Method and system for power network dynamic stability assessment based on data analysis

Abstract

The application relates to a power network dynamic stability evaluation method and system based on data analysis. Real-time operation data, environment data and a topology structure of a power network to be evaluated are acquired, and a dynamic characteristic map is constructed according to the real-time operation data, the environment data and the topology structure. The map performs space-time correlation on electrical changes of each power grid node in the power network, linkage relationships between each power grid node in the topology structure and environmental factor changes. The dynamic characteristic map is subjected to similarity matching with a preset historical disturbance event library, a stability classification signal is generated according to a matching result, a control parameter optimization strategy is generated based on the stability classification signal, and the control parameter of electronic equipment of the power network is adjusted according to the control parameter optimization strategy. The method can accurately judge the stability of the power system and simultaneously provide a corresponding adjustment control method.

Description

Technical Field

[0001] This application relates to the field of power network technology, and in particular to a method and system for evaluating the dynamic stability of power networks based on data analysis. Background Technology

[0002] Current power system stability assessment technologies are primarily based on two methodologies: The first is a real-time monitoring system based on electrical quantity thresholds. This system collects core parameters such as voltage, current, and power, and sets fixed safety thresholds to trigger alarm mechanisms. This type of technology relies on monitoring systems such as SCADA for basic data acquisition and has the advantage of fast response speed when dealing with explicit faults such as overvoltage and low frequency, but it is essentially a passive defense. The second is a stability analysis method based on equipment modeling. This method analyzes the dynamic response characteristics of the system under disturbances by establishing impedance models or state-space equations for power electronic equipment. This method explains the causes of instability at the mechanistic level and has led to several standardized solutions in areas such as oscillation suppression in renewable energy grid-connected systems.

[0003] However, with the high proportion of renewable energy integration and the frequent occurrence of extreme weather events, existing technologies are gradually revealing three limitations: First, the electrical quantity threshold monitoring method only focuses on single parameter exceedances and cannot capture the risks associated with multi-dimensional data. For example, when lightning strikes cause line tripping, events such as sudden weather changes, topology reconfiguration, and power transfer often occur coupled within a second-level timescale, but traditional monitoring systems, due to fragmented data dimensions, struggle to identify the precursory characteristics of such complex disturbances. Second, model-driven methods heavily rely on the accuracy of equipment parameters, while the control strategies of power electronic equipment such as photovoltaic inverters and energy storage converters differ significantly in actual systems. Unified modeling will lead to analysis results deviating from actual operating conditions. More critically, existing technologies generally neglect the influence mechanism of environmental factors—power fluctuations in transmission lines in strong wind areas are strongly correlated with meteorological parameters, but current assessment models do not incorporate external variables such as meteorology and geography into the analysis framework, resulting in the frequent underestimation of risks such as frequency instability during cold waves.

[0004] In existing technologies, such defects lead to two types of systemic risks for power networks: at the data level, the isolated processing of multi-source heterogeneous information makes it difficult to capture latent instability characteristics in a timely manner. For example, during a continuous voltage drop, traditional monitoring may only trigger an alarm when a preset threshold is reached, missing the optimal control window; at the decision-making level, model mismatch and missing environmental variables reduce the credibility of stability assessment results, which in turn leads to a misalignment between control strategies and actual needs. Summary of the Invention

[0005] Therefore, it is necessary to provide a data analysis-based method and system for assessing the dynamic stability of power networks that can take into account the actual comprehensive situation and thus achieve highly accurate assessment, in order to address the above-mentioned technical problems.

[0006] A data analysis-based method for assessing the dynamic stability of power networks, the method comprising: The system acquires real-time operational data, environmental data, and the topology of the power network to be evaluated. The real-time operational data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. Based on the real-time operating data, environmental data, and topology, a dynamic feature map is constructed. The dynamic feature map spatiotemporally correlates the electrical changes at each power grid node in the power network, the link relationships between each power grid node in the topology, and the changes in environmental factors. The dynamic feature map is matched with a preset historical disturbance event database for similarity, and a stability classification signal is generated based on the matching result. A control parameter optimization strategy is generated based on the stability grading signal, and the control parameters of the electronic equipment of the power network are adjusted according to the control parameter optimization strategy.

[0007] In one embodiment, constructing a dynamic feature map based on the real-time running data, environmental data, and topology includes: Key nodes on the topology are extracted according to preset requirements, and the extracted key nodes are used as nodes of the dynamic feature map. Based on the real-time operating data, the rate of change of electrical quantities at the key nodes is used as the time dimension feature of the nodes in the dynamic feature map, and the electrical quantity propagation relationship between adjacent nodes in the dynamic feature map is used as the spatial dimension feature. Based on environmental data, meteorological change trends are correlated with corresponding nodes in the dynamic feature map according to geographical location, serving as environmental disturbance features.

[0008] In one embodiment, when performing similarity matching between the dynamic feature map and a preset historical disturbance event database, different levels of matching are performed sequentially. When the matching threshold of a certain level is met, the matching of the next level is performed. When the matching condition of the current level is not met, the matching process is terminated, and the matching result of the previous level is used as the final determination.

[0009] In one embodiment, when performing similarity matching between the dynamic feature map and a preset historical disturbance event database, there are three matching levels: primary matching, intermediate matching, and advanced matching. The primary matching includes: when the real-time electrical quantity of a node in the dynamic feature map exceeds a preset safety threshold, a primary alarm is triggered. The intermediate matching includes: identifying the association pattern between environmental disturbance features and time dimension features at a certain node in the dynamic feature map, and matching the association pattern with historical association patterns in the historical disturbance event database. If the matching result is a strong association, advanced matching is performed; if the matching result is a weak association, the primary matching result is maintained. The advanced matching includes: finding corresponding historical instability events in the historical disturbance event database based on strong correlations, performing scenario deduction, and generating stability grading signals.

[0010] In one embodiment, when generating a control parameter optimization strategy based on the stability grading signal: The adjustment target is selected based on the stability grading signal, and the adjustment target includes voltage support strength or frequency response speed. It also correlates with the current operating conditions of related power electronic equipment to generate adjustment amounts for inverter droop coefficients or reactive power compensation parameters.

[0011] In one embodiment, selecting the adjustment target based on the stability grading signal includes: When the stability grading signal indicates a risk of voltage instability, the reactive current compensation parameter is adjusted first. When the stability grading signal indicates a risk of frequency instability, the active power droop coefficient is adjusted preferentially.

[0012] This application also provides a power network dynamic stability assessment system based on data analysis, the system comprising: The data fusion module is used to acquire real-time operating data, environmental data, and the topology of the power network to be evaluated. The real-time operating data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. The graph construction module is used to construct a dynamic feature graph based on the real-time operating data, environmental data and topology. The dynamic feature graph spatiotemporally correlates the electrical changes on each power grid node in the power network, the link relationships between each power grid node in the topology and the changes in environmental factors. The evaluation and analysis module is used to perform similarity matching between the dynamic feature map and a preset historical disturbance event database, and generate a stability grading signal based on the matching results. The strategy generation module is used to generate a control parameter optimization strategy based on the stability grading signal, and to adjust the control parameters of the electronic equipment of the power network according to the control parameter optimization strategy.

[0013] In one embodiment, the evaluation and analysis module includes a threshold comparison unit, a pattern matching unit, and a scenario deduction unit.

[0014] In one embodiment, the threshold comparison unit is used to trigger a primary alarm; The pattern matching unit is linked with the threshold comparison unit. When the threshold comparison unit triggers a primary alarm, the pattern matching unit calls a preset historical disturbance event library to identify the association pattern between environmental disturbance features and time dimension features at a certain node in the dynamic feature map, and matches the association pattern with the historical association patterns in the historical disturbance event library. If the matching result is a strong association, advanced matching is performed; if the matching result is a weak association, the primary matching result is maintained. The scenario deduction unit works in conjunction with the pattern matching unit. When the pattern matching unit determines that the matching result is a strong correlation, it finds the corresponding historical instability event in the historical disturbance event database based on the strong correlation, performs scenario deduction, and generates a stability grading signal.

[0015] The aforementioned data analysis-based method and system for assessing the dynamic stability of power networks acquires real-time operational data, environmental data, and the topology of the power network to be assessed. Based on these data, a dynamic feature map is constructed. This map spatiotemporally correlates electrical changes at each grid node, the connections between grid nodes in the topology, and changes in environmental factors. The dynamic feature map is then matched with a pre-defined database of historical disturbance events to generate a stability grading signal. Based on this signal, a control parameter optimization strategy is generated, and the control parameters of the power network's electronic equipment are adjusted according to the optimization strategy. This method allows for accurate assessment of power system stability and provides corresponding adjustment control methods. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a data analysis-based dynamic stability assessment method for power networks in one embodiment. Figure 2 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0018] In this application, as Figure 1 As shown, a data analysis-based method for evaluating the dynamic stability of power networks is provided, including the following steps: Step S100: Obtain real-time operating data, environmental data, and the topology of the power network to be evaluated. The real-time operating data includes the spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters.

[0019] Step S110: Construct a dynamic feature map based on real-time operating data, environmental data, and topology. The dynamic feature map spatiotemporally correlates electrical changes at each power grid node in the power network, the link relationships between each power grid node in the topology, and changes in environmental factors.

[0020] Step S120: Perform similarity matching between the dynamic feature map and the preset historical disturbance event database, and generate a stability classification signal based on the matching result.

[0021] Step S130: Generate a control parameter optimization strategy based on the stability grading signal, and adjust the control parameters of the electronic equipment of the power network according to the control parameter optimization strategy.

[0022] This application overcomes the limitations of traditional single electrical quantity monitoring by integrating power grid operation data, environmental data, and topology. It constructs a dynamic feature map to achieve multi-dimensional risk correlation characterization, and generates tiered early warning signals based on the matching of the map with a historical event database, significantly improving the accuracy of identifying latent instability risks. Finally, a control optimization strategy is generated to form closed-loop control, solving the response lag problem of traditional methods.

[0023] In step S100, multi-dimensional data related to power system stability is acquired. Among these, the spatiotemporal variations of voltage, current, and power directly reflect electrical changes. Electrical changes essentially refer to changes in the state of electrical quantities (voltage, current, power, etc.) in the power grid, and their spatiotemporal variations are extensions of these changes in time and space. For example, a significant drop in voltage at a critical node within a short period directly reflects the electrical change occurring at that node. Changes in the direction of current propagation between adjacent nodes also represent spatial manifestations of electrical changes. Environmental data, including meteorological conditions and geographical parameters, represents external factors affecting power network stability.

[0024] In this embodiment, the topology of the power network is composed of various components in the power system and their connections. The topology includes nodes, branches, and connections. Nodes are key connection points in the power network, serving as core units for power conversion, distribution, or control, including generation nodes, substation nodes, load nodes, and converter nodes. Branches are electrical pathways connecting different nodes, acting as carriers of power transmission, and include transmission lines, distribution lines, transformers, and switching equipment. Connections refer to the combination of nodes and branches, determining the network's topology, primarily manifested in physical connections and topology type.

[0025] In step S110, to reflect the dynamic changes of the power network during operation and better provide underlying support for subsequent risk assessment, fault diagnosis, and decision optimization, a dynamic feature map is constructed based on real-time operating data, environmental data, and topology. This dynamic feature map spatiotemporally correlates electrical changes at each grid node, the link relationships between grid nodes in the topology, and changes in environmental factors. By extracting the rate of change of electrical quantities, spatial propagation relationships, and environmental binding features, a strong correlation is established between meteorological / geographical disturbances and power grid dynamics, resolving misjudgments caused by environmental factors. The spatiotemporal feature fusion enables the map to predict disturbance propagation paths, providing accurate input for subsequent matching.

[0026] In this embodiment, when constructing a dynamic feature map based on real-time operating data, environmental data, and topology, the process includes: extracting key nodes on the topology according to preset requirements, and using the extracted key nodes as nodes of the dynamic feature map; based on real-time operating data, using the rate of change of electrical quantities on the key nodes as the time dimension feature of the nodes in the dynamic feature map; simultaneously, using the electrical quantity propagation relationship between adjacent nodes in the dynamic feature map as the spatial dimension feature; and based on environmental data, associating meteorological change trends with corresponding nodes in the dynamic feature map according to geographical location as environmental disturbance features.

[0027] Specifically, the selection criteria for key nodes in the power network topology can be set according to the specific application environment. For example, key nodes can be selected based on load importance, topological centrality, or fault sensitivity, and used as nodes in the dynamic feature graph.

[0028] Specifically, when mapping real-time operational data and environmental data to nodes in the dynamic feature map, both data are preprocessed. Outlier removal and time alignment are performed on the real-time operational data (spatiotemporal variation information of voltage, current, and power). Environmental data is integrated to match the time of the real-time operational data, and geographic environmental parameters are precisely bound to the node locations in the topology.

[0029] Furthermore, in the time dimension, based on preprocessed power operation data, the rate of change of voltage, current, and power at key nodes is calculated per unit time, and trend characteristics are captured using a sliding window. In the spatial dimension, based on a standardized network topology, the propagation relationship of electrical quantities between adjacent nodes is analyzed through power flow direction, impedance correlation, and electrical quantity difference ratios to establish spatial correlation characteristics. Regarding environmental disturbance characteristics, integrated meteorological data is bound to the location of power grid nodes and combined with geographical environmental parameters to generate [the data].

[0030] In this embodiment, the time dimension features are also updated according to the data acquisition frequency based on the dynamic update mechanism, and the spatial dimension features and environmental disturbance features are updated respectively when the network topology, meteorological conditions and geographical environment parameters change.

[0031] In step S120, a three-level matching mechanism is adopted to achieve progressive risk assessment of the power network to avoid overlooking risks of different levels. During the judgment, the dynamic feature map is matched with a preset historical disturbance event database for similarity. Matching is performed at different levels in sequence. When the matching threshold of a certain level is met, the matching of the next level is performed. If the matching condition of the current level is not met, the matching process is terminated, and the matching result of the previous level is used as the final judgment.

[0032] In this embodiment, when performing similarity matching between the dynamic feature map and a preset historical disturbance event database, three matching levels are included: primary matching, intermediate matching, and advanced matching. Primary matching involves triggering a primary alarm when the real-time electrical quantity of a node in the dynamic feature map exceeds a preset safety threshold. Intermediate matching involves identifying the association pattern between environmental disturbance features and time-dimensional features at a node in the dynamic feature map, and matching this association pattern with historical association patterns in the historical disturbance event database. If the matching result is a strong association, advanced matching is performed; otherwise, the primary matching result is maintained. Advanced matching involves finding corresponding historical instability events in the historical disturbance event database based on strong associations, performing scenario deduction, and generating a stability grading signal.

[0033] Specifically, in intermediate-level matching, the correlation pattern refers to the dynamic coupling relationship between meteorological / geographical environmental parameters and the electrical response of power grid nodes, quantified by the Pearson correlation coefficient ρX,Y: if ρX,Y>0.7, it is determined to be a strong correlation pattern; if 0.5≤ρX,Y≤0.7, it is determined to be a weak correlation pattern, where X is an environmental parameter vector (e.g., wind speed, humidity), and Y is an electrical quantity change vector (e.g., power, voltage). The determination of this coefficient integrates data-driven analysis and domain expert experience. In intermediate-level matching, the threshold for the Pearson correlation coefficient (e.g., >0.7 for strong correlation) is set by analyzing a large amount of historical environmental and power grid data, combining expert labeling results for "strong correlation" and "weak correlation," and referring to general statistical standards for classifying correlation strength. The aim is to ensure accurate capture of the true correlation between environmental disturbances and power grid response.

[0034] Specifically, during advanced matching, a safety signal is output when the similarity is <0.6, a warning signal is output when the similarity is 0.6 ≤ similarity <0.8, triggering device-level parameter adjustments, and an emergency control signal is output when the similarity is ≥0.8, triggering topology reconfiguration and energy storage scheduling. These output signals constitute the system's stability grading signals, used to guide subsequent precise control.

[0035] Furthermore, during advanced matching, similarity thresholds (e.g., >0.8 as an emergency control signal) are determined by constructing a multi-dimensional similarity model, combining it with the consequence classification of historical instability events, and after backtesting and risk balancing. When setting these thresholds, it is necessary to balance the risk of missed detections (failure to identify dangers in a timely manner) with the cost of false alarms (triggering unnecessary control measures) to ensure that the system can adopt the most appropriate response strategy at different similarity levels.

[0036] Specifically, the multidimensional similarity model is a comprehensive similarity calculation method that considers not only single-feature similarity between entities but also integrates multiple dimensions of similarity measures, such as attribute similarity, relational similarity, and semantic similarity. It is particularly suitable for situation determination in complex scenarios. When constructing a multidimensional similarity model, data preparation and preprocessing are first performed, including preprocessing and constructing a feature set. Then, multidimensional similarity calculations are performed on different dimensions (attributes, relations, semantics, key attribute values, etc.). Next, similarity fusion and weighting are performed, followed by similarity ranking and determination.

[0037] In this embodiment, constructing a multidimensional similarity model includes first constructing a feature set of historical instability events, such as electrical quantities like voltage, frequency, and power, and their rates of change. Then, calculating various similarities between the current system state and historical events, such as Euclidean distance and cosine similarity. Finally, weighted fusion of the similarities from each dimension yields a comprehensive similarity value, and a determination is made based on a preset threshold (e.g., >0.8) to determine whether emergency control should be triggered.

[0038] In step 130, when generating control parameter optimization strategies based on stability grading signals: select adjustment targets according to stability grading signals, including voltage support strength or frequency response speed, and associate them with the current operating conditions of relevant power electronic equipment to generate adjustment amounts for inverter droop coefficients or reactive power compensation parameters.

[0039] In this embodiment, selecting the adjustment target based on the stability grading signal includes: when the stability grading signal indicates a risk of voltage instability, the reactive current compensation parameter is adjusted first; when the stability grading signal indicates a risk of frequency instability, the active power droop coefficient is adjusted first.

[0040] Specifically, in the advanced matching stage, multi-dimensional feature vectors of the current operating state are extracted through dynamic feature maps, and cosine similarity is calculated between these vectors and the feature vectors of typical events in the historical disturbance event database. The historical disturbance event database contains three key types of data: first, power grid topology abrupt change events (such as line N-1 faults, main transformer commissioning / decommissioning, etc.), recording node voltage and branch power flow distribution before the event; second, renewable energy output fluctuation events (such as sudden power drops in wind farms, sudden power changes in photovoltaic clusters, etc.), including dynamic features such as power change rate and frequency deviation before and after the fluctuation; and third, abnormal load disturbance events (such as industrial load shocks, disordered charging of electric vehicle clusters, etc.), covering features such as load abrupt change amplitude and three-phase imbalance. By quantitatively comparing the real-time feature vectors with the historical event feature vectors, the similarity calculation results directly reflect the degree of similarity between the current state and historical dangerous events.

[0041] Furthermore, based on the similarity calculation results, the system constructs a three-level risk response mechanism: when the similarity is below 0.6, it is determined that the current state deviates significantly from historical dangerous events, outputs a safety signal, and maintains the existing operating mode; when the similarity is in the range of 0.6-0.8, it is identified as a potential risk state, outputs a warning signal, and triggers equipment-level parameter adjustments, such as dynamically adjusting the AVC voltage setpoint and optimizing generator excitation parameters; when the similarity reaches 0.8 or above, it is determined as a high-risk state, outputs an emergency control signal, automatically initiates topology reconfiguration (such as disconnecting weakly connected areas and closing the backup tie line), and calls upon the energy storage system for rapid power support. This hierarchical response mechanism, by quantifying the correlation strength between similarity and historical events, realizes a closed-loop logic from "feature matching - risk assessment - control decision," ensuring the accuracy of stability signals and the effectiveness of control measures.

[0042] The aforementioned data analysis-based dynamic stability assessment method for power networks systematically addresses three core issues in power network stability assessment: data fragmentation, model mismatch, and lack of environmental adaptability by establishing a closed-loop assessment mechanism that integrates multi-source data fusion, dynamic feature map construction, and historical event-driven approaches.

[0043] First, by integrating real-time electrical quantities, topology connections, and environmental parameters of the power grid into a unified analysis framework, we can overcome the limitations of single-dimensional data in traditional monitoring, achieve a strong correlation mapping between meteorological and geographical disturbances and electrical responses, and significantly improve the early identification capability of hidden instability risks.

[0044] Secondly, dynamic modeling based on spatiotemporal feature maps replaces the rigid analysis mode of static thresholds and fixed equipment parameters. Through the self-organized correlation of electrical quantity change rate, spatial propagation path and environmental binding features, it accurately depicts the propagation law of complex disturbances between multiple nodes, fundamentally avoiding the evaluation bias caused by model mismatch.

[0045] Finally, relying on the intelligent matching and three-level progressive evaluation mechanism of the historical event database, the real-time spectrum features and historical disturbance patterns are compared and similarity is extrapolated to generate a stability classification signal that is both predictive and adaptive. Based on this, the control parameters of the power electronic equipment are automatically optimized, forming a complete defense chain of "risk identification - strategy generation - closed-loop control".

[0046] This method enables millisecond-level online assessment and adaptive control of the dynamic stability of the power grid. It overcomes the inherent defects of traditional methods in the absence of environmental variables and significantly reduces the maintenance costs caused by system expansion and equipment iteration, providing universal protection for the power grid security under high-proportion renewable energy access.

[0047] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0048] In one embodiment, a power network dynamic stability assessment system based on data analysis is provided, comprising: a data fusion module, a graph construction module, an assessment and analysis module, and a strategy generation module, wherein: The data fusion module is used to acquire real-time operating data, environmental data, and the topology of the power network to be evaluated. The real-time operating data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. The graph construction module is used to construct a dynamic feature graph based on the real-time operating data, environmental data and topology. The dynamic feature graph spatiotemporally correlates the electrical changes on each power grid node in the power network, the link relationships between each power grid node in the topology and the changes in environmental factors. The evaluation and analysis module is used to perform similarity matching between the dynamic feature map and a preset historical disturbance event database, and generate a stability grading signal based on the matching results. The strategy generation module is used to generate a control parameter optimization strategy based on the stability grading signal, and to adjust the control parameters of the electronic equipment of the power network according to the control parameter optimization strategy.

[0049] In this embodiment, the evaluation and analysis module includes a threshold comparison unit, a pattern matching unit, and a scene deduction unit.

[0050] In this embodiment, the threshold comparison unit is used to trigger a primary alarm. The pattern matching unit works in conjunction with the threshold comparison unit. When the threshold comparison unit triggers the primary alarm, the pattern matching unit calls a preset historical disturbance event library to identify the association pattern between environmental disturbance features and time dimension features at a certain node in the dynamic feature map. It then matches the association pattern with historical association patterns in the historical disturbance event library. If the matching result is a strong association, advanced matching is performed; if the matching result is a weak association, the primary matching result is maintained. The scenario inference unit works in conjunction with the pattern matching unit. When the pattern matching unit determines that the matching result is a strong association, it finds the corresponding historical instability event in the historical disturbance event library based on the strong association, performs scenario inference, and generates a stability grading signal.

[0051] Specific limitations regarding the data analysis-based power network dynamic stability assessment system can be found in the limitations of the data analysis-based power network dynamic stability assessment method described above, and will not be repeated here. Each module in the aforementioned data analysis-based power network dynamic stability assessment system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0052] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 2 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a data analysis-based method for assessing the dynamic stability of power networks. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0053] Those skilled in the art will understand that Figure 2The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0054] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps: The system acquires real-time operational data, environmental data, and the topology of the power network to be evaluated. The real-time operational data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. Based on the real-time operating data, environmental data, and topology, a dynamic feature map is constructed. The dynamic feature map spatiotemporally correlates the electrical changes at each power grid node in the power network, the link relationships between each power grid node in the topology, and the changes in environmental factors. The dynamic feature map is matched with a preset historical disturbance event database for similarity, and a stability classification signal is generated based on the matching result. A control parameter optimization strategy is generated based on the stability grading signal, and the control parameters of the electronic equipment of the power network are adjusted according to the control parameter optimization strategy.

[0055] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor: The system acquires real-time operational data, environmental data, and the topology of the power network to be evaluated. The real-time operational data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. Based on the real-time operating data, environmental data, and topology, a dynamic feature map is constructed. The dynamic feature map spatiotemporally correlates the electrical changes at each power grid node in the power network, the link relationships between each power grid node in the topology, and the changes in environmental factors. The dynamic feature map is matched with a preset historical disturbance event database for similarity, and a stability classification signal is generated based on the matching result. A control parameter optimization strategy is generated based on the stability grading signal, and the control parameters of the electronic equipment of the power network are adjusted according to the control parameter optimization strategy.

[0056] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0057] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for evaluating the dynamic stability of power networks based on data analysis, characterized in that, The method includes: The system acquires real-time operational data, environmental data, and the topology of the power network to be evaluated. The real-time operational data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. Based on the real-time operating data, environmental data, and topology, a dynamic feature map is constructed. The dynamic feature map spatiotemporally correlates the electrical changes at each power grid node in the power network, the link relationships between each power grid node in the topology, and the changes in environmental factors. The dynamic feature map is matched with a preset historical disturbance event database for similarity, and a stability classification signal is generated based on the matching result. A control parameter optimization strategy is generated based on the stability grading signal, and the control parameters of the electronic equipment of the power network are adjusted according to the control parameter optimization strategy.

2. The power network dynamic stability assessment method based on data analysis according to claim 1, characterized in that, When constructing a dynamic feature map based on the real-time operating data, environmental data, and topology, the following are included: Key nodes on the topology are extracted according to preset requirements, and the extracted key nodes are used as nodes of the dynamic feature map. Based on the real-time operating data, the rate of change of electrical quantities at the key nodes is used as the time dimension feature of the nodes in the dynamic feature map, and the electrical quantity propagation relationship between adjacent nodes in the dynamic feature map is used as the spatial dimension feature. Based on environmental data, meteorological change trends are correlated with corresponding nodes in the dynamic feature map according to geographical location, serving as environmental disturbance features.

3. The power network dynamic stability assessment method based on data analysis according to claim 1, characterized in that, When performing similarity matching between the dynamic feature map and the preset historical disturbance event database, different levels of matching are performed sequentially. When the matching threshold of a certain level is met, the matching of the next level is performed. When the matching condition of the current level is not met, the matching process is terminated, and the matching result of the previous level is used as the final judgment.

4. The power network dynamic stability assessment method based on data analysis according to claim 3, characterized in that, When performing similarity matching between the dynamic feature map and the preset historical disturbance event database, there are three matching levels: primary matching, intermediate matching, and advanced matching. The primary matching includes: when the real-time electrical quantity of a node in the dynamic feature map exceeds a preset safety threshold, a primary alarm is triggered. The intermediate matching includes: identifying the association pattern between environmental disturbance features and time dimension features at a certain node in the dynamic feature map, and matching the association pattern with historical association patterns in the historical disturbance event database. If the matching result is a strong association, advanced matching is performed; if the matching result is a weak association, the primary matching result is maintained. The advanced matching includes: finding corresponding historical instability events in the historical disturbance event database based on strong correlations, performing scenario deduction, and generating stability grading signals.

5. The power network dynamic stability assessment method based on data analysis according to claim 1, characterized in that, When generating a control parameter optimization strategy based on the stability grading signal: The adjustment target is selected based on the stability grading signal, and the adjustment target includes voltage support strength or frequency response speed. It also correlates with the current operating conditions of related power electronic equipment to generate adjustment amounts for inverter droop coefficients or reactive power compensation parameters.

6. The power network dynamic stability assessment method based on data analysis according to claim 5, characterized in that, Selecting the adjustment target based on the stability grading signal includes: When the stability grading signal indicates a risk of voltage instability, the reactive current compensation parameter is adjusted first. When the stability grading signal indicates a risk of frequency instability, the active power droop coefficient is adjusted preferentially.

7. A power network dynamic stability assessment system based on data analysis, characterized in that, The system includes: The data fusion module is used to acquire real-time operating data, environmental data, and the topology of the power network to be evaluated. The real-time operating data includes spatiotemporal variation information of voltage, current, and power, and the environmental data includes meteorological conditions and geographical environmental parameters. The graph construction module is used to construct a dynamic feature graph based on the real-time operating data, environmental data and topology. The dynamic feature graph spatiotemporally correlates the electrical changes on each power grid node in the power network, the link relationships between each power grid node in the topology and the changes in environmental factors. The evaluation and analysis module is used to perform similarity matching between the dynamic feature map and a preset historical disturbance event database, and generate a stability grading signal based on the matching results. The strategy generation module is used to generate a control parameter optimization strategy based on the stability grading signal, and to adjust the control parameters of the electronic equipment of the power network according to the control parameter optimization strategy.

8. The power network dynamic stability assessment system based on data analysis according to claim 7, characterized in that, The evaluation and analysis module includes a threshold comparison unit, a pattern matching unit, and a scenario deduction unit.

9. The power network dynamic stability assessment system based on data analysis according to claim 8, characterized in that, The threshold comparison unit is used to trigger a primary alarm. The pattern matching unit is linked with the threshold comparison unit. When the threshold comparison unit triggers a primary alarm, the pattern matching unit calls a preset historical disturbance event library to identify the association pattern between environmental disturbance features and time dimension features at a certain node in the dynamic feature map, and matches the association pattern with the historical association patterns in the historical disturbance event library. If the matching result is a strong association, advanced matching is performed; if the matching result is a weak association, the primary matching result is maintained. The scenario deduction unit works in conjunction with the pattern matching unit. When the pattern matching unit determines that the matching result is a strong correlation, it finds the corresponding historical instability event in the historical disturbance event database based on the strong correlation, performs scenario deduction, and generates a stability grading signal.

Images