A method and device for identifying cross-regional propagation and cascade diffusion of urban group power grid risks, a computer device and a computer program product

Abstract

This application belongs to the field of power system risk assessment and power grid security prevention and control technology under extreme weather conditions. In particular, it relates to a method, device, computer equipment and computer program product for identifying cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids. It can be adapted to the complex coupling propagation characteristics of urban agglomeration power grids under extreme weather scenarios such as typhoons, and can combine historical data review and deduction with real-time dynamic analysis. It can avoid the technical defects of delayed risk prediction, blind prevention and control, and inability to provide accurate basis for cross-regional emergency mutual assistance and prevention of cascading collapse.

Description

Technical Field

[0001] This application belongs to the field of power system risk assessment and power grid security control technology under extreme weather conditions, and in particular relates to a method, device, computer equipment and computer program product for identifying cross-regional propagation and cascading diffusion of power grid risks in urban agglomerations. Background Technology

[0002] The power grids of coastal urban clusters are tightly coupled electrically and highly interconnected. Under extreme weather disturbances such as typhoons and cold waves, single-point equipment failures can easily spread to neighboring urban power grids through interconnection channels, power flow interaction, and voltage coupling, inducing power flow shifts, line overloads, and protection maloperation, forming cross-regional, cascading, and avalanche-like cascading faults.

[0003] Existing technologies suffer from the following shortcomings: Traditional risk identification often targets independent power grids in single cities, neglecting the power interaction and reactive power coupling characteristics of interconnected lines between urban clusters, and failing to characterize the cross-city propagation patterns of risks. Existing cascading failure models are mostly fault simulations under static topologies, failing to consider spatiotemporal random disturbances such as typhoon wind speeds, heavy rains, and line breaks. Furthermore, they do not fully utilize historical typhoon meteorological data, power grid fault records, and power flow anomaly samples for model calibration and scenario review, resulting in identification results that poorly reflect actual disaster conditions. There is also a lack of risk propagation tracing mechanisms, making it difficult to accurately locate risk origin nodes, transit propagation links, and critical collapse zones, and unable to distinguish between localized faults and cross-regional diffusion faults. In addition, existing technologies lack quantitative indicators of risk diffusion intensity, failing to characterize the multi-level amplification and evolution process of risks from local nodes → regions → cities → urban clusters. Summary of the Invention

[0004] To address the technical challenge mentioned in the background art of how to dynamically identify the cross-regional propagation path and multi-scale cascading diffusion of risks in urban agglomeration power grids based on historical meteorological and fault data review and calibration under spatiotemporal random disturbances caused by typhoons, this application provides a method, device, computer equipment, and computer program product for identifying cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids.

[0005] In a first aspect, embodiments of this application provide a method for identifying the cross-regional propagation and cascading diffusion of power grid risks in urban agglomerations, including: Step S1: Collect the topology parameters and operating condition data of the power grids of each city in the urban agglomeration, as well as the real-time monitoring data of the typhoon weather and air under the current typhoon event; and collect historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region, wherein the historical fault data includes historical failure records of each power equipment and power flow over-limit time sequence records of intercity tie lines. Step S2: Based on the real-time monitoring data of typhoon weather and the historical meteorological data of typhoons over the years, a non-stationary Gaussian mixture model is used to characterize the spatiotemporal stochastic evolution law of typhoon meteorological elements. The fitting parameters in the equipment vulnerability function are corrected by combining the historical fault data of the power grid. The random failure probability of power equipment in each city power grid described by the topology parameters at different times is calculated to screen out high-risk fault nodes as the initial risk triggering source. Step S3: Based on the topology parameters and operating condition data collected in Step S1, calculate the active power flow transfer sensitivity, voltage coupling correlation and reactive power interaction coefficient of inter-city node pairs, so as to construct a directed weighted adjacency matrix for cross-regional risk propagation of urban agglomeration power grid, so as to quantify the direction and weight of risk transmission between different nodes. Step S4: Starting from the initial risk trigger source, define the risk propagation neighborhood using the directed weighted adjacency matrix; wherein, the neighborhood of node j is all directed weighted adjacency matrices. For node i with a value greater than 0, a cascaded diffusion inference model is constructed by integrating Markov random processes and improved cellular automata. Historical meteorological data of typhoons from previous years and historical fault data of the power grid in the same region are retrieved from step S1. The risk propagation time series statistical law under historical typhoon disasters of the same level is extracted from the historical meteorological data of typhoons from previous years and historical fault data of the power grid in the same region. Based on the statistical law, the parameters of the node state transition probability formula in the cascaded diffusion inference model are optimized. The multi-scale cascaded process of risk spreading within the city, between cities, and across the entire urban agglomeration is inferred in layers to generate a risk propagation time series path chain. Step S5: Based on the risk propagation time-series path chain generated in Step S4, calculate the four-dimensional risk indicators of propagation breadth, diffusion depth, spread rate, and power loss amplitude for each propagation link to construct a four-dimensional risk assessment index matrix; use the entropy weight-TOPSIS algorithm to calculate the four-dimensional risk assessment index matrix to obtain the comprehensive risk proximity of each link, classify the risk level according to the proximity, and output high-risk blocking nodes, key weak sections, and risk diffusion boundaries to complete the identification of cross-regional risk propagation and cascading diffusion.

[0006] The technical advantages of the first aspect of this application are as follows: First, by integrating a typhoon-induced spatiotemporal meteorological stochastic model with a device vulnerability function calibrated using historical fault data, it solves the problems of traditional methods ignoring spatiotemporal disturbance patterns and failing to analyze historical scenarios, thus accurately locating the initial risk triggering source. Second, by constructing a multi-dimensional directed weighted adjacency matrix integrating power flow, voltage, and reactive power, it overcomes the limitations of traditional methods that rely solely on static topology, enabling the quantification of the direction and intensity of risk propagation across cities. Third, by optimizing state transitions based on a Markov-cellular automata model and incorporating historical propagation patterns, it achieves multi-scale dynamic extrapolation from within cities to intercity areas and then to the entire region. Finally, through four-dimensional quantitative rating, it accurately identifies high-risk links and collapse thresholds. Therefore, this invention achieves a leap from "single-point static assessment" to "cross-regional dynamic cascading identification." It can adapt to the complex coupling and propagation characteristics of urban agglomeration power grids under typhoon scenarios, combining historical data analysis with real-time dynamic judgment, and avoiding the technical shortcomings of delayed risk prediction, blind prevention and control, and the inability to provide accurate evidence for cross-regional emergency mutual assistance and preventing cascading collapses.

[0007] In one embodiment of the first aspect, in step S2, the probability density function of the non-stationary Gaussian mixture model is: ; in, For meteorological element variables; For the first The weights of each Gaussian component; For time-varying expectations; Standard deviation; The number of Gaussian components.

[0008] The technical advantage of this embodiment lies in: by introducing the time-varying expectation parameter μ k (t) can accurately characterize the non-stationary spatiotemporal stochastic evolution of typhoon path deviation, wind speed, and rainfall intensity, overcoming the shortcomings of traditional static distribution models that cannot describe the dynamic movement of typhoons. This makes the generated meteorological disturbance sequence closer to the real physical process, providing more accurate external driving conditions for subsequent equipment failure probability calculations, thereby significantly improving the identification accuracy of initial risk triggering sources.

[0009] In one embodiment of the first aspect, in step S2, the device vulnerability function is a piecewise nonlinear probability model:

[0010] in, This represents the random failure probability of electrical equipment. Real-time wind speed; These are the minimum wind speed threshold that the equipment can withstand and the maximum wind speed threshold that the equipment can withstand, respectively. The parameters are fitted to the fragile shape.

[0011] The technical advantages of this embodiment are as follows: by employing a piecewise nonlinear model and introducing shape fitting parameters λ and η, it is possible to flexibly distinguish the differentiated failure characteristics of different equipment (such as lines, transformers, and towers) under typhoons. Simultaneously, by calibrating these parameters using historical fault data, the calculation of equipment failure probability no longer relies on theoretical assumptions but is based on actual disaster statistical patterns, significantly improving the model's engineering adaptability and calculation accuracy in real typhoon scenarios.

[0012] In one embodiment of the first aspect, in step S3, the elements of the directed weighted adjacency matrix Calculated using the following formula:

[0013] in, For nodes To the node Risk transmission weight; Sensitivity to active power flow transfer; Voltage coupling sensitivity; Reactive power interaction sensitivity; Assign coefficients to the weights, and satisfy the following conditions: .

[0014] The technical advantage of this embodiment lies in its integration of three types of electrical coupling sensitivity—power flow, voltage, and reactive power—to quantify the risk transmission capability between cities, overcoming the limitations of traditional single-dimensional (power flow only) approaches. By flexibly adjusting the contribution of these three factors through weighting coefficients, it can more comprehensively characterize the complex cross-coupling propagation characteristics of urban agglomeration power grids under typhoon disturbances. The definition of positive and negative signs for matrix elements allows this invention to clearly distinguish between two risk propagation modes: forward diffusion and reverse backflow, providing crucial directional information for subsequent source tracing analysis.

[0015] In one embodiment of the first aspect, in step S4, the nodal state transition probability formula of the cascade diffusion derivation model is:

[0016] in, For a node to have a state Transfer to The transition probability; the parameters in the node state transition probability formula include: The reliability coefficient of the protection device; This will contribute to the incremental shift of trends; This refers to the rated transmission capacity of the branch line. This represents the probability of device loss. Specifically, by retrieving historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region, the statistical law of risk propagation time series under historical typhoon disasters of the same level is extracted, and the parameters in the node state transition probability formula are optimized.

[0017] The technical advantage of this embodiment is that it reduces the failure probability of weather-driven equipment. Operating characteristics of protection devices and the physical constraints of the current By incorporating a unified state transition probability framework, joint modeling of meteorology, electrical systems, and protection systems was achieved. In particular, by introducing the propagation time sequence patterns of historical typhoons of the same magnitude to optimize the model parameters, the inference model can reuse historical disaster evolution experience, thereby making the cascading diffusion process more consistent with the actual fault logic and solving the pain point of existing models that are "static topology inference and detached from historical review".

[0018] In one embodiment of the first aspect, step S5, which involves calculating the four-dimensional risk assessment index matrix to obtain the comprehensive risk proximity of each link, and classifying the risk level based on the proximity, includes: First, the objective weights of each risk indicator are calculated using the entropy weight method. The entropy weight formula for the item index is:

[0019] in, For the first The link is in the 1st Normalized ratio of the indicator; This represents the total number of propagation links; As evaluation indicator dimensions; Then, the relative proximity of each propagation link is calculated using the TOPSIS algorithm. Based on the relative proximity, the degree of danger is ranked and the risk level is classified; whereby, These are the Euclidean distances from the evaluation object to the optimal and worst ideal solutions, respectively. The greater the relative proximity, the higher the overall risk level of the link.

[0020] The technical advantages of this embodiment are as follows: It employs the entropy weighting method for objective weighting, avoiding the subjectivity of manually determining indicator weights. This ensures that the evaluation results rely entirely on the dispersion of the data itself, making it more objective and fair. Combined with the TOPSIS method to calculate relative proximity, it can quantitatively rank a large number of propagation links, accurately identifying high-risk blocking nodes, key weak points, and risk diffusion boundaries. This provides dispatchers with clear priorities and targets for implementing prevention and control measures such as "emergency isolation, backup support, and zoned control," effectively solving the problems of "lagging risk prediction and blind prevention and control" in existing technologies.

[0021] In one embodiment of the first aspect, the propagation breadth represents the number of affected nodes, the diffusion depth represents the number of cascade propagation levels, the propagation rate represents the number of newly added faulty nodes per unit time, and the power loss amplitude represents the total power loss caused by the fault.

[0022] The technical advantage of this embodiment lies in its construction of an evaluation system based on four orthogonal dimensions: spatial breadth (propagation breadth), temporal depth (diffusion depth), dynamic speed (spread rate), and energy intensity (power loss). This system enables multi-faceted and comprehensive quantification of the multi-level amplification and evolution of risk from "local nodes → areas → cities → urban clusters." This provides a scientific and comprehensive quantitative basis for subsequent risk level classification, avoiding the one-sidedness of single-indicator evaluations and making the identification results more valuable for decision-making.

[0023] Secondly, this application also proposes a device for identifying the cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids, comprising: The historical data collection and real-time data acquisition module is used to collect the topology parameters and operating condition data of the power grids of various cities in the urban agglomeration, as well as the real-time monitoring data of typhoon weather and air quality under the current typhoon event; and to collect historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region. The historical fault data includes historical failure records of various power equipment and power flow over-limit time sequence records of intercity tie lines. The initial risk triggering source identification module is used to characterize the spatiotemporal random evolution law of typhoon meteorological elements based on the real-time monitoring data of typhoon time and air meteorology and the historical meteorological data of typhoons over the years, and to combine the historical fault data of the power grid to correct the fitting parameters in the equipment vulnerability function, calculate the random failure probability of power equipment in each city power grid described by the topology parameters at different times, so as to screen out high-risk fault nodes as initial risk triggering sources. The intercity risk propagation directed weighted matrix construction module is used to calculate the active power flow transfer sensitivity, voltage coupling correlation degree and reactive power interaction coefficient of intercity node pairs based on the topology parameters and operating condition data collected by the historical data collection and real-time data acquisition module, so as to construct the inter-regional risk propagation directed weighted adjacency matrix of the urban agglomeration power grid and quantify the direction and weight of risk transmission between different nodes. The multi-scale cascaded diffusion simulation module is used to define the risk propagation neighborhood starting from the initial risk triggering source and using the directed weighted adjacency matrix. The neighborhood of node j consists of all nodes i with directed weights A_ij>0. A cascaded diffusion simulation model is constructed by fusing Markov random processes and improved cellular automata. The module retrieves historical typhoon meteorological data and historical power grid fault data collected by the historical data collection and real-time data acquisition modules. It extracts the risk propagation time-series statistical patterns under historical typhoon disasters of the same magnitude from these data. Based on these statistical patterns, it optimizes the parameters of the node state transition probability formula in the cascaded diffusion simulation model, and performs hierarchical simulation of the multi-scale cascaded process of risk spread within cities, between cities, and across the entire urban agglomeration, generating a risk propagation time-series path chain. The risk quantification and link identification module is used to calculate the four-dimensional risk indicators of propagation breadth, diffusion depth, spread rate, and power loss amplitude for each propagation link based on the risk propagation time-series path chain generated by the multi-scale cascade diffusion inference module, so as to construct a four-dimensional risk assessment index matrix. The entropy weight-TOPSIS algorithm is used to calculate the four-dimensional risk assessment index matrix to obtain the comprehensive risk proximity of each link. The risk level is divided according to the proximity, and high-risk blocking nodes, key weak sections, and risk diffusion boundaries are output to complete the identification of cross-regional propagation and cascade diffusion of risks.

[0024] Thirdly, this application also proposes a computer device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the method described in the first aspect.

[0025] Fourthly, this application also proposes a computer program product comprising a computer program that, when executed by a computer device, implements the steps of the method described in the first aspect.

[0026] It should be noted that the technical effects of the second to fourth aspects mentioned above are explained in the description of the technical effects of the first aspect, and will not be repeated here. Attached Figure Description

[0027] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1A flowchart illustrating a main embodiment of a method for identifying cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids provided in this application; Figure 2 This application provides a schematic diagram of the multi-spatial-scale cascade diffusion evolution mechanism. Figure 3 A structural block diagram of a device for identifying cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids provided in this application; Figure 4 A schematic diagram of an embodiment of the computer device provided in this application for implementing the method for identifying cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids. Detailed Implementation

[0029] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0030] It should be understood that, when used in this application specification and the appended claims, the term "comprising" indicates the presence of the described features, integrals, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or a collection thereof.

[0031] It should also be understood that the term “and / or” as used in this application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0032] As used in this application specification and the appended claims, the term "if" may be interpreted, depending on the context, as "when," "once," "in response to determination," or "in response to detection." Similarly, the phrase "if determined" or "if detected [the described condition or event]" may be interpreted, depending on the context, as meaning "once determined," "in response to determination," "once detected [the described condition or event]," or "in response to detection [the described condition or event]."

[0033] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), unless otherwise expressly and specifically defined.

[0034] Furthermore, in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0035] References to "one embodiment" or "some embodiments" in this specification mean that one or more embodiments of this application include a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, the phrases "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized.

[0036] The terms “including,” “comprising,” “having,” and variations thereof all mean “including but not limited to,” unless otherwise specifically emphasized.

[0037] To solve the aforementioned technical problems, firstly, refer to... Figure 1 and Figure 2 This application provides a method for identifying the cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids. This method is executed by a computer device and includes: Step S1 (Historical Data Collection and Real-time Data Acquisition): Collect the topology parameters and operating condition data of the power grids of each city in the urban agglomeration, as well as the real-time monitoring data of typhoon weather and air under the current typhoon event; and collect historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region. The historical fault data includes historical failure records of each power equipment and power flow over-limit time sequence records of intercity tie lines. In practical implementation, this step can be based on existing Energy Management Systems (EMS), Wide Area Measurement Systems (WAMS), and dynamic monitoring systems to collect real-time topological parameters (such as node, branch, and tie-line connections), equipment parameters (such as line impedance and transformer rated capacity), and real-time operating condition data (such as node voltage and branch power flow) of the power grids in each city within the urban agglomeration. Simultaneously, it can also collect real-time interactive data from intercity tie lines and obtain real-time meteorological monitoring data such as typhoon path, wind speed, and rainfall under the current typhoon event.

[0038] Secondly, historical meteorological data on typhoon passages over the years (including typhoon paths, wind speed sequences, and rainfall distribution) can be simultaneously collected from historical databases, along with historical power grid fault records for the region (including historical failure records of various power equipment, fault types, and times), and historical power flow overrun records for intercity connecting lines. Based on these massive historical samples, the model's basic parameters can be pre-calibrated, and the initial patterns of disaster evolution can be reviewed, providing statistical support for subsequent steps.

[0039] Finally, the power grids of each city are divided into independent electrical subnets, and the power nodes within the subnets and the coupling links between the subnets are defined. Combining equipment vulnerability, meteorological disturbance intensity and network topology margin, a multi-dimensional risk factor set of the urban agglomeration power grid is constructed, and the normalized model of the multi-subnet coupled topology is completed.

[0040] Step S2 (Initial Risk Trigger Source Identification): Based on the real-time monitoring data of typhoon weather and the historical meteorological data of typhoons over the years, a non-stationary Gaussian mixture model is used to characterize the spatiotemporal stochastic evolution law of typhoon meteorological elements. The fitting parameters in the equipment vulnerability function are corrected by combining the historical fault data of the power grid. The random failure probability of power equipment in each city power grid described by the topology parameters at different times is calculated to screen out high-risk fault nodes as initial risk trigger sources. It is understandable that this step is based on the multi-source data collected in step S1 to accurately identify the initial risk triggering source of the power grid under the spatiotemporal random disturbance of the typhoon; In practical implementation, a non-stationary Gaussian mixture model (GMM) can first be used to characterize the spatiotemporal stochastic evolution of typhoon meteorological elements (such as wind speed and rainfall). In the embodiments of this application, the non-stationary Gaussian mixture model can be understood as a meteorological model, specifically, the probability density function (probability density formula of meteorological random disturbance) of this meteorological model is expressed as:

[0041] in, For meteorological variables (wind speed or rainfall); For the first The weights of each Gaussian component; The time-varying expectation is used to characterize the non-stationary movement characteristics of typhoons; Standard deviation; The number of Gaussian components. Through time-series sampling, a continuous sequence of meteorological disturbances can be generated, providing external driving conditions for calculating the probability of equipment failure.

[0042] Secondly, based on the statistical data on equipment shutdown and failure under historical typhoon disasters collected in step S1, the fitting parameters in the equipment vulnerability model are corrected. This embodiment uses a piecewise nonlinear probability model to characterize the failure patterns of power equipment under typhoon conditions, and its vulnerability function is as follows:

[0043] In the formula: This represents the probability of random failure of the equipment. Real-time wind speed; These are the minimum wind speed threshold that the equipment can withstand and the maximum wind speed threshold that the equipment can withstand, respectively. The vulnerability shape fitting parameters can be used to differentiate the disaster resistance capabilities of different power equipment such as lines, transformers, and towers.

[0044] Then, based on the aforementioned meteorological model and vulnerability function, the probability of random failure of power equipment at different times and in different regions can be calculated. A dynamic threshold discrimination method is used to screen out high-risk fault nodes and distinguish three types of initial disturbance sources: meteorological direct faults, power flow induced faults, and coupled derivative faults. Finally, a time-series set of initial fault scenarios is generated as the initial risk triggering source for subsequent cascaded simulations.

[0045] Step S3 (Construction of Directed Weighted Matrix for Intercity Risk Propagation): Based on the topology parameters and operating condition data collected in Step S1, calculate the active power flow transfer sensitivity, voltage coupling correlation, and reactive power interaction coefficient of intercity node pairs to construct a directed weighted adjacency matrix for cross-regional risk propagation in the urban agglomeration power grid, so as to quantify the direction and weight of risk transmission between different nodes. Understandably, this step is based on electrical coupling sensitivity to construct a directed weighted adjacency matrix that quantifies the risk transmission capability between cities; Specifically, based on the topology parameters and operating condition data collected in step S1, the following three types of quantitative indicators can be calculated first: ① Active power flow transfer sensitivity ① Used to measure the impact of power changes at node i on power flow at node j; ② Voltage coupling sensitivity This is used to measure the degree of coupling between the voltage change at node i and the voltage at node j; ③ Reactive power interaction sensitivity. , used to measure the interactive impact of reactive power changes at node i on reactive power at node j; Secondly, by combining the above three types of indicators and optimizing the weighting ratio with reference to historical intercity power interaction anomaly data, a directed weighted adjacency matrix for cross-regional risk propagation is constructed. The elements of the directed weighted adjacency matrix... expression:

[0046] in, For nodes To the node Risk transmission weight; Sensitivity to active power flow transfer; Voltage coupling sensitivity; Reactive power interaction sensitivity; Assign coefficients to the weights, and satisfy the following conditions: The positive and negative signs of the matrix elements can respectively represent the two propagation modes of forward diffusion and reverse backflow, thereby quantifying the ease and direction of risk transmission at different nodes and different connection sections.

[0047] Step S4 (Multi-scale Cascaded Diffusion Inference Step): Starting from the initial risk triggering source, define the risk propagation neighborhood using the directed weighted adjacency matrix; wherein, the neighborhood of node j is all directed weighted adjacency matrices. For node i with a value greater than 0, a cascaded diffusion inference model is constructed by integrating Markov random processes and improved cellular automata. Historical meteorological data of typhoons from previous years and historical fault data of the power grid in the same region are retrieved from step S1. The risk propagation time series statistical law under historical typhoon disasters of the same level is extracted from the historical meteorological data of typhoons from previous years and historical fault data of the power grid in the same region. Based on the statistical law, the parameters of the node state transition probability formula in the cascaded diffusion inference model are optimized. The multi-scale cascaded process of risk spreading within the city, between cities, and across the entire urban agglomeration is inferred in layers to generate a risk propagation time series path chain. As can be understood, this step starts with the initial risk trigger source and uses the risk propagation neighborhood defined by the directed weighted adjacency matrix (the neighborhood of node j is all directed weighted adjacency matrices). (Node i > 0), integrate Markov stochastic process and improved cellular automata model to construct cascade diffusion extrapolation model to simulate the dynamic evolution of risk under continuous typhoon disturbance; As an example, we can first define the state set of a power grid node as {normal, overloaded, faulty, out of service}. The node state transition probability formula is as follows:

[0048] In the formula: For a node to have a state Transfer to The transition probability; The reliability coefficient of the protection device; This will contribute to the incremental shift of trends; This refers to the rated transmission capacity of the branch line. This represents the probability of device loss. Secondly, retrieve historical meteorological data of typhoons from step S1 and historical fault data of the power grid in the same region, and extract statistical patterns of risk propagation time series under historical typhoon disasters of the same magnitude (such as risk propagation paths and cascading evolution time series). Based on these statistical patterns, optimize the parameters in the above node state transition probability formula (such as...). , The statistical distribution of the model is used, and the neighborhood update rule of the cellular automata is combined to make the inference model more consistent with the real failure evolution logic of extreme disasters.

[0049] Finally, the multi-scale cascading process of risk is simulated through the following methods: ① the scale of intra-city spread, simulating the propagation of local faults within the urban power grid; ② the scale of inter-city propagation, simulating the penetration and transfer of risks to neighboring cities through inter-city links; ③ the scale of city cluster-wide spread, simulating the superposition and resonance of risks throughout the entire city cluster. The fault propagation evolution path is fully reproduced, recording the risk propagation time sequence nodes, jump links, propagation delays, and fault amplification factors, generating a visualized risk propagation time sequence path chain.

[0050] Step S5 (Risk Quantification and Link Identification): Based on the risk propagation time-series path chain generated in Step S4, calculate the four-dimensional risk indicators of propagation breadth, diffusion depth, spread rate, and power loss amplitude for each propagation link to construct a four-dimensional risk assessment index matrix; use the entropy weight-TOPSIS algorithm to calculate the four-dimensional risk assessment index matrix to obtain the comprehensive risk proximity of each link, classify the risk level according to the proximity, and output high-risk blocking nodes, key weak sections, and risk diffusion boundaries to complete the identification of cross-regional risk propagation and cascading diffusion.

[0051] It is understandable that this step, based on the propagation path chain generated in step S4, quantifies and rates the cascading risks and identifies high-risk propagation links. As an example, firstly, a four-dimensional risk assessment index system is constructed for each risk propagation link: ① propagation breadth (number of affected nodes), ② diffusion depth (number of cascaded levels), ③ propagation rate (number of faulty nodes per unit time), and ④ power loss amplitude (total power loss). This forms a four-dimensional risk assessment index matrix.

[0052] Secondly, this embodiment uses the entropy-weighted TOPSIS algorithm to comprehensively evaluate the above matrix; firstly, the weight of each index is calculated using the entropy-weighted method, and then... The entropy weight formula for the item index is:

[0053] in, For the first The link is in the 1st Normalized ratio of the indicator; This represents the total number of propagation links; The evaluation metrics are dimensional (this example uses M=4 for illustration); Then, the relative proximity of each link is calculated using the TOPSIS algorithm. :

[0054] in, , respectively, are the Euclidean distances from the evaluation object to the optimal ideal solution and the worst ideal solution. A higher value indicates a higher overall risk level for the link; Finally, based on proximity Based on the magnitude of the risk and historical cases of high-risk section failures during disasters, the risk is classified into three warning levels: minor spread, localized spread, and cascading collapse. High-risk blocking nodes, key weak sections, and risk spread boundaries are output, thereby enabling the identification of cross-regional propagation patterns and the assessment of cascading spread trends of urban power grid risks under typhoon scenarios.

[0055] Secondly, such as Figure 3 As shown, the present invention also provides a device for identifying the cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids, comprising: The historical data collection and real-time data acquisition module 10 is used to collect the topology parameters and operating condition data of the power grids of each city in the urban agglomeration, as well as the real-time monitoring data of the typhoon weather and air under the current typhoon event; and to collect historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region. The historical fault data includes historical failure records of each power equipment and power flow over-limit time sequence records of intercity tie lines. The initial risk triggering source identification module 20 is used to characterize the spatiotemporal random evolution law of typhoon meteorological elements by using a non-stationary Gaussian mixture model based on the real-time monitoring data of typhoon time and air meteorology and the historical meteorological data of typhoons over the years, and to combine the historical fault data of the power grid to correct the fitting parameters in the equipment vulnerability function, calculate the random failure probability of power equipment in each city power grid described by the topology parameters at different times, so as to screen out high-risk fault nodes as initial risk triggering sources. The intercity risk propagation directed weighted matrix construction module 30 is used to calculate the active power flow transfer sensitivity, voltage coupling correlation degree and reactive power interaction coefficient of intercity node pairs based on the topology parameters and operating condition data collected by the historical data collection and real-time data acquisition module, so as to construct the inter-regional risk propagation directed weighted adjacency matrix of the urban agglomeration power grid and quantify the direction and weight of risk transmission between different nodes. The multi-scale cascaded diffusion simulation module 40 is used to define the risk propagation neighborhood starting from the initial risk triggering source and using the directed weighted adjacency matrix; wherein, the neighborhood of node j is all nodes i with directed weights A_ij>0, and a cascaded diffusion simulation model is constructed by integrating Markov random processes and improved cellular automata; the historical meteorological data of typhoons over the years and the historical fault data of the power grid in the same region collected by the historical data collection and real-time data acquisition module are retrieved, and the risk propagation time series statistical law under historical typhoon disasters of the same level is extracted from the historical meteorological data of typhoons over the years and the historical fault data of the power grid in the same region. Based on the statistical law, the parameters of the node state transition probability formula in the cascaded diffusion simulation model are optimized, and the multi-scale cascaded process of risk spreading within the city, between cities, and throughout the urban agglomeration is simulated in layers, generating a risk propagation time series path chain; The risk quantification and link identification module 50 is used to calculate the four-dimensional risk indicators of propagation breadth, diffusion depth, spread rate, and power loss amplitude for each propagation link based on the risk propagation time-series path chain generated by the multi-scale cascade diffusion inference module, so as to construct a four-dimensional risk assessment index matrix; the entropy weight-TOPSIS algorithm is used to calculate the four-dimensional risk assessment index matrix to obtain the comprehensive risk proximity of each link, classify the risk level according to the proximity, and output high-risk blocking nodes, key weak sections, and risk diffusion boundaries to complete the identification of cross-regional propagation and cascade diffusion of risks.

[0056] It should be noted that the urban agglomeration power grid risk cross-regional propagation and cascading diffusion identification device in this embodiment can be understood as a software analysis tool, which can be installed in the computer device of the aforementioned embodiment. The computer device calls this device through the processor, and then runs the specific implementation scheme in the above embodiments of the urban agglomeration power grid risk cross-regional propagation and cascading diffusion identification method.

[0057] Thirdly, please refer to Figure 4 , Figure 4 A schematic diagram of an embodiment of a computer device for implementing a method for identifying the cross-regional propagation and cascading diffusion of risks in urban agglomeration power grids, as provided in this application, is shown below. Figure 4 As shown, the computer device 1 in this embodiment includes: at least one processor 10, a memory 11, and a computer program 12 stored in the memory 11 and executable on the at least one processor 10. When the processor 10 executes the computer program 12, it implements the steps in the embodiment of the method for identifying cross-regional propagation and cascading diffusion of urban agglomeration power grid risks in this application.

[0058] The computer device may be a desktop computer, laptop, smartphone, PDA, or other computing device.

[0059] Figure 4The computer device shown may include, but is not limited to, processor 10 and memory 11. Those skilled in the art will understand that... Figure 4 The computer device 1 is merely an example and does not constitute a limitation on the computer device 1. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0060] The information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and the resulting technical effects, please refer to the method embodiments section.

[0061] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is merely an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiments can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. Furthermore, the specific names of the functional units and modules are only for easy differentiation and are not intended to limit the scope of protection of this application. The specific working process of the units and modules in the above system can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0062] This application also provides a computer program product, which, when executed by a processor, implements the steps in the various method embodiments described above.

[0063] This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.

[0064] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include at least: any entity or device capable of carrying computer program code to a photographing device / terminal device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium. Examples include USB flash drives, portable hard drives, magnetic disks, or optical disks. In some jurisdictions, according to legislation and patent practice, computer-readable media cannot be electrical carrier signals or telecommunication signals.

[0065] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0066] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for identifying the cross-regional propagation and cascading diffusion of power grid risks in urban agglomerations, characterized in that, include: Collect topology parameters and operating condition data of the power grids in each city within the urban agglomeration, as well as real-time monitoring data of typhoon weather and atmospheric conditions during the current typhoon event; It also collects historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region. The historical fault data includes historical failure records of various power equipment and power flow over-limit time sequence records of intercity tie lines. Based on the real-time monitoring data of typhoon weather and the historical meteorological data of typhoons over the years, a non-stationary Gaussian mixture model is used to characterize the spatiotemporal random evolution law of typhoon meteorological elements. The fitting parameters in the equipment vulnerability function are corrected by combining the historical fault data of the power grid. The random failure probability of power equipment in each city power grid described by the topology parameters at different times is calculated to screen out high-risk fault nodes as the initial risk triggering source. Based on the topology parameters and operating condition data, the active power flow transfer sensitivity, voltage coupling correlation and reactive power interaction coefficient of inter-city node pairs are calculated to construct a directed weighted adjacency matrix for cross-regional risk propagation in urban agglomeration power grids, so as to quantify the direction and weight of risk transmission between different nodes. Starting from the initial risk trigger source, the risk propagation neighborhood is defined by the directed weighted adjacency matrix; wherein, the neighborhood of node j is all directed weighted adjacency matrices. For node i with a value greater than 0, a cascaded diffusion inference model is constructed by integrating Markov random processes and improved cellular automata. Historical meteorological data of typhoons from previous years and historical fault data of the power grid in the same region are retrieved from step S1. The risk propagation time series statistical law under historical typhoon disasters of the same level is extracted from the historical meteorological data of typhoons from previous years and historical fault data of the power grid in the same region. Based on the statistical law, the parameters of the node state transition probability formula in the cascaded diffusion inference model are optimized. The multi-scale cascaded process of risk spreading within the city, between cities, and across the entire urban agglomeration is inferred in layers to generate a risk propagation time series path chain. Based on the risk propagation timeline path chain, four-dimensional risk indicators—propagation breadth, diffusion depth, spread rate, and power loss amplitude—are calculated for each propagation link to construct a four-dimensional risk assessment indicator matrix. The four-dimensional risk assessment indicator matrix is ​​then calculated to obtain the comprehensive risk proximity of each link. Based on the proximity, risk levels are classified, and high-risk blocking nodes, key weak sections, and risk diffusion boundaries are output to complete the identification of cross-regional risk propagation and cascading diffusion.

2. The method as described in claim 1, characterized in that, The probability density function of the non-stationary Gaussian mixture model is: ; in, For meteorological element variables; For the first The weights of each Gaussian component; For time-varying expectations; Standard deviation; The number of Gaussian components.

3. The method as described in claim 2, characterized in that, The device vulnerability function is a piecewise nonlinear probability model: in, The probability of random failure of electrical equipment; Real-time wind speed; These are the minimum wind speed threshold that the equipment can withstand and the maximum wind speed threshold that the equipment can withstand, respectively. The parameters are fitted to the fragile shape.

4. The method as described in claim 1, characterized in that, The elements of the directed weighted adjacency matrix Calculated using the following formula: in, For nodes To the node Risk transmission weight; Sensitivity to active power flow transfer; Voltage coupling sensitivity; Reactive power interaction sensitivity; Assign coefficients to the weights, and satisfy the following conditions: .

5. The method as described in claim 1, characterized in that, The formula for the node state transition probability of the cascaded diffusion model is as follows: in, For a node to have a state Transfer to The transition probability; the parameters in the node state transition probability formula include: The reliability coefficient of the protection device; This will contribute to the incremental shift of trends; This refers to the rated transmission capacity of the branch line. This represents the probability of device loss. Specifically, by retrieving historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region, the statistical law of risk propagation time series under historical typhoon disasters of the same level is extracted, and the parameters in the node state transition probability formula are optimized.

6. The method as described in claim 1, characterized in that, The calculation of the four-dimensional risk assessment index matrix yields the comprehensive risk proximity of each link, and the risk level is classified according to the magnitude of the proximity, including: The objective weights of each risk indicator are calculated using the entropy weight method. The entropy weight formula for the item index is: in, For the first The link is in the 1st Normalized ratios for each indicator; This represents the total number of propagation links; As evaluation indicator dimensions; The relative proximity of each propagation link is calculated using the TOPSIS algorithm. Based on the relative proximity, the degree of danger is ranked and the risk level is classified; whereby, These are the Euclidean distances from the evaluation object to the optimal and worst ideal solutions, respectively. The greater the relative proximity, the higher the overall risk level of the link.

7. The method as described in any one of claims 1 to 6, characterized in that, The propagation breadth represents the number of affected nodes, the propagation depth represents the number of cascade propagation levels, the propagation rate represents the number of newly added faulty nodes per unit time, and the power loss amplitude represents the total power loss caused by the fault.

8. A device for identifying the cross-regional propagation and cascading diffusion of risks in urban power grids, characterized in that, include: The historical data collection and real-time data acquisition module is used to collect the topology parameters and operating condition data of the power grids of various cities in the urban agglomeration, as well as the real-time monitoring data of typhoon weather and air quality under the current typhoon event; and to collect historical meteorological data of typhoons over the years and historical fault data of the power grid in the same region. The historical fault data includes historical failure records of various power equipment and power flow over-limit time sequence records of intercity tie lines. The initial risk triggering source identification module is used to characterize the spatiotemporal random evolution law of typhoon meteorological elements based on the real-time monitoring data of typhoon time and air meteorology and the historical meteorological data of typhoons over the years, and to combine the historical fault data of the power grid to correct the fitting parameters in the equipment vulnerability function, calculate the random failure probability of power equipment in each city power grid described by the topology parameters at different times, so as to screen out high-risk fault nodes as initial risk triggering sources. The intercity risk propagation directed weighted matrix construction module is used to calculate the active power flow transfer sensitivity, voltage coupling correlation degree and reactive power interaction coefficient of intercity node pairs based on the topology parameters and operating condition data collected by the historical data collection and real-time data acquisition module, so as to construct the inter-regional risk propagation directed weighted adjacency matrix of the urban agglomeration power grid and quantify the direction and weight of risk transmission between different nodes. The multi-scale cascaded diffusion simulation module is used to define the risk propagation neighborhood starting from the initial risk triggering source and using the directed weighted adjacency matrix. The neighborhood of node j consists of all nodes i with directed weights A_ij>0. A cascaded diffusion simulation model is constructed by fusing Markov random processes and improved cellular automata. The module retrieves historical typhoon meteorological data and historical power grid fault data collected by the historical data collection and real-time data acquisition modules. It extracts the risk propagation time-series statistical patterns under historical typhoon disasters of the same magnitude from these data. Based on these statistical patterns, it optimizes the parameters of the node state transition probability formula in the cascaded diffusion simulation model, and performs hierarchical simulation of the multi-scale cascaded process of risk spread within cities, between cities, and across the entire urban agglomeration, generating a risk propagation time-series path chain. The risk quantification and link identification module is used to calculate the four-dimensional risk indicators of propagation breadth, diffusion depth, spread rate, and power loss amplitude for each propagation link based on the risk propagation time-series path chain generated by the multi-scale cascade diffusion inference module, so as to construct a four-dimensional risk assessment index matrix; calculate the comprehensive risk proximity of each link by the four-dimensional risk assessment index matrix, classify the risk level according to the proximity, and output high-risk blocking nodes, key weak sections, and risk diffusion boundaries to complete the identification of cross-regional propagation and cascade diffusion of risks.

9. A computer device 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 computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a computer device, implements the steps of the method as described in any one of claims 1 to 7.

Images