, and topology changes, respectively. Y1, Y2, and ΔY1 represent the topology matrix of sub-microgrid 1, the topology matrix of sub-microgrid 2, and the change in the topology matrix when a topology change occurs in sub-microgrid 1, respectively.

[0154] Warnings about specific anomalies and their locations will be sent to the relevant microgrid area, and error data analysis or topology checks will be initiated.

[0155] Once the abnormal data comparison is complete, the abnormal identification information will be communicated to each layer of the network layer and broadcast to all nodes.

[0156] In this embodiment of the invention, the sub-micronet area is a sub-micronet of the physical layer, corresponding to the micronet level of the network layer, and can be regarded as a proxy node.

[0157] In a preferred embodiment, the network layer employs a distribution network-level coordinated control strategy, an inter-regional control strategy, and a microgrid-based control strategy to achieve energy dispatching of the microgrid group (specifically, energy dispatching refers to achieving voltage stability and power balance of the microgrid group); wherein:

[0158] The distribution network hierarchical coordination control strategy collects voltage estimation data for each tie line through cooperation at other levels to obtain the overall operation information of the distribution network. A global optimization algorithm is then used to perform global-level optimization control on the distribution network and microgrid clusters, setting regional optimization targets for lower-level areas (the optimization target being the switching power of the sub-microgrids). This guides all controllable distributed generation units, flexible load units, and interconnection switches on feeders within the entire distribution network, and sets regional optimization targets within the microgrid group (the optimization target is the target value of the interconnection line output power). Coordination coefficient k of the connecting line i );

[0159] Inter-regional control strategy: All controllable distributed power sources and energy storage units within an autonomous region on a feeder are coordinated and controlled at the tie-line level. The hierarchical distributed controller at the network layer achieves regional coordinated control by controlling the various source-network coordination controllers within its jurisdiction.

[0160] The microgrid control strategy divides the control structure of the sub-microgrid into two layers. The first layer adopts a droop control strategy based on equal-cost micro-increment to achieve economical distribution of active power within the microgrid (i.e., proportional distribution). The second layer adopts a microgrid DG control strategy based on the consistency principle to control the information interaction between distributed power sources on the same distribution node, thereby achieving secondary adjustment of the microgrid frequency, voltage, and reactive power, and avoiding problems such as unbalanced active power distribution and system instability caused by frequency instability. This achieves voltage stability and power balance of the microgrid group and completes the energy dispatch of the microgrid group.

[0161] In a preferred embodiment, the distribution network-level coordination control strategy is specifically as follows:

[0162] Distribution network coordination and control is primarily implemented at the distribution network level. The distribution network level should monitor the regional switching power indicators in real time and conduct real-time assessments of the status of each microgrid area.

[0163] P i microgrid (t)=P i microgrid-in (t)-P i microgrid-out (t) (50)

[0164] In the formula, P i microgrid(t) represents the switching power of microgrid i at time t; P i microgrid-in (t) represents the power flowing into microgrid i at time t; P i microgrid-out (t) represents the power outflow value of microgrid i at time t.

[0165] At fixed intervals, the global optimization target value is calculated based on the current state of the distribution network and the state assessment results. This target value is then decomposed and distributed to each control area (i.e., each level). The control area responds to external disturbances in real time according to specific control strategies (i.e., the control strategies corresponding to each level), and coordinates the distributed energy resources within the control area to minimize the gap between each control area and the global target. As a result, the operation of the entire system is closer to the global target optimization value.

[0166] In a preferred embodiment, the inter-regional control strategy is specifically as follows:

[0167] Distribution network coordination and control is primarily implemented at the distribution network level. The distribution network level should monitor the regional switching power indicators in real time and conduct real-time assessments of the status of each microgrid area.

[0168] P i microgrid (t)=P i microgrid-in (t)-P i microgrid-out (t) (51)

[0169] In the formula, P i microgrid (t) represents the switching power of microgrid i at time t; P i microgrid-in (t) represents the power flowing into microgrid i at time t; P i microgrid-out (t) represents the power outflow value of microgrid i at time t. The global optimization management system calculates the global optimization target value based on the current and predicted state of the power grid at fixed intervals, and then decomposes it and distributes the local target value to each control area. The control area responds to external disturbances in real time according to specific control strategies, coordinates the distributed energy within the control area, and minimizes the gap between each control area and the global target, so that the operation of the entire system is closer to the global target optimization value.

[0170] In a preferred embodiment, the microgrid control strategy is specifically as follows:

[0171] Microgrid control is primarily implemented at the microgrid level, controlling the active and reactive power outputs of each controllable distributed generation (DG). To achieve microgrid control objectives, the sub-microgrid control structure can be divided into two layers. Primary control employs a droop control strategy based on equal-cost incremental increases to achieve economical allocation of active power within the microgrid. Secondary control, based on a consistency-principle-based microgrid DG control strategy, utilizes information exchange between distributed generation sources to achieve secondary adjustments to the microgrid's frequency, voltage, and reactive power.

[0172] In the primary control, distributed droop control is used to achieve proportional power distribution. However, since droop control is a differential control, it will produce frequency deviation. Therefore, in the secondary control, the frequency deviation will be further corrected.

[0173] An embodiment of the present invention also provides a method for information interaction and control of microgrid groups, comprising the following steps:

[0174] S100: Based on the interconnection topology of the microgrid group, acquire local measurement data of the microgrid group;

[0175] S200 performs interactive information calculations based on local measurement data, conducts data interaction through private data negotiation, and identifies anomalies through consensus-based anomaly detection.

[0176] S300, the business layer, is used to implement three functions in the interaction mechanism: anomaly identification, state estimation, and control information interaction.

[0177] It should be noted that the steps in the method provided by the present invention can be implemented using the corresponding structure in the system. Those skilled in the art can refer to the technical solution of the system to implement the steps of the method. That is, the embodiments in the system can be understood as preferred examples of the method, and will not be elaborated here.

[0178] The technical solutions provided by the above embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0179] The microgrid group information interaction and control system provided in the above embodiments of the present invention can be modeled as three layers, namely, physical layer, network layer, and service layer, such as... Figure 1 As shown.

[0180] (1) Physical layer

[0181] The physical layer describes the interconnection topology of the microgrid cluster. The microgrid cluster information interaction and control system provided in the above embodiments of the present invention is applicable to centralized, distributed, and hybrid physical topology microgrid clusters. A centralized physical topology allows a central energy manager to centrally process the interaction information of each microgrid; however, due to the increased number of microgrids connected, information congestion may occur, preventing timely fulfillment of power system dispatch requirements. Distributed physical topology microgrid clusters can be interconnected through certain connection methods. When there is a supply-demand imbalance or fault in the microgrid, energy can be effectively redistributed through internal energy management, ensuring stable operation of the microgrid cluster. The hybrid physical topology combines the advantages of both centralized and distributed systems, alleviating information congestion issues and effectively protecting microgrid data privacy, achieving distributed data protection. However, all three structural methods are essentially interconnections of microgrid clusters, and therefore can be abstracted as topology interconnections at the network layer, with microgrids as the objects.

[0182] (2) Network layer

[0183] The network layer abstracts the corresponding information interaction objects as proxy nodes for the structural units that make up the microgrid cluster. The structure of the abstracted information interaction objects is as follows: Figure 2 As shown, the information interaction structure presents a tree-like structure, which can be divided into three levels according to the scope of information interaction: distribution network level, tie line level, and microgrid level.

[0184] For microgrid level m k There are: (1) Internal nodes: All adjacent nodes of an internal node are in the sub-microgrid area, and its state variable vector is denoted as x. ik (2) Boundary node: A node in the microgrid level that is connected to the tie-line level, and its state variable vector is denoted as x. bk The microgrid agent acquires measurement data within the microgrid, including active power, reactive power, current, and voltage, performs regional grid-level state estimation, and returns the voltage estimation data of the boundary nodes to the boundary nodes, thereby transmitting it to the tie-line level.

[0185] In this embodiment of the invention, the internal nodes of the microgrid level refer to specific physical nodes that can be connected to the DG, and an entire microgrid corresponds to one agent.

[0186] For the tie-line level l m While the tie-line agent obtains its own measurement data at the induction tie-line, it also collects voltage estimates from the boundary nodes of adjacent microgrid levels, performs corresponding data exchange and private data negotiation processes, and completes the mutual comparison of data to complete anomaly identification.

[0187] At the distribution network level, the distribution network agent collects voltage distribution data for each tie line, and can eventually obtain the voltage distribution of the entire network, thereby effectively scheduling energy for the entire microgrid group.

[0188] (3) Business layer

[0189] The business layer is used to implement three functions in the interaction mechanism: anomaly identification, state estimation, and control information exchange. The microgrid group information interaction and control system aims to achieve hierarchical information exchange between microgrid groups, enabling coordinated control management and interactive operation between the microgrid and distribution network levels. In data acquisition, it ensures the observability of the microgrid group, and on this basis, protects data privacy within the microgrid group, ensuring data accuracy and real-time performance, which is reflected in the data within the real-time sensing area. The hierarchical microgrid group information interaction and control system is also beneficial for application in large-scale microgrid groups, improving the flexibility and economy of microgrid operation and control, and fully realizing the potential of microgrids.

[0190] The specific process of the information calculation and interaction method of the microgrid group information interaction and control system provided in the above embodiments of the present invention is described as follows.

[0191] In a microgrid cluster, each sub-microgrid only communicates with its adjacent sub-microgrids. Using a control center for information exchange would not only increase the construction and maintenance costs of communication but also compromise information privacy. Therefore, the above embodiments of this invention provide a reliable information exchange method, enabling each sub-microgrid to achieve reliable information exchange while ensuring information privacy. Microgrid clusters involve dynamic physical (power exchange, physical connections) and informational (measurement data, node voltage state estimation data, time information data, network anomaly identification data, control information) interactions among multiple microgrids; therefore, a hierarchical and partitioned information exchange method can be established. The information calculation and interaction method provided in the above embodiments of this invention mainly includes the following three processes: private data negotiation, consensus anomaly detection, and control information interaction, such as... Figure 3 As shown. First, for the microgrid layer m k Specifically, regional measurement updates trigger the local state estimation module of the microgrid and initiate a private data negotiation process with adjacent tie-line levels to exchange measurement data and boundary value estimation data. The results of the data exchange will proceed to the consensus anomaly detection phase, where network topology anomalies and bad data are identified. Microgrid areas identified as having anomalies will have their anomaly information recorded and relevant sub-microgrid areas will be alerted, proactively requesting bad data detection and updating state data. Control information interaction can employ hierarchical and partitioned control information interaction, reading and adjusting real-time updated state data to achieve effective interaction of information flow and control flow within the microgrid group.

[0192] (1) Privacy data negotiation

[0193] To effectively protect data privacy within the microgrid area and to accelerate decentralized computing efficiency, a framework of parallel computing by each microgrid agent is adopted, and boundary data is exchanged in the form of encrypted data, such as... Figure 4 As shown.

[0194] When the measurement of microgrid k is updated, the microgrid level m k To its adjacent tie-line level l m Send encrypted message 1, where encryption involves a unique digital signature and authentication, and may employ an asymmetric encryption algorithm. Message 1 contains: (1) measurement data (node ​​injected active power / node injected reactive power / voltage amplitude / measurement error) related to the boundary nodes and extended nodes (internal nodes willing to share with the tie-line level); (2) topology information willing to share with the extended nodes at the tie-line level; (3) the microgrid proxy address as the sending terminal microgrid area and the tie-line proxy address as the receiving terminal tie-line area; (4) the current time T. start .

[0195] The topology of the communication network of a microgrid cluster can be represented as an undirected graph, where each sub-microgrid is considered a node, and the branches in the graph represent communication connections between sub-microgrids. Under normal circumstances, this undirected graph is connected, and its Laplace matrix L can be obtained based on its topology. Let N... i Let |N| be the set of nodes that have a communication connection with node i. i | represents the number of nodes in the set, then the elements L of the Laplace matrix are... ij Defined as:

[0196]

[0197] According to the finite-time consistency theory, for a Laplace matrix L of a connected graph, if there exist K independent non-zero eigenvalues ​​λ2≠λ3≠...≠λ K+1 If ≠0, then the values ​​at each node in the graph will reach the average value after K iterations, that is:

[0198]

[0199] The iterative rule for the finite-time consensus algorithm for each node in the network is as follows:

[0200]

[0201] In the formula: v i (s) represents the updated value of node i after the s-th iteration; ω ii (s) and ω ij (s) is the weight factor at the s-th iteration.

[0202] The weight coefficients are updated according to the following formula during each iteration to ensure convergence after the Kth iteration.

[0203]

[0204] In the formula λ s+1 Let be the (s+1)th independent non-zero eigenvalue of the Laplace matrix L.

[0205] Time consistency algorithms can still transmit information effectively even when the communication topology changes.

[0206] Upon receiving message 1, the tie-line area will trigger the same parallel Local Estimation (LE) calculation as the microgrid area. Weighted least squares is used here as the parallel LE algorithm within the microgrid.

[0207] Consider the following measurement equation:

[0208] z = h(x) + e (56)

[0209] In the formula: z is the measurement vector in the sub-microgrid; x is the state variable vector in the sub-microgrid, with a dimension less than z; e is the measurement error vector, whose covariance matrix is ​​R, which is usually assumed to be a diagonal matrix, with the diagonal elements being the variance σ of each measurement data. 2 h(x) represents the functional relationship between the state variable vector and the measurement vector.

[0210] The objective of WLS (Weighted Least Squares) is to find the optimal estimate that minimizes the following objective function.

[0211]

[0212] Optimal estimate The solution can be obtained through the following loop:

[0213] G (p) (x (p) )Δx (p) =H T (x (p) W (p) [z (p) -h(x (p) (58)

[0214] In the formula: p represents the p-th iteration; Δx (p) =x (p+1) -x (p) ; G is the Jacobian matrix representing the measurement function h(x); (p) =H T (x(p) )R -1 H(x (p) ), is the gain matrix, and inverting it yields the estimation error covariance matrix of the state variables:

[0215]

[0216] After parallel computation is completed, the tie-line agent will send encrypted information (Information 2) to the neighboring micronet agent. Information 2 includes: (1) the estimated value of the boundary node (node ​​voltage amplitude / phase angle / estimation error); (2) the address of the tie-line area as the sending end and the address of the micronet area as the receiving end; (3) the current time T. end .

[0217] After receiving message 2, the micro-network agent triggers the next module, namely consensus anomaly detection, to perform data comparison and anomaly identification.

[0218] (2) Consensus Anomaly Detection

[0219] After receiving message 2, the micro-network agent triggers the next module, consensus anomaly detection, to perform data comparison and anomaly identification, such as... Figure 5 As shown.

[0220] 1) After receiving information 2, the microgrid needs to perform the following data consistency check (criterion 1) to determine whether the consistency check is acceptable:

[0221]

[0222] In the formula: and These are micro-network agents m s and contact line agent l s The estimated value for the same boundary node i; It is m s The local estimate of G is the i-th diagonal element; c is the proportionality coefficient, and when c = 1, 2, 3, it indicates that the intervals falling into the standard Gaussian distribution are 68.2%, 95.4%, and 99.7% of the total interval, respectively. Simulation results show that c = 1 can achieve an acceptable balance between recognition accuracy and anomaly threshold confusion.

[0223] After Criterion 1 is completed, the microgrid agent will draft data consistency check requirements, including the following information: (1) Microgrid agent and tie-line agent addresses; (2) Microgrid agent's estimates of the boundary nodes (node ​​voltage amplitude / phase angle / estimation error); (3) Tie-line agent's estimates of the boundary nodes (node ​​voltage amplitude / phase angle / estimation error); (4) Consistency status (i.e., 0 for acceptance, 1 for rejection); (5) T in the first encrypted information (Information 1). start T in the second encrypted information (information 2)end .

[0224] 2) Address information from data consistency rejections can help locate anomalies. If the consistency rejection comes from multiple neighboring micronet agents, the common contact line agent is suspicious; if only one sub-micronet agent rejects the consistency check, the anomaly may only occur in that micronet agent.

[0225] Since a proportionality coefficient of 1 in criterion 1 only covers 68.2% of the interval, environmental noise may cause criterion 1 to be rejected. To identify bad data and environmental noise, the absolute deviation Δx of the boundary estimate between the historical time t-1 and the current time t of the rejected microgrid query is used. t,t-1 If Δx t,t-1 Exceeding the threshold Normal max The data is identified as being influenced by bad data; the threshold FC is exceeded. max This is determined to be due to bad data. Threshold Normal max FC max The value of can be obtained from the error of the largest estimate in the historical data without anomalies, as follows:

[0226]

[0227]

[0228] According to (61) and (62) there are

[0229]

[0230] In the above formulas: These are estimates for scenarios with bad data, environmental noise, data latency, and topological changes, respectively. is the true value; g(·) is the functional relationship between the measured value and the state estimate; z1 is the measured value under normal conditions of sub-microgrid area 1. This represents the measurement deviation due to bad data, environmental noise, data latency, and topology changes.

[0231] Warnings about specific anomalies and their locations will be sent to the relevant microgrid area, and error data analysis or topology checks will be initiated.

[0232] 3) After the abnormal data comparison is completed, the abnormal identification information will be notified to each layer of the network layer and broadcast to all nodes.

[0233] (3) Control information interaction

[0234] Following the aforementioned process, the microgrid cluster achieves real-time data perception within its region. To ensure the realization of an interactive mechanism for large-scale microgrids, enhance the flexibility and economy of microgrid operation and control, and fully realize the potential of microgrids, effective hierarchical and zoned control information exchange is also required. Adopting a hierarchical and zoned control approach diverts some communication channels, primarily addressing two needs: ① active control needs; ② monitoring needs. Active control needs mainly include information exchange with distributed power sources, energy storage systems, lower-level controllers, and distribution network switches to facilitate a multi-level, graded absorption mode for the microgrid cluster. Monitoring needs mainly include information exchange with distribution network power flow and other information requiring interface display terminals to help dispatchers understand the microgrid cluster's operating status through information exchange and respond promptly to anomalies.

[0235] Actively controlling demand includes:

[0236] (1) Distribution network hierarchical coordination control strategy

[0237] Active distribution network coordination control, as the first-level control, collects the operation information of the entire distribution network through cooperation with other levels. It performs global optimization control on the distribution network (upper-level grid) and microgrid groups through global optimization algorithms, sets regional optimization targets for lower-level areas, guides the optimization control of all controllable distributed generation units, flexible load units and interconnection switches on feeders, and sets regional optimization targets within the microgrid.

[0238] Specifically:

[0239] Distribution network coordination and control is primarily implemented at the distribution network level. The distribution network level should monitor the regional switching power indicators in real time and conduct real-time assessments of the status of each microgrid area.

[0240] P i microgrid (t)=P i microgrid-in (t)-P i microgrid-out (t) (64)

[0241] In the formula, P i microgrid (t) represents the switching power of microgrid i at time t; P i microgrid-in (t) represents the power flowing into microgrid i at time t; P i microgrid-out(t) represents the power outflow value of microgrid i at time t. The global optimization management system calculates the global optimization target value based on the current and predicted state of the power grid at fixed intervals, and then decomposes it and distributes the local target value to each control area. The control area responds to external disturbances in real time according to specific control strategies, coordinates the distributed energy within the control area, and minimizes the gap between each control area and the global target, so that the operation of the entire system is closer to the global target optimization value.

[0242] (2) Inter-regional control strategy

[0243] The second level of control is local area control. The distribution network (upper-level grid) is interconnected with a group of microgrids, which dispatches lower-level units within the region. The tie-line level needs to achieve internal stability within its connected areas and reduce mutual influence between different areas, while minimizing disturbances to the distribution network. The tie-line level controller, as the second-level control unit, coordinates all controllable distributed generation and energy storage units within an autonomous area on the feeder. The hierarchical distributed controller achieves regional coordinated control by controlling the various source-grid coordination controllers within its jurisdiction.

[0244] Specifically:

[0245] Inter-regional control is achieved through the tie-line level. The tie-line level needs to achieve internal stability within its connected regions and reduce mutual interference between different regions, while minimizing disturbances to the distribution network. The tie-line level controller, as a second-level control unit, coordinates all controllable distributed generation and energy storage units within an autonomous region on a feeder. The hierarchical distributed controller achieves regional coordinated control by controlling the various source-grid coordination controllers within its jurisdiction. The tie-line level forms the objective function for controlling sub-microgrid i:

[0246]

[0247] In the formula: P FCEi For the tie-line control error of sub-micronet i; k i ΔP is the coordination coefficient of the tie line containing sub-micronet i; F P is the difference between the actual tie-line output power and its target value. F The current output power of the tie line (positive for power flowing from the busbar into the feeder); The target value for the tie-line output power; ΔP microgrid-i P is the difference between the actual power exchange value of sub-microgrid i and its optimization target value; microgrid-i The actual value of the switching power of sub-micronet i at the current moment; Let k be the target value for the switching power of sub-microgrid i. i , These are all target values ​​obtained after calculation by a global optimization algorithm, and are managed by global optimization.

[0248] The system calculates and distributes the data at fixed intervals, and the inter-regional connection layer unit updates the data simultaneously after receiving the three target values.

[0249] (3) Microgrid control strategy

[0250] The third layer of control is microgrid-level control, which controls the active and reactive power output of each controllable DG (distributed generation). To achieve the microgrid control objectives, the sub-microgrid control structure can be divided into two layers. The primary control employs a droop control strategy based on equal-cost incremental increases to achieve economical distribution of active power within the microgrid. The secondary control, through information exchange between distributed generation sources, achieves secondary adjustments to the microgrid's frequency, voltage, and reactive power, avoiding imbalances in active power distribution and system instability caused by frequency instability. Therefore, a microgrid DG control strategy based on the consistency principle can be constructed to control distributed generation sources on the same distribution node.

[0251] like Figure 6 As shown, the control flow within this microgrid is as follows:

[0252] Microgrid control is primarily implemented at the microgrid level, controlling the active and reactive power outputs of each controllable distributed generation (DG). To achieve microgrid control objectives, the sub-microgrid control structure can be divided into two layers. Primary control employs a droop control strategy based on equal-cost incremental increases to achieve economical allocation of active power within the microgrid. Secondary control, based on a consistency-principle-based microgrid DG control strategy, utilizes information exchange between distributed generation sources to achieve secondary adjustments to the microgrid's frequency, voltage, and reactive power.

[0253] In the primary control, distributed droop control is used to achieve proportional power distribution. However, since droop control is a differential control, it will produce frequency deviation. Therefore, in the secondary control, the frequency deviation will be further corrected.

[0254] An embodiment of the present invention also provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it can be used to run any of the systems described in the above embodiments, or to execute any of the methods described in the above embodiments.

[0255] An embodiment of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, can be used to run any of the systems described in the above embodiments, or to execute any of the systems described in the above embodiments.

[0256] Optionally, the memory is used to store programs; the memory may include volatile memory, such as random-access memory (RAM), such as static random-access memory (SRAM), double data rate synchronous dynamic random-access memory (DDR SDRAM), etc.; the memory may also include non-volatile memory, such as flash memory. The memory is used to store computer programs (such as application programs, functional modules, etc. that implement the above methods), computer instructions, etc., and the aforementioned computer programs, computer instructions, etc., can be partitioned and stored in one or more memories. Furthermore, the aforementioned computer programs, computer instructions, data, etc., can be accessed by the processor.

[0257] The aforementioned computer programs, computer instructions, etc., can be stored in partitions within one or more memory locations. Furthermore, the aforementioned computer programs, computer instructions, data, etc., can be accessed by a processor.

[0258] A processor is used to execute a computer program stored in memory to implement the various steps of the methods involved in the above embodiments. For details, please refer to the relevant descriptions in the preceding method embodiments.

[0259] The processor and memory can be separate structures or integrated structures. When the processor and memory are separate structures, they can be coupled together via a bus.

[0260] The microgrid group information interaction and control system and method provided in the above embodiments of the present invention fully utilize user data within the microgrid (i.e., local measurement data of the microgrid group), resolving the contradiction between the observability of the microgrid group and the privacy of user data. Furthermore, while protecting privacy, it only exchanges and stores boundary values, reducing transmission and storage requirements. Employing a decentralized framework, it utilizes a parallel information interaction mechanism between the distribution network level, tie-line level, and microgrid level, allowing for flexible partitioning and parallel computation among microgrids, effectively alleviating computational pressure. It is also compatible with communication devices with different measurement sampling rates, triggering private data negotiation and consensus anomaly detection based on measurement data, achieving more flexible microgrid group information interaction. It is scalable; this interaction mechanism is applicable to large-scale microgrid groups, effectively improving the flexibility and economy of microgrid operation control through distributed state estimation, and maximizing the potential of the microgrid. The above embodiments of the present invention enable each sub-microgrid to achieve reliable information interaction while ensuring information privacy, effectively responding to network anomalies, improving the ability to observe state data, and laying the foundation for subsequent microgrid group energy management and scheduling.

[0261] The microgrid group information interaction and control system and method provided in the above embodiments of the present invention

[0262] Any matters not covered in the above embodiments of the present invention are well-known in the art.

[0263] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various modifications or variations within the scope of the claims, which do not affect the essence of the present invention.

Claims

1. A microgrid group information interaction and control system, characterized in that, include: Physical layer, network layer, and service layer; among which: The physical layer is used to describe the interconnection topology that makes up the microgrid group; Based on the interconnected topology of the microgrid cluster, the network layer abstracts the corresponding information interaction objects into proxy nodes, obtains local measurement data of the microgrid cluster through the proxy nodes, performs data interaction through private data negotiation, and identifies anomalies through consensus anomaly detection; at the same time, based on the local measurement data, it generates a control strategy for energy scheduling of the microgrid cluster. The service layer is used to realize information interaction with the network layer and obtain the required anomaly identification information, state estimation information and control strategy information; The network layer employs a distribution network-level coordinated control strategy, an inter-regional control strategy, and a microgrid-internal control strategy to achieve energy dispatching of the microgrid group; wherein: The distribution network hierarchical coordination control strategy collects voltage estimation data for each tie line to obtain the operation information of the entire distribution network. It performs global optimization control on the distribution network and microgrid group through a global optimization algorithm, sets regional optimization targets for the lower-level area, namely the exchange power of the sub-microgrid, guides all controllable distributed generation units, flexible load units and tie switches on feeders in the entire distribution network, and sets regional optimization targets within the microgrid group, namely the target value of tie line output power and tie line coordination coefficient. The inter-regional control strategy coordinates all controllable distributed power sources and energy storage units within an autonomous region on a feeder through a tie-line hierarchy. The microgrid control strategy divides the control structure of the sub-microgrid into two layers. The first layer adopts a droop control strategy based on equal cost micro-increment to realize the proportional distribution of active power within the microgrid. The second layer adopts a microgrid DG control strategy based on the consistency principle to control the information interaction between distributed power sources on the same distribution node, realize the secondary adjustment of microgrid frequency, voltage, and reactive power, achieve voltage stability and power balance of the microgrid group, and complete the energy dispatch of the microgrid group.

2. The microgrid group information interaction and control system according to claim 1, characterized in that, The microgrid clusters include: centralized physical topology microgrid clusters, distributed physical topology microgrid clusters, and hybrid physical topology microgrid clusters.

3. The microgrid group information interaction and control system according to claim 1, characterized in that, The network layer includes: a microgrid layer, a tie-line layer, and a distribution network layer; wherein: The microgrid layer includes: internal nodes and boundary nodes; the internal nodes correspond to the structural units of the physical layer and are used to acquire the measurement data of the microgrid and perform state estimation calculations at the regional grid level; the boundary nodes are used to transmit the calculation results of the state estimation to the tie-line layer. The tie-line level receives its own measurement data at the induction tie-line, and at the same time receives voltage estimates from the boundary nodes of the microgrid level. It exchanges voltage estimate data between boundary nodes through private data negotiation and identifies anomalies through consensus anomaly detection. The distribution network level collects voltage distribution data from each tie-line level to obtain the voltage distribution of the entire network. Based on the voltage distribution, it provides control optimization targets for the network layer to implement control strategies for energy dispatching of the microgrid group.

4. The microgrid group information interaction and control system according to claim 3, characterized in that, In the microgrid layer, measurement data of the microgrid is acquired, and state estimation calculations at the regional grid level are performed, including: We employ a weighted least squares algorithm as the parallel local state estimation algorithm for each region, considering the following measurement equation: z = h(x) + e (1) In the formula: z is the measurement vector in the sub-microgrid area; x is the state variable vector in the sub-microgrid area, with a dimension less than z; e is the measurement error vector, and its covariance matrix is ​​R. The covariance matrix R is set as a diagonal matrix, and the diagonal elements are the variances σ of each measurement data. 2 h(x) represents the functional relationship between the state variable vector and the measurement vector. The objective of the weighted least squares algorithm is to find the optimal estimate that minimizes the following objective function: In the formula, J(x) is the objective function for the desired state estimation, and z i For the i-th measurement data, h i (x) is the i-th measurement function, R ii is the i-th diagonal element of the covariance matrix, and m is the number of measurement data; The optimal estimate is obtained by solving the following equation iteratively: G (p) (x (p) )Δx (p) =H T (x (p ))R -1 [z-h(x (p) )] (3) In the formula, p represents the p-th iteration; Δx (p) =x (p+1) -x (p) ; The Jacobian matrix representing the measurement function h(x); G (p) =H T (x (p) )R -1 H(x (p) ), where is the gain matrix.

5. The microgrid group information interaction and control system according to claim 3, characterized in that, In the aforementioned tie-line layer, voltage estimation data exchange at boundary nodes is conducted through private data negotiation, and anomaly identification is performed through consensus anomaly detection, including: The private data negotiation employs a parallel computing approach across microgrid agents to exchange voltage estimation data at boundary nodes in encrypted form, including: When the measurement data of microgrid k is updated, the microgrid level m k To its adjacent tie-line level l m Send first encrypted information, which includes: measurement data related to the boundary node and the extended node, topology information of the extended node, micronet proxy address as the sending micronet layer and tie-line proxy address as the receiving tie-line layer, and the current time T. start Among them, extended nodes are internal nodes that are willing to share with the tie-line level; The topology of the communication network of the microgrid cluster is represented as an undirected graph, where each sub-microgrid is considered a node in the graph, and the branches in the graph represent communication connections between sub-microgrids. Under normal circumstances, the undirected graph is connected, and its Laplace matrix L is obtained based on its topology. Let N... i Let |N| be the set of nodes that have a communicative connection with node i in an undirected graph. i | represents the number of nodes in the set, then the elements L of the Laplace matrix L are... ij Defined as: According to the finite-time consistency theory, for a Laplace matrix L of a connected graph, if there exist K independent non-zero eigenvalues ​​λ2≠λ3≠...≠λ K+1 If ≠0, then the values ​​at each node in the graph will reach the average value after K iterations, that is: The finite-time consensus algorithm iteration rule for each node in the graph is as follows: In the formula: v * (s) represents the updated value of node * after the s-th iteration; ω ii (s) and ω ij (s) is the weight factor at the s-th iteration; The weight coefficients are updated according to the following formula during each iteration to ensure convergence after the Kth iteration: In the formula, λ s+1 Let be the (s+1)th independent non-zero eigenvalue of the Laplacian matrix L; After receiving the first information, the tie-line layer triggers the same parallel local estimation calculation as the microgrid layer. The parallel local estimation calculation at the microgrid layer uses weighted least squares and includes: Consider the following measurement equation: z = h(x) + e (8) In the formula: z is the measurement vector in the sub-microgrid area; x is the state variable vector in the sub-microgrid area, with a dimension less than z; e is the measurement error vector, whose covariance matrix is ​​R, wherein the covariance matrix R is a diagonal matrix, and the diagonal elements in the matrix are the variances σ of each measurement data. 2 h(x) represents the functional relationship between the state variable vector and the measurement vector. Calculate the optimal estimate x that minimizes the following objective function: In the formula, J(x) is the objective function for the desired state estimation, and z i For the i-th measurement data, h i (x) is the i-th measurement function, R ii is the i-th diagonal element of the covariance matrix, and m is the number of measurement data; Optimal estimate The following solution is obtained through a loop: G (p) (x (p) )Δx (p) =H T (x (p) )W (p) [z (p) -h(x (p) )] (10) In the formula: p represents the p-th iteration; Δx (p) =x (p+1) -x (p) ; The Jacobian matrix representing the measurement function h(x); G (p) =H T (x (p) )R -1 H(x (p) Let be the gain matrix. Inverting the gain matrix yields the estimation error covariance matrix of the state variables. After parallel computation is completed, the tie-line layer sends a second encrypted message to the adjacent microgrid layer. The second encrypted message includes: the estimated value of the boundary node x = [v; θ], the tie-line proxy address of the sending tie-line layer and the microgrid proxy address of the receiving microgrid layer, and the current time T. end Where v is the node voltage magnitude vector and θ is the node voltage phase angle vector; The consensus anomaly detection includes: When the microgrid layer receives the second encrypted information, it triggers consensus anomaly detection, performing data comparison and anomaly identification; wherein: After the microgrid layer receives the second encrypted information, it performs the following data consistency check to determine whether the consistency check is acceptable: In the formula: and These are micro-network agents m s and contact line agent l s The estimated value for the same boundary node i; For micro-network agent m s The local estimate of G is the i-th diagonal element; c is the proportionality coefficient. When c = 1, 2, 3, it indicates that the intervals falling into the standard Gaussian distribution are 68.2%, 95.4%, and 99.7% of the total interval, respectively. After the data consistency check is completed, the micronet agent will draft the data consistency check requirements, including: the micronet agent and tie-line agent addresses, the micronet agent's estimate of the boundary nodes, the tie-line agent's estimate of the boundary nodes, the consistency status, and the time T when the first encrypted message was generated. start and the time T when the second encrypted information is generated end The consistency states include: acceptance and rejection; if the rejection in the consistency state comes from multiple adjacent micronet agents, the common contact line agent may be abnormal; if the rejection in the consistency state is from a single micronet agent, that micronet agent may be abnormal. For a microgrid agent with a consistency rejection status, query the absolute deviation Δx of the boundary node estimate between historical time t-1 and current time t. t,t-1 If Δx t,t-1 Exceeding the set threshold Normal max If Δx t,t-1 Exceeding the set threshold FC max If the value is less than 1, it is considered to be negative data; among which, the threshold Normal max and FC max The value of is obtained from the error of the largest estimate in the historical data without anomalies, where: According to equations (47) and (48), we get: In the formula: These are estimates for scenarios with bad data, environmental noise, data latency, and topological changes, respectively. is the true value; g(·) is the functional relationship between the measured value and the state estimate; z1 is the measured value under normal conditions of sub-microgrid area 1. These represent the measurement deviation values ​​for bad data, environmental noise, data latency, and topology changes, respectively. Y1, Y2, and ΔY1 are the topology matrices of the first sub-microgrid, the second sub-microgrid, and the change in the topology matrix when the first sub-microgrid undergoes a topology change, respectively. Warnings about specific anomalies and their locations will be sent to the relevant microgrid area, and error data analysis or topology checks will be initiated. Once the abnormal data comparison is complete, the abnormal identification information will be communicated to each layer of the network layer and broadcast to all nodes.

6. The microgrid group information interaction and control system according to claim 1, characterized in that, The distribution network hierarchical coordination control strategy includes: The distribution network-level coordinated control is achieved through the distribution network level, which monitors the regional exchange power index of the distribution network in real time and evaluates the status of each microgrid region in real time. P i microgrid (t)=P i microgrid-in (t)-P i microgrid-out (t) (16) In the formula, P i microgrid (t) represents the power exchanged by sub-microgrid i at time t; P i microgrid-in (t) represents the power flowing into sub-microgrid i at time t; P i microgrid-out (t) represents the power outflow value of sub-microgrid i at time t; At fixed intervals, the global optimization target value is calculated based on the current state of the distribution network and the state assessment results. After being decomposed, the regional optimization target value is issued to the control areas of each level. The control areas respond to external disturbances in real time according to their corresponding control strategies, and coordinate the distributed energy within the control areas to minimize the gap between their respective control areas and the global target. As a result, the operation of the entire distribution network is closer to the global target optimization value.

7. The microgrid group information interaction and control system according to claim 1, characterized in that, The inter-regional control strategy includes: The inter-regional control is achieved through a tie-line hierarchy, which coordinates all controllable distributed generation and energy storage units within an autonomous region on a control feeder, forming the objective function for controlling sub-microgrid i: In the formula: P FCEi For the tie-line control error of sub-microgrid i; k i ΔP is the coordination factor of the tie line where sub-microgrid i is located; F P is the difference between the actual tie-line output power and its target value. F This represents the current output power of the tie line, where the power flowing from the busbar into the feeder is positive. The target value for the tie-line output power; ΔP microgrid-i P is the difference between the actual power exchange value and its optimized target value of sub-microgrid i; microgrid-i This represents the actual value of the exchange power of sub-microgrid i at the current moment; Let k be the target value of the switching power of sub-microgrid i; where k i , These are the global optimization target values ​​calculated and issued by the distribution network level at fixed intervals in the distribution network level coordination control, respectively. The tie line level updates simultaneously after receiving the three global optimization target values.

8. The microgrid group information interaction and control system according to claim 1, characterized in that, The microgrid control strategy includes: The microgrid control is implemented through the microgrid hierarchy, and the controlled objects are the active and reactive power outputs of each controllable distributed power source.

9. A method for information interaction and control of a microgrid group, characterized in that, include: Describe the interconnection topology that makes up a microgrid group; Based on the interconnected topology of the microgrid cluster, the corresponding information interaction objects are abstracted as proxy nodes. The local measurement data of the microgrid cluster is obtained through the proxy nodes, and data interaction is carried out through private data negotiation and anomaly identification is carried out through consensus anomaly perception. At the same time, a control strategy for energy scheduling of the microgrid cluster is generated. Obtain the necessary anomaly identification information, state estimation information, and control strategy information to achieve information exchange within the microgrid group; employ distribution network-level coordinated control strategies, inter-regional control strategies, and intra-microgrid control strategies to achieve energy dispatching of the microgrid group; wherein: The distribution network hierarchical coordination control strategy collects voltage estimation data for each tie line to obtain the operation information of the entire distribution network. It performs global optimization control on the distribution network and microgrid group through a global optimization algorithm, sets regional optimization targets for the lower-level area, namely the exchange power of the sub-microgrid, guides all controllable distributed generation units, flexible load units and tie switches on feeders in the entire distribution network, and sets regional optimization targets within the microgrid group, namely the target value of tie line output power and tie line coordination coefficient. The inter-regional control strategy coordinates all controllable distributed power sources and energy storage units within an autonomous region on a feeder through a tie-line hierarchy. The microgrid control strategy divides the control structure of the sub-microgrid into two layers. The first layer adopts a droop control strategy based on equal cost micro-increment to realize the proportional distribution of active power within the microgrid. The second layer adopts a microgrid DG control strategy based on the consistency principle to control the information interaction between distributed power sources on the same distribution node, realize the secondary adjustment of microgrid frequency, voltage, and reactive power, achieve voltage stability and power balance of the microgrid group, and complete the energy dispatch of the microgrid group.

10. A terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it can be used to run the system of any one of claims 1-8, or to execute the method of claim 9.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program can be used to run the system of any one of claims 1-8, or to perform the method of claim 9.

Images