Power supply network operation state prediction method and related equipment

Abstract

The invention relates to the technical field of power supply networks, and discloses a power supply network operation state prediction method and related equipment, and the method comprises the steps: determining a vulnerability function of a target node in a power supply network; determining the damage condition of the target node based on the vulnerability function and the exposure degree of the target node; calculating the comprehensive reliability measure of the target node based on the damage condition; constructing a probability graph model based on the comprehensive reliability measure and the relationship between the target node and the adjacent nodes; and predicting the operation state of the power supply network based on the probabilistic graph model. According to the prediction method, the vulnerability function of the node is determined firstly, and then the damage condition of the node is evaluated in combination with the exposure degree of the node. And then, determining the comprehensive reliability measure of the node by using a heuristic search algorithm. And then analyzing the relationship among the nodes to construct a probabilistic graph model. And finally, state reasoning is carried out through a belief propagation algorithm, and the operation state of the whole power supply network is predicted.

Description

Technical Field

[0001] The present invention relates to the technical field of power supply networks, and in particular, to a method, device, electronic device, storage medium, and product for predicting the operating state of a power supply network. Background Art

[0002] In recent years, extreme climate disasters have occurred frequently worldwide, and the chain reaction of urban systems has led to serious consequences. As a key urban infrastructure, the power supply network needs to quickly evaluate its operating state during extreme disasters to provide strong guarantees for the normal life of residents and subsequent rescue operations.

[0003] In related technologies, the state of nodes in a power supply network is usually predicted through power grid modeling methods. Currently, the power grid modeling methods mainly include a power flow-based power grid model and a network flow-based power grid model. For the former, the prior vulnerability between different nodes is not considered. When multiple nodes at the urban scale are damaged, it is difficult to obtain data and slow to calculate using this modeling method, which cannot meet the requirements of rapid perception. For the latter, the numerical quantification results of the correlation between nodes are not considered, and the global operating state of the power grid cannot be monitored through the states of a few nodes. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in related technologies to some extent. For this purpose, the first object of the present invention is to propose a method for predicting the operating state of a power supply network, which evaluates the damage of nodes through the vulnerability function of nodes and combines the exposure of nodes. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of nodes. After that, the relationship between nodes is analyzed to construct a probabilistic graph model. Finally, state inference is performed through the belief propagation algorithm to predict the operating state of the entire power supply network.

[0005] The second object of the present invention is to propose a device for predicting the operating state of a power supply network.

[0006] The third object of the present invention is to propose an electronic device.

[0007] The fourth object of the present invention is to propose a non-transitory computer-readable storage medium.

[0008] The fifth object of the present invention is to propose a computer program product.

[0009] To achieve the above object, an embodiment of the first aspect of the present invention provides a method for predicting the operating state of a power supply network, including: determining the vulnerability function of a target node in the power supply network; determining the damage condition of the target node based on the vulnerability function and the exposure degree of the target node; calculating the comprehensive reliability measure of the target node based on the damage condition; constructing a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes; predicting the operating state of the power supply network based on the probability graph model.

[0010] In addition, the method for predicting the operating state of a power supply network according to the above embodiment of the present invention may further have the following additional technical features:

[0011] According to some embodiments of the present invention, determining the vulnerability function of a target node in the power supply network includes: obtaining the network flow model of the power supply network, and determining the target node based on the importance of several nodes in the network flow model; performing vulnerability analysis on the target node and obtaining the analysis result; determining the vulnerability function of the target node based on the analysis result.

[0012] According to some embodiments of the present invention, determining the damage condition of the target node based on the vulnerability function and the exposure degree of the target node includes: respectively determining the exposure degree of the target node in several disaster scenarios; determining the physical damage probability of the target node based on the vulnerability function and the exposure degree; determining the damage condition of the target node in several disaster scenarios based on the physical damage probability.

[0013] According to some embodiments of the present invention, calculating the comprehensive reliability measure of the target node based on the damage condition includes: for each of several disaster scenarios, respectively calculating the node function normal probability of the target node in each disaster scenario based on the damage condition; decomposing the network flow model, and determining the sum of the node function normal probabilities of the target node in several disaster scenarios based on the node function normal probability of the target node in each disaster scenario; determining the shortest connected path between the target node and the source point; determining the comprehensive reliability measure of the target node based on the sum of probabilities and the shortest connected path.

[0014] According to some embodiments of the present invention, the damage condition includes the physical damage degree of the target node in the target disaster scenario; respectively calculating the node function normal probability of the target node in each disaster scenario based on the damage condition includes: determining whether there is a connected path between the target node and the source point based on the breadth-first search algorithm; and determining whether the physical damage degree of the target node in the target disaster scenario is less than or equal to the threshold; in response to determining that there is a connected path between the target node and the source point, and the physical damage degree of the target node in the target disaster scenario is less than or equal to the threshold, determining the state of the target node as the target state; determining the node function normal probability of the target node in the target state in each of several disaster scenarios.

[0015] According to some embodiments of the present invention, a probability graph model is constructed based on a comprehensive reliability measure and the relationship between a target node and its adjacent nodes, including: determining a state probability matrix between the target node and its adjacent nodes according to the comprehensive reliability measure; using the current states of the target node and its adjacent nodes as latent variables and the state probability matrix as a potential function to transform a network flow model and generate a probability graph model.

[0016] For the prediction method of the operating state of a power supply network according to an embodiment of the present invention, starting from known state nodes, the task of sensing the operating state of the power grid is changed into a function optimization task. Obtaining the state probability matrix between adjacent nodes through a heuristic search algorithm avoids the problem of dimensional explosion in a large-scale network and the slow convergence problem of the Monte Carlo method, and can dynamically adjust the prior vulnerability of nodes according to actual disaster situations, improving the prediction accuracy. In addition, the present invention also uses the belief propagation algorithm to accelerate iterative convergence through a triangulation process, and can predict the state of all unknown node sets from the state of a single node, thereby predicting the overall operating state of the power supply network.

[0017] An embodiment of the second aspect of the present invention provides a prediction device for the operating state of a power supply network, including: a function determination module configured to determine a vulnerability function of a target node in the power supply network; a damage situation determination module configured to determine the damage situation of the target node based on the vulnerability function and the exposure degree of the target node; a comprehensive reliability measure determination module configured to calculate the comprehensive reliability measure of the target node based on the damage situation; a probability model construction module configured to construct a probability graph model based on the comprehensive reliability measure and the relationship between the target node and its adjacent nodes; and an operating state prediction module configured to predict the operating state of the power supply network based on the probability graph model.

[0018] For the prediction device of the operating state of a power supply network according to an embodiment of the present invention, the device evaluates the damage situation of a node through the vulnerability function of the node and in combination with the exposure degree of the node. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed to construct a probability graph model. Finally, state inference is performed through the belief propagation algorithm to predict the operating state of the entire power supply network.

[0019] An embodiment of the third aspect of the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, and when the processor executes the program, it implements the above-mentioned prediction method for the operating state of a power supply network.

[0020] The electronic device according to an embodiment of the present invention evaluates the damage condition of a node by means of the vulnerability function of the node and in combination with the exposure degree of the node by executing the above-mentioned prediction method for the operation state of the power supply network. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed and then a probabilistic graph model is constructed. Finally, state inference is performed through the belief propagation algorithm to predict the operation state of the entire power supply network.

[0021] An embodiment of the fourth aspect of the present invention provides a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute the above-mentioned prediction method for the operation state of the power supply network.

[0022] The storage medium according to an embodiment of the present invention evaluates the damage condition of a node by means of the vulnerability function of the node and in combination with the exposure degree of the node by executing the above-mentioned prediction method for the operation state of the power supply network. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed and then a probabilistic graph model is constructed. Finally, state inference is performed through the belief propagation algorithm to predict the operation state of the entire power supply network.

[0023] An embodiment of the fifth aspect of the present invention provides a computer program product including computer program instructions which, when executed on a computer, cause the computer to execute the above-mentioned prediction method for the operation state of the power supply network.

[0024] The computer program product according to an embodiment of the present invention evaluates the damage condition of a node by means of the vulnerability function of the node and in combination with the exposure degree of the node by executing the above-mentioned prediction method for the operation state of the power supply network. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed and then a probabilistic graph model is constructed. Finally, state inference is performed through the belief propagation algorithm to predict the operation state of the entire power supply network.

[0025] In summary, after determining the vulnerability function of the node, the present invention evaluates the damage condition of the node in combination with the exposure degree of the node. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed and then a probabilistic graph model is constructed. Finally, state inference is performed through the belief propagation algorithm to predict the operation state of the entire power supply network.

[0026] In the present invention, starting from known state nodes, the task of perceiving the power grid operation state is transformed into a function optimization task. By using a heuristic search algorithm to obtain the state probability matrix between adjacent nodes, the problems of dimensional explosion in large-scale networks and slow convergence of the Monte Carlo method are avoided. The task of dynamically adjusting the prior vulnerability of nodes according to actual disaster situations can improve the prediction accuracy. In addition, the present invention also uses the belief propagation algorithm to accelerate iterative convergence through the triangulation process, and can predict the states of all unknown node sets from the state of a single node, and then predict the overall operation state of the power supply network.

[0027] Additional aspects and advantages of the present invention will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 FIG. is a flowchart of a method for predicting the operation state of a power supply network according to some embodiments of the present invention;

[0029] Figure 2 FIG. is a distribution diagram of the number of initially damaged nodes in a power supply network according to some embodiments of the present invention;

[0030] Figure 3 FIG. is a distribution diagram of the number of nodes damaged by cascading in a power supply network according to some embodiments of the present invention;

[0031] Figure 4 FIG. is a schematic diagram of a device for predicting the operation state of a power supply network according to some embodiments of the present invention;

[0032] Figure 5 FIG. is a block diagram of an electronic device according to some embodiments of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments and the accompanying drawings.

[0034] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should have the ordinary meanings understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the embodiments of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Words such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "linked" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right", etc. are only used to represent relative position relationships. When the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0035] As described in the background art section, in recent years, extreme climate disasters have occurred frequently worldwide, and the chain reaction of urban systems has led to serious consequences. As a key urban infrastructure, the power supply network needs to quickly evaluate its operating status during extreme disasters to provide strong guarantees for the normal life of residents and subsequent rescue.

[0036] The applicant found in the process of implementing the present invention that in the related art, the state of nodes in the power supply network is usually predicted through power grid modeling methods. Currently, the power grid modeling methods mainly include power flow-based power grid models and network flow-based power grid models. For the former, the prior vulnerability between different nodes is not considered. When multiple nodes at the urban scale are damaged, it is difficult to obtain data and slow to calculate using this modeling method, and it cannot meet the requirements of rapid perception. The latter does not consider the numerical quantification results of the correlation between nodes, and it is impossible to monitor the global operating status of the power grid through the status of a few nodes.

[0037] Therefore, the damage situation of nodes is evaluated through the vulnerability function of nodes and combined with the exposure degree of nodes. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of nodes. After that, the relationship between nodes is analyzed and a probabilistic graph model is constructed. Finally, state inference is performed through the belief propagation algorithm to predict the operating status of the entire power supply network.

[0038] Next, the prediction method, device, electronic device, storage medium, and computer program product for the operating status of the power supply network proposed in the embodiments of the present invention will be described with reference to the accompanying drawings.

[0039] Refer to Figure 1 , which is a flowchart of the prediction method for the operating status of the power supply network according to some embodiments of the present invention.

[0040] As Figure 1 shown, the prediction method for the operating status of the power supply network in the embodiments of the present invention may include the following steps:

[0041] S101, determine the vulnerability function of the target node in the power supply network.

[0042] In some alternative embodiments, the prediction method in the present invention can be implemented by a network prediction model. Specifically, the network prediction model can first obtain the data information of power supply nodes in the corresponding Geographic Information System (GIS) of the power supply network, and then determine the adjacency relationship between power supply nodes and generate a network flow model of the power supply network.

[0043] In some alternative embodiments, the power supply network is a complex system composed of various types of nodes and the lines connecting them. It may include several nodes, such as power plants, substations, transformers, transmission lines, distribution lines, etc. These nodes undertake different functions in the power supply network and cooperate together to achieve the transmission and distribution of electric energy. In the power supply network, not all power supply nodes are meaningful. Therefore, before generating the network flow model, the network prediction model will first screen and analyze all the power supply nodes in the power supply network, and then determine the relatively important power supply nodes. Then, a network flow model is generated based on these relatively important power supply nodes.

[0044] In some alternative embodiments, after obtaining the network flow model of the power supply network, the network prediction model needs to further determine the importance of each power supply node therein, so as to identify the target nodes whose importance is higher than a preset threshold. Specifically, the network prediction model first quantitatively evaluates the importance of each node based on the characteristics and connection relationships of the nodes in the network flow model, using specific algorithms or evaluation criteria, such as node betweenness centrality, degree centrality, etc. Then, by setting an importance threshold, the nodes whose importance is higher than this threshold are selected as target nodes. It can be understood that the setting of the importance threshold needs to comprehensively consider the network scale, the total number of nodes and the requirements for the importance of key nodes in the specific application scenario, so as to ensure that the selected target nodes have a significant influence and key role in the power supply network. Specifically, the target nodes extracted by the network prediction model according to the importance of power supply nodes can be divided into three levels, namely, power plant—(transmission network)—substation—(distribution network)—transformer.

[0045] In some alternative embodiments, after determining the target nodes, the network prediction model can perform vulnerability analysis on these target nodes. Specifically, the network prediction model can comprehensively evaluate the possible physical damage degrees of these target nodes in different disaster scenarios through multi-dimensional information such as the physical attributes of the target nodes, their geographical locations, surrounding environmental factors, and historical operation data. For example, for a substation located in an earthquake-prone area, its structural seismic performance needs to be analyzed emphatically; while for a transmission line located in a low-lying area, the impact caused by flood immersion needs to be considered emphatically. Through the above process, the network prediction model can obtain the vulnerability characteristics of each target node in various disaster scenarios, and then obtain the vulnerability analysis results. Finally, the network prediction model can determine the corresponding vulnerability function for each target node.

[0046] In some alternative embodiments, the foregoing vulnerability function can be expressed as:

[0047] V(n) = f(material strength, structural form, geographical location, historical maintenance record)

[0048] Wherein, V(n) represents the vulnerability of node n, and f represents function calculation.

[0049] S102. Determine the damage condition of the target node based on the vulnerability function and the exposure degree of the target node.

[0050] In some alternative embodiments, the network prediction model first evaluates the exposure degree of each target node in multiple preset disaster scenarios. It can be understood that the determination of the exposure degree needs to comprehensively consider factors such as the geographical location of the target node, the surrounding environment, and the characteristics of the disaster. For example, for a target node located in a high flood-risk area, its exposure degree in the flood disaster scenario will be relatively high.

[0051] In some alternative embodiments, through the foregoing vulnerability function and exposure degree, the network prediction model can further calculate the physical damage probability of the target node in each disaster scenario. Among them, the foregoing physical damage probability is used to reflect the probability that the target node may suffer physical damage in a specific disaster scenario.

[0052] Among them, the probability of physical damage can be expressed as:

[0053] P(damaged) = V(n) × E(n)

[0054] Wherein, P(damaged) represents the damage probability of node n, V(n) is the vulnerability of the node, and E(n) is the exposure degree of the node.

[0055] In some alternative embodiments, the network prediction model can determine the damage status of the target node under different disaster scenarios based on the above physical damage probability. Through this process, the network prediction model can convert the quantified physical damage probability into a more specific description of the damage status, such as slight damage, moderate damage, or severe damage, etc., so as to more intuitively understand the expected performance of the target node in the disaster.

[0056] S103, calculate the comprehensive reliability measure of the target node based on the damage status.

[0057] In some alternative embodiments, for each disaster scenario, the network prediction model can use the breadth-first search algorithm to check whether there is a connected path between the target node and the source point. Among them, the aforementioned source point represents the starting point of power supply, such as a power plant or a substation, etc. It is the starting point of power transmission and is responsible for delivering electrical energy to each node in the entire network.

[0058] In some alternative embodiments, through the breadth-first search algorithm, the network prediction model can systematically explore all possible paths between the target node and the source point in the network flow model, ensuring that no potential connections are missed. At the same time, the network prediction model can also evaluate the physical damage degree of the target node in the target disaster scenario and determine whether it exceeds the threshold. It can be understood that this threshold is comprehensively set according to the physical characteristics of the node, historical data, and the characteristics of the disaster, aiming to distinguish whether the node can maintain its basic functions in the disaster.

[0059] In some alternative embodiments, if there is a connected path between the target node and the source point, and its physical damage degree is within an acceptable range (i.e., less than or equal to the threshold), then mark this node as the target state. Among them, the target state indicates that the target node is in a normal operating state in the current disaster scenario, and it can provide support for subsequent power transmission and distribution.

[0060] Next, for the target node in the target state, the network prediction model further determines its probability of normal function under each disaster scenario. This probability takes into account not only the physical damage degree of the node, but also its position, connection relationship, and other relevant factors in the network.

[0061] In some alternative embodiments, in order to more accurately reflect the overall reliability of the node under multiple disaster scenarios. The network prediction model can also decompose the network flow model, and then, based on the decomposition of the network flow model, aggregate the probabilities of normal function of the target node under all disaster scenarios to obtain a comprehensive probability sum. It can be understood that by decomposing the network flow model, the network prediction model can more carefully analyze the mutual influence and dependence relationship between nodes, so as to more accurately calculate the probability of normal function of each node.

[0062] Among them, the decomposition process of the network prediction model for the network flow model can be expressed as:

[0063]

[0064] Among them: Φ(G) - the probability of satisfying the target event in graph G; f(R0) - the probability of path R0 being connected; - the probability of path R0 not being connected; n1n2…n k - the nodes in path R0; - the probability that node n1 is not physically damaged while node n2 is physically damaged; Φ(G k ) - the sub - graph structure obtained by removing node n k from graph G.

[0065] In some alternative embodiments, after obtaining the sum of probabilities, the network prediction model further needs to determine the shortest connected path between the target node and the source node. After that, the network prediction model can combine the sum of probabilities and the shortest connected path to calculate the comprehensive reliability measure of the target node.

[0066] It can be understood that the comprehensive reliability measure comprehensively considers the functional state of the target node under different disaster scenarios and the connectivity in the network, providing a comprehensive basis for evaluating the reliability of the target node and the operating state of the entire power supply network. In this way, the network prediction model can not only accurately evaluate the reliability of a single node, but also provide scientific support for the optimization and emergency management of the entire power supply network.

[0067] S104. Construct a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes.

[0068] S105. Predict the operating state of the power supply network based on the probability graph model.

[0069] In some alternative embodiments, after obtaining the comprehensive reliability measure of the target node, the network prediction model can determine the state probability matrix between it and adjacent nodes based on the comprehensive reliability measure of the target node. This matrix can reflect the transition probabilities between adjacent nodes in different states. For example, the probability of transitioning from the normal state to the failure state, or the probability of remaining in the normal state, etc. It can be understood that the construction of the state probability matrix needs to comprehensively consider factors such as the physical characteristics of the nodes, the network topology structure, and historical operation data to ensure its accuracy and practicality.

[0070] In some alternative embodiments, next, the network prediction model can regard the current states of the target node and its adjacent nodes as latent variables, and use the constructed state probability matrix as a potential function to transform the network flow model and obtain a probabilistic graphical model. Then, the operating state of the power supply network can be predicted through the probabilistic graphical model. Among them, the aforementioned latent variables represent various states that the nodes may be in during actual operation, such as normal operation, slightly damaged, severely damaged, etc. The aforementioned potential function can be used to quantify the tendency or possibility of state transitions between nodes, providing key parameter support for the probabilistic graphical model.

[0071] In some alternative embodiments, when the network prediction model transforms the network flow model, it can combine the attributes of nodes and edges with the state probability matrix to transform the traditional network flow model into a probabilistic graphical model. The probabilistic graphical model can graphically display the probabilistic dependence relationships between nodes, making the analysis of complex network systems more intuitive and efficient.

[0072] It can be understood that the probabilistic graphical model generated by the method in the present invention not only retains the topological structure and basic characteristics of the original network flow model, but also incorporates node states and probability information, providing a solid foundation for subsequent network operating state prediction and reliability analysis. This model can be used to simulate the operation of the power supply network under different disaster scenarios, predict the failure probability of nodes, and evaluate the reliability of the entire network, thereby providing a scientific basis for emergency management and network optimization.

[0073] In some alternative embodiments, after obtaining the probabilistic graphical model, the network prediction model can also obtain a real-time disaster risk field based on disaster deduction data information, real-time social media, and data information of the monitoring network. Then, the probability distribution in the probabilistic graphical model can be dynamically adjusted through the real-time disaster risk field. And by deploying the belief propagation algorithm and combining the operating states of some monitoring points, the operating state information of all nodes in the power supply network can be determined to achieve the state prediction of the entire power supply network.

[0074] In some alternative embodiments, dynamically updating the urban risk field mainly includes, based on the simulation results of the disaster model, correcting the simulation data through the real-time disaster intensity feedback from each monitoring point in the city and the public opinion monitoring situation on social media such as Weibo and forums to obtain more accurate disaster intensity information at the urban scale.

[0075] In some alternative embodiments, summarizing the operating conditions of monitoring nodes mainly includes, on the premise that communication remains unobstructed, monitoring the operating conditions of some nodes (such as 30%).

[0076] In some alternative embodiments, predicting the operating conditions of the remaining nodes based on graph reasoning mainly includes: reading the potential function in the probabilistic graph model, taking the operating conditions of the known nodes as environmental parameters, the state of each remaining unknown node as a discrete variable, and the potential function as the conditional probability, which constrains its value state; defining the node set as {N}, deploying the belief propagation algorithm, and iteratively updating the state probability distribution of the target node set according to the following equations (1) and (2) until convergence. The selection of the node set {N} can be flexibly distributed according to the needs of the target task, and at least a single element can be taken. In this case, the calculated result is the marginal probability distribution of a single node.

[0077]

[0078] Equation (1) represents the belief transfer process from node i to its neighbor node a, where m represents the strength of the belief and N(i) represents the neighborhood of node i.

[0079]

[0080] Equation (2) represents the information receiving process of node i, where N(i) represents the neighborhood of node i, and m a→i (x i ) represents the strength of the belief transferred from neighbor node a to i.

[0081] In some alternative embodiments, for the sensing task, the node set {N} is defined as all the remaining nodes. At this time, the belief propagation algorithm is equivalent to solving a maximum a posteriori probability (MAP) problem, and the state with the maximum probability is selected as the prediction result.

[0082] Figure 2 is the distribution diagram of the initial damaged node numbers in the power supply network according to some embodiments of the present invention. As Figure 2 shown, it can be seen that the number of initially damaged nodes shows an obvious normal trend. According to the normal distribution, it can be considered that when the number of initially damaged nodes is less than 10, it is a minor disaster; when it is 11 - 14, it is a general disaster; and when it is more than 15, it is a severe disaster. Then the corresponding inference accuracy can be referred to Table 1 below:

[0083] Table 1 Inference accuracy under different initial disaster intensities and proportions of known state nodes

[0084]

[0085]

[0086] As can be seen from the table, the accuracy of various methods decreases as the severity of the disaster increases. In the case of minor disasters, the accuracy of the BP method can reach 80% when the proportion of known state nodes is 30%, which is nearly 30% higher than that of traditional physics-based reliability methods. As the severity of the disaster increases, BP relies on more node states to provide more information for inference. In minor disasters, the 10% and 30% known proportions have little impact on the accuracy of the BP method. However, for severe disasters, the inference accuracy is 10% higher when the known proportion is 30% compared to 10%.

[0087] Figure 3 is a distribution diagram of the number of cascading failure nodes in the power supply network according to some embodiments of the present invention. As Figure 3 shown, it is divided according to different degrees of cascading disasters. It can be seen that the number of cascading damaged nodes shows a long-tailed normal distribution. Below 6 is defined as a minor cascading disaster (80%), 7 - 9 is a general cascading disaster (16%), and above 10 is a severe cascading disaster (4%). The corresponding chart is as follows:

[0088] Table 2 Inference accuracy under different cascading disaster intensities and proportions of known state nodes

[0089]

[0090]

[0091] As can be seen from the table, the proportion of severe cascading disasters is not high, but the prediction accuracy of the physics-based reliability method is close to random selection, indicating that the traditional physics-based reliability method is ineffective for severe cascading disasters. At the same time, in the case of cascading disasters above medium level, there are certain differences between the comprehensive reliability method and the physics-based reliability method, which also proves the rationality of the comprehensive reliability calculation process. Under severe cascading disasters, as the number of known nodes increases, the prediction accuracy of the BP method improves significantly. When the known proportion is 30%, the prediction accuracy is 30% higher than that of the physics-based reliability method.

[0092] In summary, it can be seen that the present method improves and optimizes the traditional inference method, and is a new method that can make full use of disaster information for faster and more accurate state inference.

[0093] In summary, after determining the vulnerability function of the node, the present invention evaluates the damage of the node in combination with the exposure of the node. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed to construct a probabilistic graph model. Finally, state inference is performed through the belief propagation algorithm to predict the operating state of the entire power supply network.

[0094] In the present invention, starting from a known state node, the power grid operation state perception task is transformed into a function optimization task. Obtaining the state probability matrix between adjacent nodes through a heuristic search algorithm avoids the dimensional explosion problem in a large-scale network and the slow convergence problem of the Monte Carlo method, and can dynamically adjust the task of the prior vulnerability of nodes according to the actual disaster situation, improving the prediction accuracy. In addition, the present invention also uses the belief propagation algorithm to accelerate iterative convergence through the triangulation process, and can predict the states of all unknown node sets from the state of a single node, and then predict the overall operation state of the power supply network.

[0095] It should be noted that the method of the embodiment of the present invention can be executed by a single device, such as a computer or a server. The method of this embodiment can also be applied to a distributed scenario and completed by the cooperation of multiple devices. In this case of a distributed scenario, one of the multiple devices can only execute one or more steps of the method of the embodiment of the present invention, and these multiple devices will interact with each other to complete the above method.

[0096] It should be noted that some embodiments of the present invention have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be executed in a different order than in the above embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0097] Corresponding to the above embodiments, the present invention also proposes a prediction device for the operation state of a power supply network.

[0098] As Figure 4 shown, the prediction device for the operation state of the power supply network according to the embodiment of the present invention includes: a function determination module 410, a damage situation determination module 420, a comprehensive reliability measure determination module 430, a probability model construction module 440, and an operation state prediction module 450.

[0099] Among them, the function determination module 410 is configured to: determine the vulnerability function of the target node in the power supply network; the damage situation determination module 420 is configured to: determine the damage situation of the target node based on the vulnerability function and the exposure degree of the target node; the comprehensive reliability measure determination module 430 is configured to: calculate the comprehensive reliability measure of the target node based on the damage situation. The probability model construction module 440 is configured to: construct a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes. The operation state prediction module 450 is configured to: construct a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes.

[0100] In some embodiments of the present invention, the function determination module 410 is specifically configured to: obtain a network flow model of a power supply network, and determine a target node based on the importance degrees of several nodes in the network flow model; perform vulnerability analysis on the target node and obtain an analysis result; and determine a vulnerability function of the target node based on the analysis result.

[0101] In some embodiments of the present invention, the damage condition determination module 420 is specifically configured to: respectively determine the exposure degrees of the target node in several disaster scenarios; determine the physical damage probability of the target node based on the vulnerability function and the exposure degree; and determine the damage condition of the target node in several disaster scenarios based on the physical damage probability.

[0102] In some embodiments of the present invention, the comprehensive reliability measure determination module 430 is specifically configured to: for each of several disaster scenarios, calculate the node function normal probability of the target node in each disaster scenario based on the damage condition, including: calculating the node function normal probability of the target node in each disaster scenario based on the damage condition, including: determining whether there is a connected path between the target node and the source point based on the breadth-first search algorithm; and determining whether the physical damage degree of the target node in the target disaster scenario is less than or equal to a threshold; in response to determining that there is a connected path between the target node and the source point and the physical damage degree of the target node in the target disaster scenario is less than or equal to the threshold, determining the state of the target node as the target state; determining the node function normal probability of the target node in the target state in each of several disaster scenarios; decomposing the network flow model, and determining the sum of the node function normal probabilities of the target node in several disaster scenarios based on the node function normal probability of the target node in each disaster scenario; determining the shortest connected path between the target node and the source point; and determining the comprehensive reliability measure of the target node based on the sum of probabilities and the shortest connected path.

[0103] In some embodiments of the present invention, the probability model construction module 440 is specifically configured to: determine a state probability matrix between the target node and adjacent nodes according to the comprehensive reliability measure; use the current states of the target node and adjacent nodes as hidden variables and the state probability matrix as a potential function to transform the network flow model and generate a probabilistic graphical model.

[0104] It should be noted that for the details not disclosed in the prediction device for the operation state of the power supply network in the embodiments of the present invention, please refer to the details disclosed in the prediction method for the operation state of the power supply network in the embodiments of the present invention, which will not be elaborated herein.

[0105] In summary, the prediction device for the operating state of the power supply network according to the embodiments of the present invention evaluates the damage condition of a node through the vulnerability function of the node and in combination with the exposure degree of the node. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed and a probabilistic graph model is constructed. Finally, state inference is performed through the belief propagation algorithm to predict the operating state of the entire power supply network.

[0106] For the sake of convenience of description, when describing the above system, various modules are described separately according to their functions. Of course, when implementing the present invention, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0107] The system of the above embodiment is used to implement the corresponding method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0108] Corresponding to the above embodiment, the present invention also proposes an electronic device.

[0109] Reference Figure 5 , which is a block diagram of an electronic device according to some embodiments of the present invention, showing a more specific schematic diagram of the hardware structure of the electronic device provided in this embodiment. The device may include: a processor 510, a memory 520, an input / output interface 530, a communication interface 540, and a bus 550. Among them, the processor 510, the memory 520, the input / output interface 530, and the communication interface 540 are communicatively connected to each other inside the device through the bus 550.

[0110] The processor 510 may be implemented in a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0111] The memory 520 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 520 may store an operating system and other application programs. When implementing the technical solutions provided in the embodiments of this specification through software or firmware, the relevant program codes are stored in the memory 520 and are called and executed by the processor 510.

[0112] The input / output interface 530 is used to connect to the input / output module to achieve information input and output. The input / output module can be configured as a component in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Among them, the input devices can include keyboards, mice, touchscreens, microphones, various sensors, etc., and the output devices can include displays, speakers, vibrators, indicator lights, etc.

[0113] The communication interface 540 is used to connect to the communication module (not shown in the figure) to achieve communication interaction between this device and other devices. Among them, the communication module can achieve communication through wired means (such as USB, network cable, etc.) or through wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0114] The bus 550 includes a path for transmitting information between various components of the device (such as the processor 510, the memory 520, the input / output interface 530, and the communication interface 540).

[0115] It should be noted that although the above device only shows the processor 510, the memory 520, the input / output interface 530, the communication interface 540, and the bus 550, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device may also only include the components necessary to implement the solution of the embodiments of this specification, and do not have to include all the components shown in the figure.

[0116] The electronic device of the above embodiment is used to implement the corresponding method in any of the foregoing embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0117] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present invention also provides a non-transitory computer-readable storage medium. The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a computer to execute the method of any of the above embodiments.

[0118] The above non-transitory computer-readable storage medium can be any available medium or data storage device accessible by a computer, including but not limited to magnetic memories (such as floppy disks, hard disks, magnetic tapes, magneto-optical discs (MO), etc.), optical memories (such as CDs, DVDs, BDs, HVDs, etc.), and semiconductor memories (such as ROM, EPROM, EEPROM, non-volatile memories (NAND FLASH), solid-state drives (SSD)), etc.

[0119] The computer instructions stored in the storage medium of the above embodiment are used to cause a computer to execute the method of any of the above exemplary method parts, and have the beneficial effects of the corresponding method embodiments, which will not be elaborated here.

[0120] Based on the same inventive concept, corresponding to the method of any of the above embodiments, the present invention further provides a computer program product, including computer program instructions, which, when executed on a computer, cause the computer to execute the above method for predicting the operating state of a power supply network.

[0121] According to the computer program product of the embodiments of the present invention, by executing the above method for predicting the operating state of a power supply network, the damage condition of a node is evaluated through the vulnerability function of the node and in combination with the exposure degree of the node. Then, a heuristic search algorithm is used to determine the comprehensive reliability measure of the node. After that, the relationship between nodes is analyzed and a probabilistic graph model is constructed. Finally, state inference is performed through a belief propagation algorithm to predict the operating state of the entire power supply network.

[0122] In addition, although the operations of the method of the present invention are described in a specific order in the drawings, this does not require or imply that these operations must be performed in that specific order, or that all of the shown operations must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart may be changed in the order of execution. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution.

[0123] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0124] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present invention should have the ordinary meanings understood by those of ordinary skill in the art to which the present invention pertains. The "first", "second" and similar terms used in the embodiments of the present invention do not denote any order, quantity or importance, but are only used to distinguish different components. Words such as "comprising" or "including" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. Words such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Upper", "lower", "left", "right", etc. are only used to represent relative positional relationships, and when the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0125] Although the spirit and principles of the present invention have been described with reference to several specific embodiments, it should be understood that the present invention is not limited to the specific embodiments disclosed, and the division of each aspect does not mean that the features in these aspects cannot be combined for benefits. Such a division is only for the convenience of expression. The present invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. The scope of the appended claims is to be accorded the broadest interpretation so as to encompass all such modifications and equivalent structures and functions.

Claims

1. A method for predicting the operating state of a power supply network, characterized in that, Including: Determine the vulnerability function of the target node in the power supply network; Determine the damage condition of the target node based on the vulnerability function and the exposure degree of the target node; Calculate the comprehensive reliability measure of the target node based on the damage condition; Construct a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes; Predict the operating state of the power supply network based on the probability graph model.

2. The prediction method for the operating state of the power supply network according to claim 1, characterized in that The determining the vulnerability function of the target node in the power supply network includes: Obtain the network flow model of the power supply network, and determine the target node based on the importance of several nodes in the network flow model; Conduct a vulnerability analysis on the target node and obtain the analysis result; Determine the vulnerability function of the target node based on the analysis result.

3. The prediction method for the operating state of the power supply network according to claim 1, characterized in that, The determining the damage condition of the target node based on the vulnerability function and the exposure degree of the target node includes: Respectively determine the exposure degrees of the target node under several disaster scenarios; Determine the physical damage probability of the target node based on the vulnerability function and the exposure degree; Determine the damage condition of the target node under several disaster scenarios based on the physical damage probability.

4. The prediction method for the operating state of the power supply network according to claim 1, characterized in that, The calculating the comprehensive reliability measure of the target node based on the damage condition includes: For each of several disaster scenarios, calculate the node function normal probability of the target node under each disaster scenario based on the damage condition; Decompose the network flow model, and based on the node function normal probability of the target node under each disaster scenario, determine the sum of the node function normal probabilities of the target node under several disaster scenarios; Determine the shortest connected path between the target node and the source point; Determine the comprehensive reliability measure of the target node based on the sum of probabilities and the shortest connected path.

5. The prediction method for the operating state of the power supply network according to claim 4, characterized in that, The damage condition includes the physical damage degree of the target node under the target disaster scenario; The calculating the node function normal probability of the target node under each disaster scenario based on the damage condition includes: Determine whether there is a connected path between the target node and the source point based on the breadth-first search algorithm; And determine whether the physical damage degree of the target node under the target disaster scenario is less than or equal to the threshold; In response to determining that there is a connected path between the target node and the source point, and the physical damage degree of the target node under the target disaster scenario is less than or equal to the threshold, determine the state of the target node as the target state; Determine the node function normal probability of the target node in the target state under each of several disaster scenarios.

6. The prediction method for the operation state of the power supply network according to claim 1, wherein The constructing a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes includes: Determine the state probability matrix between the target node and adjacent nodes according to the comprehensive reliability measure; Take the current states of the target node and the adjacent nodes as hidden variables, and the state probability matrix as the potential function, transform the network flow model and generate a probability graph model.

7. A prediction device for the operating state of a power supply network, comprising: A function determination module, configured to: determine the vulnerability function of a target node in the power supply network; A damage situation determination module, configured to: determine the damage situation of the target node based on the vulnerability function and the exposure degree of the target node; A comprehensive reliability measure determination module, configured to: calculate the comprehensive reliability measure of the target node based on the damage situation; A probability model construction module, configured to: construct a probability graph model based on the comprehensive reliability measure and the relationship between the target node and adjacent nodes; An operating state prediction module, configured to: predict the operating state of the power supply network based on the probability graph model.

8. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium, which stores computer instructions for causing a computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, comprising computer program instructions, which when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 6.

Images