A method and system for security evaluation of multi-level power distribution unit fusion features

Abstract

The application discloses a kind of multi-level distribution unit fusion feature safety evaluation method and system, it is related to multi-level distribution unit fusion technical field, the application includes data fusion processing, feature extraction, safety evaluation coefficient calculation, safety state analysis and control instruction generation, the consistency and availability of multi-source heterogeneous data are effectively promoted by data fusion processing, by comprehensively extracting multidimensional features such as steady state, transient state and event, system operating state can be fully described;Further based on multiple safety evaluation coefficients and comprehensive safety evaluation coefficients, hierarchical analysis from data to function to control layer is realized, the accuracy of safety state judgment is improved;Finally, safety state identification and strategy library are combined to dynamically generate control instruction, realize the closed-loop management from state perception to decision execution, greatly improve the response speed and control precision of system in response to abnormal working condition, enhance the reliability and toughness of distribution system operation.

Description

Technical Field

[0001] This application relates to the field of multi-level power distribution unit integration technology, specifically to a security assessment method and system for the integration features of multi-level power distribution units. Background Technology

[0002] With the rapid development of smart grids, the types of equipment connected to power distribution systems are increasing daily, resulting in an exponential growth in the amount of data generated, including multi-source heterogeneous data such as operating status, load information, and fault records. Traditional power distribution system safety management typically relies on independent subsystems for local monitoring and control, lacking the ability for global, multi-level data integration and comprehensive analysis. Therefore, it is necessary to construct an integrated processing mechanism that can integrate data fusion, feature extraction, multi-dimensional safety assessment, and control decision-making to improve the intelligence level of overall power distribution system safety management.

[0003] Existing technologies, such as the invention application patent with announcement number CN120377484A, disclose a method and system for real-time monitoring and intelligent early warning of power transmission and distribution equipment based on multi-source data fusion. This relates to the field of power system safety monitoring and early warning technology. The system comprises the following steps: S1, acquisition: collecting operational data of the power transmission and distribution equipment from multiple data sources; S2, fusion: fusing the collected multi-source data and extracting key feature information; S3, monitoring: real-time monitoring of the operational status of the power transmission and distribution equipment based on the fused data; S4, early warning: generating intelligent early warning signals based on the monitoring results; and S5, storage: comprehensively analyzing and storing the monitoring and early warning results.

[0004] Regarding the above-mentioned solutions, the inventors of this application have found that the above-mentioned technologies have at least the following technical problems: 1. Currently, there is a lack of extracting three types of features—steady state, transient state, and event—from fused data, and no normalized feature set is generated, making it impossible to achieve multi-dimensional and all-round perception of security status: it cannot enhance the ability to identify different types of security risks, is not conducive to capturing instantaneous threats such as shocks and harmonics, and is not conducive to tracking operational logic and chain risks.

[0005] 2. Currently, there is a lack of hierarchical assessment from the three dimensions of data, function, and control, as well as a comprehensive level. This makes it impossible to accurately locate and assess safety risks in a hierarchical manner, reflect the correctness and coordination of the power distribution unit in performing its intended functions, reflect the matching degree between the control strategy and the actual state of the system, and reflect the effectiveness of the control commands. There is no intuitive safety status index and label generated based on multi-dimensional assessment coefficients, and the intuitive and hierarchical output of safety status cannot be achieved. Summary of the Invention

[0006] To address the aforementioned technical shortcomings, the purpose of this application is to provide a safety assessment method and system for the integrated characteristics of multi-level power distribution units.

[0007] To solve the above-mentioned technical problems, this application adopts the following technical solution: In the first aspect, this application provides a security assessment method for the fusion characteristics of multi-level power distribution units. The method includes the following steps: S1, data fusion processing: Based on the pre-acquired data of each level of power distribution units, data alignment and cleaning are performed, and fused data is generated through a weighted fusion algorithm.

[0008] S2. Feature Extraction: Based on the fused data, extract steady-state features, transient features, and event features to generate a normalized feature set.

[0009] S3. Calculation of security assessment coefficients: Based on the normalized feature set, the data fusion security assessment coefficient, the function fusion security assessment coefficient, and the control fusion security assessment coefficient are analyzed and obtained, and then the comprehensive security assessment coefficient is obtained.

[0010] S4. Security Status Analysis: Based on the data fusion security assessment coefficient, function fusion security assessment coefficient, control fusion security assessment coefficient and comprehensive security assessment coefficient, the current security status index is analyzed and a security status identifier is generated.

[0011] S5. Control command generation: Based on the security status identifier and the preset control strategy library, generate the final control command set.

[0012] Preferably, the data of each level of power distribution unit includes high-voltage power distribution data, medium-voltage power distribution data, low-voltage power distribution data, and user-side intelligent power distribution data.

[0013] Preferably, the step of performing data alignment and cleaning, and generating fused data through a weighted fusion algorithm, includes: performing data alignment and cleaning based on pre-acquired data of each level of distribution unit to obtain cleaned data of each level of distribution unit; then generating a unified data view based on the cleaned data of each level of distribution unit using a spatiotemporal mapping algorithm; and finally integrating the data based on the unified data view through a weighted fusion algorithm to generate fused data.

[0014] Preferably, the steady-state characteristics include the effective voltage value and load rate, the transient characteristics include fault characteristics, and the event characteristics include switch displacement events.

[0015] Preferably, the step of extracting steady-state features, transient features, and event features based on the fused data to generate a normalized feature set includes: calculating the effective voltage value and load rate based on the fused data using a steady-state feature extraction algorithm to generate a steady-state feature set; extracting fault features using wavelet transform based on the fused data using a transient feature extraction algorithm to generate a transient feature set; detecting switch displacement events based on the fused data using an event feature extraction algorithm to generate an event feature set; and standardizing the steady-state feature set, transient feature set, and event feature set using a feature normalization processing algorithm to generate a normalized feature set.

[0016] Preferably, the step of analyzing and deriving the data fusion security assessment coefficient, functional fusion security assessment coefficient, and control fusion security assessment coefficient based on the normalized feature set includes: A1, through calculation formulas. Determine the data fusion security assessment coefficient , This is represented by the ID corresponding to the feature in the normalized feature set. , It represents the total number of features in the normalized feature set. Represented as the first The variance of each feature, Represented as the first The weighting factors corresponding to the variance of each feature.

[0017] A2. Through calculation formula Determine the functional integration safety assessment coefficient , Represented as the first One characteristic, Represented as the first The entropy of each feature Represented as the first The first feature and the first Mutual information between features This is represented by the ID corresponding to the feature in the normalized feature set. , It represents the total number of features in the normalized feature set; and .

[0018] A3. Through calculation formula Determine the control fusion security assessment coefficient , This is represented as a control command. Represented as the first Features and control commands The correlation coefficient, This is represented as a sensitivity weighting parameter, and is preset to 0.1. Represented as control commands to the first The partial derivatives of each feature.

[0019] Preferably, the analysis yields a comprehensive safety assessment coefficient, including: through a calculation formula. derive the comprehensive safety assessment coefficient ,in , and These represent the weighting factors corresponding to the data fusion security assessment coefficient, the functional fusion security assessment coefficient, and the control fusion security assessment coefficient, respectively.

[0020] Preferably, generating a safety status identifier includes: comparing the current safety status index with the lower limit and the upper limit of the current safety status index; if the current safety status index is greater than or equal to the upper limit of the current safety status index, the current safety status is normal, and the safety status identifier is set to 0; if the current safety status index is greater than or equal to the lower limit of the current safety status index but less than the upper limit of the current safety status index, the current safety status is warning, and the safety status identifier is set to 1; if the current safety status index is less than the lower limit of the current safety status index, the current safety status is dangerous, and the safety status identifier is set to 2.

[0021] Preferably, generating the final control instruction set based on the security status identifier and the preset control strategy library includes: triggering the rule engine based on the security status identifier to generate an initial control instruction set, and then optimizing the parameters of the initial control instruction set based on the control strategy library to generate the final control instruction set.

[0022] In its second aspect, this application provides a system for a security assessment method of multi-level distribution unit fusion characteristics, comprising: a data fusion processing module, which performs data alignment and cleaning based on pre-acquired data of each level of distribution unit, and generates fused data through a weighted fusion algorithm.

[0023] The feature extraction module extracts steady-state features, transient features, and event features based on the fused data, and generates a normalized feature set.

[0024] The security assessment coefficient calculation module, based on the normalized feature set, analyzes and derives the data fusion security assessment coefficient, the function fusion security assessment coefficient, and the control fusion security assessment coefficient, and then analyzes and derives the comprehensive security assessment coefficient.

[0025] The security status analysis module analyzes and derives the current security status index based on the data fusion security assessment coefficient, function fusion security assessment coefficient, control fusion security assessment coefficient, and comprehensive security assessment coefficient, and generates a security status identifier.

[0026] The control command generation module generates the final control command set based on the safety status identifier and the preset control strategy library.

[0027] The beneficial effects of this application are as follows: 1. The safety assessment method and system for multi-level power distribution unit fusion features provided by this application effectively improves the consistency and availability of multi-source heterogeneous data through data fusion processing. By comprehensively extracting multi-dimensional features such as steady state, transient state and events, it can comprehensively characterize the system's operating status. Furthermore, based on multiple safety assessment coefficients and comprehensive safety assessment coefficients, it realizes hierarchical analysis from data to function to control layer, improving the accuracy of safety status judgment. Finally, by combining safety status identifiers and policy library to dynamically generate control commands, it realizes closed-loop management from status perception to decision execution, greatly improving the system's response speed and control accuracy in dealing with abnormal operating conditions, and enhancing the reliability and resilience of the power distribution system operation.

[0028] 2. This application aligns, cleans, and weights the data from multi-source, heterogeneous power distribution units at all levels, improving data quality and consistency. It resolves the data chaos caused by diverse data sources, inconsistent formats, and different acquisition timescales, providing a high-quality, highly consistent, standardized data foundation for subsequent analysis. The application extracts steady-state, transient, and event features in parallel from the fused data and generates a normalized feature set to achieve multi-dimensional and comprehensive perception of safety status. This enhances the ability to identify different types of safety risks: steady-state features are helpful in detecting chronic problems such as overload and imbalance; transient features are helpful in capturing instantaneous threats such as shocks and harmonics; and event features are helpful in tracking operational logic and cascading risks. The combination of these three features constructs a three-dimensional safety risk perception network.

[0029] 3. This application conducts a layered assessment from three dimensions—data, function, and control—as well as a comprehensive level, to achieve precise positioning and layered assessment of safety risks: the data fusion security assessment coefficient reflects the reliability and accuracy of the perception layer data; the function fusion security assessment coefficient reflects the correctness and coordination of the power distribution unit in performing its predetermined functions; the control fusion security assessment coefficient reflects the matching degree between the control strategy and the actual state of the system, as well as the effectiveness of the control commands; based on the multi-dimensional assessment coefficients, an intuitive safety status index and label are generated; the intuitive and hierarchical output of the safety status is realized, supporting trend prediction and early warning of the safety status.

[0030] 4. Based on the safety status identifier, this application automatically generates the final control instruction set from the preset strategy library; realizes the automation and rapid response of safety control, ensures the scientificity and optimization of the control strategy; forms a complete "perception-decision-execution" closed loop, completes the mapping from the information space to the physical world, and truly realizes the active safety protection of the power distribution network. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, 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.

[0032] Figure 1 This is a flowchart illustrating the implementation steps of the method described in this application.

[0033] Figure 2 This is a schematic diagram of the system structure connection of this application. Detailed Implementation

[0034] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0035] Please see Figure 1 As shown, this application provides a security assessment method for the fusion characteristics of multi-level power distribution units in the first aspect, including: S1, data fusion processing: based on the pre-acquired data of power distribution units at all levels, data alignment and cleaning are performed, and fused data is generated through a weighted fusion algorithm.

[0036] In a specific example, the data of each level of power distribution unit includes high-voltage power distribution data, medium-voltage power distribution data, low-voltage power distribution data, and user-side intelligent power distribution data.

[0037] In a specific example, the data alignment and cleaning, and the generation of fused data through a weighted fusion algorithm, include: performing data alignment and cleaning based on pre-acquired data of each level of distribution unit to obtain cleaned data of each level of distribution unit; then generating a unified data view based on the cleaned data of each level of distribution unit using a spatiotemporal mapping algorithm; and finally integrating the data based on the unified data view through a weighted fusion algorithm to generate fused data.

[0038] It should be noted that the process of aligning and cleaning the data based on the pre-acquired data of each level of distribution unit to obtain cleaned data of each level of distribution unit includes: synchronizing the high-voltage distribution data, medium-voltage distribution data, low-voltage distribution data, and user-side intelligent distribution data using a timestamp alignment module based on the pre-acquired data of each level of distribution unit to eliminate time zone differences and sampling frequency inconsistencies; then removing outliers, missing values, and noise data through data cleaning, and filling in missing points using interpolation to obtain cleaned data of each level of distribution unit.

[0039] Furthermore, the timestamp alignment module uses GPS global positioning time reference for hardware-level synchronization to ensure data time consistency; data cleaning is based on the Z-score algorithm to identify outliers, where the Z-score threshold is set to 3, and points exceeding the threshold are marked as outliers and removed.

[0040] It should be noted that, based on the cleaned data from each level of power distribution unit, a unified data view is generated using a spatiotemporal mapping algorithm, including: the formulas derived from the spatiotemporal mapping algorithm.

[0041] Derive a unified data view , Represented as a spacetime mapping function, This represents the data set of each level of power distribution unit after cleaning. Represented as a time dimension parameter, Represented as spatial dimension parameters, Represented as the number corresponding to the data point. , Represented as the total number of data points. Represented as the first The initial weights corresponding to each data point

[0042] Represented as the first Each cleaned data point Represented as the time decay coefficient, Represented as spatial attenuation coefficient, This is represented as the target time point in the time dimension parameters; Represented as the first The target time point corresponding to each data point; Represented as the target spatial location in the spatial dimension parameters; Represented as the first The spatial coordinates of each data point; It is represented as the Euclidean distance norm.

[0043] Furthermore, the spatiotemporal mapping algorithm maps distribution data scattered across different time and spatial points to a unified data view, emphasizing the contributions of data points in adjacent time and space, thereby generating a smooth and continuous data representation. The unified data view is a multi-dimensional matrix; a time decay coefficient is used to control the impact of temporal proximity; a spatial decay coefficient is used to control the impact of spatial proximity; both the time and spatial decay coefficients are adjusted according to the actual power grid topology. For example, for high-voltage distribution data, the time decay coefficient is set to 0.1 to reflect slower time changes; for user-side smart distribution data, the time decay coefficient is set to 0.5 to capture rapid fluctuations.

[0044] It should be noted that, based on a unified data view, data is integrated using a weighted fusion algorithm to generate fused data, including: using a weighted fusion algorithm... The fused data was obtained ,in Represented as a weighted fusion function, This represents the ID corresponding to the feature dimension in the unified data view. , This represents the number of dimensions of the features in the unified data view; Represented as the first The fusion weights corresponding to each feature Represented as the first in the unified data view 1 eigenvector; Represented as the first The normalization factor for each feature.

[0045] Furthermore, the weighted fusion algorithm integrates data by assigning different feature weights. The fusion weights for each feature are dynamically calculated based on historical data and feature importance. For example, high-voltage distribution data has a higher weight, such as 0.4, due to its strong stability; user-side smart distribution data has a lower weight, such as 0.1, due to its greater volatility. The normalization factor is calculated using the Min-Max normalization method to ensure that all features are fused uniformly.

[0046] S2. Feature Extraction: Based on the fused data, extract steady-state features, transient features, and event features to generate a normalized feature set.

[0047] In a specific instance, the steady-state characteristics include the effective voltage value and load rate, the transient characteristics include fault characteristics, and the event characteristics include switch displacement events.

[0048] In a specific example, the step of extracting steady-state features, transient features, and event features based on the fused data to generate a normalized feature set includes: calculating the effective voltage value and load rate based on the fused data using a steady-state feature extraction algorithm to generate a steady-state feature set; extracting fault features using wavelet transform based on the fused data using a transient feature extraction algorithm to generate a transient feature set; detecting switch displacement events based on the fused data using an event feature extraction algorithm to generate an event feature set; and standardizing the steady-state feature set, transient feature set, and event feature set using a feature normalization processing algorithm to generate a normalized feature set.

[0049] It should be noted that, based on the fused data, the steady-state feature extraction algorithm calculates the effective voltage value and load rate to generate a steady-state feature set. The specific process is as follows: The steady-state feature extraction algorithm includes effective voltage value calculation and load rate calculation. The effective voltage value calculation is based on AC voltage sampling to obtain several voltage sample values. The calculation process is as follows: the sum of the squares of each voltage sample value is divided by the total number of voltage sample values ​​to obtain the average voltage value, and the square root is taken to obtain the effective voltage value. The load rate calculation is based on the ratio of active power to rated power. The calculation process is as follows: the load rate equals the actual active power divided by the rated power multiplied by 100%. The steady-state feature set includes the effective voltage value and the load rate.

[0050] It should be noted that the transient feature extraction algorithm uses discrete wavelet transform to process voltage or current signals to detect transient changes caused by faults. The wavelet transform decomposes the signal into approximation coefficients and detail coefficients through multi-resolution analysis. Fault features are extracted from the detail coefficients, for example, by detecting abrupt changes through threshold processing, thereby generating a transient feature set. The transient feature set includes the detail coefficient energy and peak amplitude.

[0051] Furthermore, through the formula of wavelet transform Obtaining wavelet coefficients , Indicates the signal at and The characteristic intensity below; It is represented as a scaling parameter, which controls the scaling of the wavelet and is inversely proportional to the frequency; This is expressed as a translation parameter, controlling the position of the wavelet; Represented as complex conjugate, The input signal is a time series of voltage or current. The mother wavelet function is used; fault feature extraction is performed by analyzing the high-frequency components of the detail coefficients, such as calculating the energy or peak value of the detail coefficients, to identify fault events.

[0052] It should be noted that the event feature extraction algorithm monitors the switch state signal and uses the state change detection algorithm to identify displacement events; the formula for switch displacement event detection is as follows:

[0053] Obtain the switch displacement event flag 1 indicates that a displacement event has occurred, and 0 indicates that no event has occurred; Represented as the switch state at a given time point The value, This is represented as a threshold parameter, set to 0.5 to detect state changes; the event feature set includes switch displacement event flags, used to capture discrete events in the power grid.

[0054] It should be noted that the feature normalization processing algorithm uses the Min-Max normalization method to map features of different scales to... Ranges are used to ensure consistency of features during fusion and model processing.

[0055] This application aligns, cleans, and weights data from multi-source, heterogeneous power distribution units at all levels, improving data quality and consistency. It resolves the data chaos caused by diverse data sources, inconsistent formats, and different acquisition timescales, providing a high-quality, highly consistent, standardized data foundation for subsequent analysis. The application extracts steady-state, transient, and event-related features in parallel from the fused data, generating a normalized feature set to achieve multi-dimensional and comprehensive perception of safety status. This enhances the ability to identify different types of safety risks: steady-state features are helpful in detecting chronic problems such as overload and imbalance; transient features are helpful in capturing instantaneous threats such as shocks and harmonics; and event features are helpful in tracking operational logic and cascading risks. The combination of these three features constructs a three-dimensional safety risk perception network.

[0056] S3. Calculation of security assessment coefficients: Based on the normalized feature set, the data fusion security assessment coefficient, the function fusion security assessment coefficient, and the control fusion security assessment coefficient are analyzed and obtained, and then the comprehensive security assessment coefficient is obtained.

[0057] In a specific example, the process of analyzing and deriving the data fusion security assessment coefficient, functional fusion security assessment coefficient, and control fusion security assessment coefficient based on the normalized feature set includes: A1, through calculation formulas... Determine the data fusion security assessment coefficient , This is represented by the ID corresponding to the feature in the normalized feature set. , It represents the total number of features in the normalized feature set. Represented as the first The variance of each feature, Represented as the first The weighting factors corresponding to the variance of each feature.

[0058] It should be noted that, , We obtain the weight factors corresponding to the variance of each feature through factor analysis. First, we condense the information of the variance of each feature, then obtain the variance explained rate after rotation, and finally obtain the weights by summing the variance explained rates.

[0059] It should be noted that factor analysis is a well-known technique. It is a multivariate statistical analysis method that starts by studying the internal dependencies of variables and reduces some variables with complex relationships to a few comprehensive factors. Information condensation is expressed as the calculation of the median. The variance explained rate is the amount of information extracted by the factors. Variance explained rate = eigenvalues ​​ / total number of analysis terms. The rotated variance explained rate is expressed as the variance explained by the factors after maximum variance rotation.

[0060] A2. Through calculation formula Determine the functional integration safety assessment coefficient , Represented as the first One characteristic, Represented as the first The entropy of each feature Represented as the first The first feature and the first Mutual information between features This is represented by the ID corresponding to the feature in the normalized feature set. , It represents the total number of features in the normalized feature set; and .

[0061] It should be noted that, through the calculation formula The result is the The entropy of each feature, where The probability distribution of the features is represented by the calculation formula. The probability distribution estimation specifically employs the histogram method to extract probability values ​​from the normalized feature set.

[0062] Furthermore, the functional fusion security assessment coefficient represents the information integrity and independence during the functional fusion process, with a higher value indicating higher security; the entropy of a feature reflects the uncertainty of that feature; and the mutual information between two features represents the degree of dependence between the features.

[0063] A3. Through calculation formula Determine the control fusion security assessment coefficient , This is represented as a control command. Represented as the first Features and control commands The correlation coefficient, This is represented as a sensitivity weighting parameter, and is preset to 0.1. Represented as control commands to the first The partial derivatives of each feature.

[0064] It should be noted that, Correlation coefficient equals Covariance divided by and The standard deviation.

[0065] Furthermore, the control fusion security assessment coefficient represents the responsiveness and stability of the control fusion process, with higher values ​​indicating higher security; the correlation coefficient between the feature and the control command reflects the linear relationship between the feature and the control command; the partial derivative of the control command with respect to the i-th feature represents the degree of influence of feature changes on the control command, calculated through automatic differentiation.

[0066] In a specific instance, the analysis yields a comprehensive security assessment coefficient, including: through a calculation formula. derive the comprehensive safety assessment coefficient ,in , and These represent the weighting factors corresponding to the data fusion security assessment coefficient, the functional fusion security assessment coefficient, and the control fusion security assessment coefficient, respectively.

[0067] It should be noted that, , , , The data fusion security assessment coefficient, functional fusion security assessment coefficient, and control fusion security assessment coefficient were obtained through factor analysis.

[0068] S4. Security Status Analysis: Based on the data fusion security assessment coefficient, function fusion security assessment coefficient, control fusion security assessment coefficient and comprehensive security assessment coefficient, the current security status index is analyzed and a security status identifier is generated.

[0069] It should be noted that, through the calculation formula Calculate the current security status index , This is represented by the weighting factor corresponding to the comprehensive safety assessment coefficient. These are weighting factors corresponding to the data fusion security assessment coefficient, the functional fusion security assessment coefficient, and the control fusion security assessment coefficient.

[0070] Furthermore, , , ; and obtained through factor analysis and .

[0071] In a specific example, generating a safety status identifier includes: comparing the current safety status index with the lower limit and the upper limit of the current safety status index; if the current safety status index is greater than or equal to the upper limit of the current safety status index, the current safety status is normal, and the safety status identifier is set to 0; if the current safety status index is greater than or equal to the lower limit of the current safety status index but less than the upper limit of the current safety status index, the current safety status is warning, and the safety status identifier is set to 1; if the current safety status index is less than the lower limit of the current safety status index, the current safety status is dangerous, and the safety status identifier is set to 2.

[0072] It should be noted that the upper limit of the current security status index indicates a high level of security; the lower limit of the current security status index indicates a low level of security; the generation of the security status identifier is achieved by querying a predefined mapping table, which is stored in memory to improve response speed.

[0073] This application employs a tiered assessment approach across three dimensions—data, function, and control—as well as a comprehensive level, to achieve precise location and tiered evaluation of safety risks. The data fusion safety assessment coefficient reflects the reliability and accuracy of the perception layer data; the function fusion safety assessment coefficient reflects the correctness and coordination of the power distribution unit in performing its intended functions; and the control fusion safety assessment coefficient reflects the matching degree between the control strategy and the actual system state, as well as the effectiveness of control commands. Based on these multi-dimensional assessment coefficients, an intuitive safety status index and identifier are generated, enabling the intuitive and hierarchical output of safety status and supporting trend prediction and early warning of safety status.

[0074] S5. Control command generation: Based on the security status identifier and the preset control strategy library, generate the final control command set.

[0075] In a specific example, generating the final control instruction set based on the security status identifier and the preset control strategy library includes: triggering the rule engine based on the security status identifier to generate an initial control instruction set, and then optimizing the parameters of the initial control instruction set based on the control strategy library to generate the final control instruction set.

[0076] It should be noted that the rule engine is triggered based on the security status identifier to generate an initial control instruction set; the rule engine uses a predefined rule library to map the security status identifier to the corresponding control action; the security status identifier includes normal, warning and danger, which correspond to different control instruction set generation strategies.

[0077] Furthermore, the rules engine uses hardware logic circuits to quickly match status flags with instructions, ensuring real-time system response.

[0078] It should be noted that the parameters of the initial control instruction set are optimized based on the control strategy library to generate the final control instruction set. The parameter optimization algorithm employs an iterative approach, using system performance indicators as the objective function. It minimizes the objective function by adjusting the instruction parameter values. These performance indicators include energy consumption and stability metrics. The optimization process is based on historical operating data and uses automatic differentiation technology to calculate gradients, ensuring optimization efficiency and accuracy. The instruction synthesis algorithm combines the optimized parameter values ​​with instruction templates to generate control instructions that conform to communication protocol specifications, serving as the final control instruction set. The instruction templates are pre-stored in the system and support multiple industrial communication protocol formats. The synthesis process uses string formatting to ensure that the generated instructions can be correctly parsed and executed by the target device.

[0079] Based on safety status identification, this application automatically generates the final control instruction set from a preset strategy library; it realizes the automation and rapid response of safety control, ensuring the scientific and optimal nature of the control strategy; it forms a complete "perception-decision-execution" closed loop, completing the mapping from the information space to the physical world, and truly realizing the active safety protection of the power distribution network.

[0080] Please see Figure 2 As shown, in a second aspect, this application provides a system for a security assessment method of multi-level power distribution unit fusion features.

[0081] The system 100 of the safety assessment method for the fusion characteristics of multi-level power distribution units described in this invention can be installed in an electronic device. Depending on the functions implemented, the system 100 may include a data fusion processing module 101, a feature extraction module 102, a safety assessment coefficient calculation module 103, a safety status analysis module 104, and a control command generation module 105. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, stored in the memory of the electronic device.

[0082] In this embodiment, the functions of each module / unit are as follows: The data fusion processing module performs data alignment and cleaning based on the pre-acquired data of each level of power distribution unit, and generates fused data through a weighted fusion algorithm.

[0083] The feature extraction module extracts steady-state features, transient features, and event features based on the fused data, and generates a normalized feature set.

[0084] The security assessment coefficient calculation module, based on the normalized feature set, analyzes and derives the data fusion security assessment coefficient, the function fusion security assessment coefficient, and the control fusion security assessment coefficient, and then analyzes and derives the comprehensive security assessment coefficient.

[0085] The security status analysis module analyzes and derives the current security status index based on the data fusion security assessment coefficient, function fusion security assessment coefficient, control fusion security assessment coefficient, and comprehensive security assessment coefficient, and generates a security status identifier.

[0086] The control command generation module generates the final control command set based on the safety status identifier and the preset control strategy library.

[0087] This application provides a safety assessment method and system for multi-level power distribution unit fusion features. Through data fusion processing, it effectively improves the consistency and availability of multi-source heterogeneous data. By comprehensively extracting multi-dimensional features such as steady-state, transient, and event-related characteristics, it can comprehensively characterize the system's operating status. Furthermore, based on multiple safety assessment coefficients and a comprehensive safety assessment coefficient, it achieves hierarchical analysis from data to function to control layer, improving the accuracy of safety status judgment. Finally, by combining safety status identifiers and a strategy library to dynamically generate control commands, it achieves closed-loop management from status perception to decision execution, significantly improving the system's response speed and control accuracy in dealing with abnormal operating conditions, and enhancing the reliability and resilience of the power distribution system.

[0088] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0089] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0090] Furthermore, the functional modules in the various embodiments of the present invention 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 in the form of hardware plus software functional modules.

[0091] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0092] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0093] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A safety assessment method for the integrated features of multi-level power distribution units, characterized in that, include: S1. Data Fusion Processing: Based on the pre-acquired data from each level of power distribution unit, data alignment and cleaning are performed, and fused data is generated through a weighted fusion algorithm; S2. Feature Extraction: Based on the fused data, extract steady-state features, transient features, and event features to generate a normalized feature set; The steady-state characteristics include the effective voltage value and load rate, the transient characteristics include fault characteristics, and the event characteristics include switch displacement events. Based on the fused data, steady-state features, transient features, and event features are extracted to generate a normalized feature set, including: Based on the fused data, the effective voltage value and load rate are calculated using a steady-state feature extraction algorithm to generate a steady-state feature set; based on the fused data, fault features are extracted using a transient feature extraction algorithm with wavelet transform to generate a transient feature set; based on the fused data, switch displacement events are detected using an event feature extraction algorithm to generate an event feature set; based on the steady-state feature set, transient feature set, and event feature set, a feature normalization processing algorithm is used to standardize the data to generate a normalized feature set. S3. Security assessment coefficient calculation: Based on the normalized feature set, the data fusion security assessment coefficient, the function fusion security assessment coefficient, and the control fusion security assessment coefficient are analyzed and obtained, and then the comprehensive security assessment coefficient is obtained. The analysis based on the normalized feature set yields data fusion security assessment coefficients, functional fusion security assessment coefficients, and control fusion security assessment coefficients, including: A1. Through calculation formula Determine the data fusion security assessment coefficient , This is represented by the ID corresponding to the feature in the normalized feature set. , It represents the total number of features in the normalized feature set. Represented as the first The variance of each feature, Represented as the first The weighting factors corresponding to the variance of each feature; A2. Through calculation formula Determine the functional integration safety assessment coefficient , Represented as the first One characteristic, Represented as the first The entropy of each feature Represented as the first The first feature and the first Mutual information between features This is represented by the ID corresponding to the feature in the normalized feature set. , It represents the total number of features in the normalized feature set; and ; A3. Through calculation formula Determine the control fusion security assessment coefficient , This is represented as a control command. Represented as the first Features and control commands The correlation coefficient, This is represented as a sensitivity weighting parameter, and is preset to 0.

1. Represented as control commands to the first Partial derivatives of each feature; S4. Security Status Analysis: Based on the data fusion security assessment coefficient, function fusion security assessment coefficient, control fusion security assessment coefficient, and comprehensive security assessment coefficient, the current security status index is analyzed and a security status identifier is generated. S5. Control command generation: Based on the security status identifier and the preset control strategy library, generate the final control command set.

2. The safety assessment method for the integrated characteristics of multi-level power distribution units according to claim 1, characterized in that, The data from each level of power distribution unit includes high-voltage power distribution data, medium-voltage power distribution data, low-voltage power distribution data, and user-side intelligent power distribution data.

3. The safety assessment method for the integrated characteristics of multi-level power distribution units according to claim 1, characterized in that, The process of data alignment and cleaning, and generating fused data through a weighted fusion algorithm, includes: Based on the pre-acquired data of each level of distribution unit, data alignment and cleaning are performed to obtain cleaned data of each level of distribution unit; then, based on the cleaned data of each level of distribution unit, a spatiotemporal mapping algorithm is used to generate a unified data view; and finally, based on the unified data view, the data is integrated through a weighted fusion algorithm to generate fused data.

4. The safety assessment method for the integrated features of multi-level power distribution units according to claim 1, characterized in that, The analysis yields a comprehensive security assessment coefficient, including: Through calculation formula derive the comprehensive safety assessment coefficient ,in , and These represent the weighting factors corresponding to the data fusion security assessment coefficient, the functional fusion security assessment coefficient, and the control fusion security assessment coefficient, respectively.

5. The safety assessment method for the integrated characteristics of multi-level power distribution units according to claim 1, characterized in that, The generation of the security status identifier includes: The current safety status index is compared with its lower limit and upper limit. If the current safety status index is greater than or equal to its upper limit, the current safety status is normal, and the safety status flag is set to 0. If the current safety status index is greater than or equal to its lower limit but less than its upper limit, the current safety status is warning, and the safety status flag is set to 1. If the current safety status index is less than its lower limit, the current safety status is dangerous, and the safety status flag is set to 2.

6. The safety assessment method for the integrated features of multi-level power distribution units according to claim 1, characterized in that, The final control instruction set is generated based on the security status identifier and the preset control policy library, including: The rule engine is triggered based on the security status identifier to generate an initial control instruction set, and then the parameters of the initial control instruction set are optimized based on the control policy library to generate a final control instruction set.

7. A system for implementing a security assessment method for the fusion features of a multi-level power distribution unit as described in any one of claims 1-6, characterized in that, include: The data fusion processing module performs data alignment and cleaning based on pre-acquired data from various levels of power distribution units, and generates fused data through a weighted fusion algorithm; The feature extraction module extracts steady-state features, transient features, and event features based on the fused data, and generates a normalized feature set. The security assessment coefficient calculation module, based on the normalized feature set, analyzes and derives the data fusion security assessment coefficient, the function fusion security assessment coefficient, and the control fusion security assessment coefficient, and then analyzes and derives the comprehensive security assessment coefficient. The security status analysis module, based on the data fusion security assessment coefficient, function fusion security assessment coefficient, control fusion security assessment coefficient, and comprehensive security assessment coefficient, analyzes and derives the current security status index and generates a security status identifier. The control command generation module generates the final control command set based on the safety status identifier and the preset control strategy library.

Images