Method and device for identifying abnormal data of transformer substation, electronic equipment and storage medium

Abstract

The invention provides a transformer substation abnormal data identification method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a power grid topological structure corresponding to a transformer substation and measurement data of a plurality of measurement nodes; determining a target operation scene of a power grid where the transformer substation is located according to the measurement data; determining a target data processing rule of the measurement data according to the target operation scene based on a preset corresponding relationship between the operation scene and the data processing rule; processing the measurement data by applying a target data processing rule to obtain target measurement data required for abnormal data identification in the target operation scene; aiming at each kind of electrical quantity data in various kinds of electrical quantity data contained in the target measurement data, based on a power grid topological structure, determining a dynamic reference standard corresponding to the electrical quantity data; and determining abnormal data in the target measurement data according to the dynamic reference datum. The method is used for accurately identifying abnormal data in the transformer substation and improving the operation and maintenance efficiency of the transformer substation.

Description

Technical Field

[0001] This application relates to the field of power system technology, and in particular to a method, device, electronic equipment and storage medium for identifying abnormal data in substations. Background Technology

[0002] With the advancement of new power system construction, the amount of measurement data generated by multiple sources of devices in substations, such as protection devices, monitoring and control devices, and phasor measurement units (PMUs), is increasing explosively. Simultaneously, new power systems place higher demands on data accuracy. In the operation of these new power systems, any abnormal data can trigger erroneous adjustments and controls, thereby affecting grid safety and enterprise economic benefits. Therefore, how to quickly and accurately identify abnormal data in substations has become a focus of industry attention.

[0003] In related technologies, the identification of abnormal data in substations mainly relies on setting fixed thresholds to judge whether real-time measurement data reported by multi-source devices exceeds the limit. For example, if the real-time measurement data is greater than or equal to the set fixed threshold, the real-time measurement data is identified as abnormal data. Alternatively, a reference benchmark is established based on the mean or median of historical measurement data. If the real-time measurement data is greater than or equal to the reference benchmark, the real-time measurement data is identified as abnormal data. However, there is a problem of low accuracy in identifying abnormal data, which makes it difficult to meet the operation and maintenance needs of substations. Summary of the Invention

[0004] This application provides a method, device, electronic equipment, and storage medium for identifying abnormal data in substations, so as to accurately identify abnormal data in substations and improve the efficiency of substation operation and maintenance.

[0005] Firstly, this application provides a method for identifying abnormal data in a substation. The substation has multiple measurement nodes, and the method includes:

[0006] Acquire the power grid topology corresponding to the substation and the measurement data of multiple measurement nodes. The measurement data includes various electrical quantity data.

[0007] Based on the measurement data, determine the target operating scenario of the power grid where the substation is located;

[0008] Based on the pre-defined correspondence between operating scenarios and data processing rules, the target data processing rules for measurement data are determined according to the target operating scenario.

[0009] The measurement data is processed using target data processing rules to obtain the target measurement data required for anomaly data identification in the target operating scenario.

[0010] For each type of electrical quantity data included in the target measurement data, a dynamic reference benchmark corresponding to the electrical quantity data is determined based on the power grid topology.

[0011] Based on the dynamic reference baseline, identify abnormal data in the target measurement data.

[0012] In one possible implementation, for each type of electrical quantity data included in the target measurement data, a dynamic reference benchmark corresponding to the electrical quantity data is determined based on the power grid topology, including:

[0013] Based on the power grid topology, determine the electrical distance between any two measurement nodes among multiple measurement nodes;

[0014] Based on the electrical distance, the normalized topology weight of each of the multiple measurement nodes is determined. The normalized topology weight reflects the degree of influence of the measurement node on the electrical quantity data of other measurement nodes.

[0015] For each type of electrical quantity data, a weighted M-estimation method is used to determine the dynamic reference benchmark corresponding to the electrical quantity data based on the normalized topology weights.

[0016] In one possible implementation, the dynamic reference datum satisfies the following formula:

[0017]

[0018] In the formula, For the first The dynamic reference benchmark corresponding to the electrical quantity data =1,2,… , A pre-defined constant; These are candidate reference values ​​during the optimization process; For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Various electrical quantity data; The number of measurement nodes in the substation; For the first Normalized topological weights of each measurement node; This is the loss function.

[0019] In one possible implementation, the target operating scenario of the power grid where the substation is located is determined based on measurement data, including:

[0020] Determine the comprehensive disturbance index of the measurement data. The comprehensive disturbance index reflects the degree of comprehensive disturbance of the electrical quantities of the substation at a specific moment.

[0021] If the comprehensive disturbance index is less than or equal to the first set threshold, the target operating scenario is determined to be in steady state.

[0022] If the comprehensive disturbance index is greater than the first set threshold, and the comprehensive disturbance index is less than or equal to the second set threshold, then the target operating scenario is determined to be transient.

[0023] If the overall disturbance index is greater than the second set threshold, the target operating scenario is determined to be in the topology switching state.

[0024] In one possible implementation, the combined disturbance index satisfies the following formula:

[0025]

[0026] In the formula, For measurement data in The overall disturbance index at any given time; This is the instantaneous rate of change weighting coefficient, used to adjust the proportion of instantaneous changes in the state vector in the comprehensive disturbance index; This is the recent volatility weighting coefficient, used to adjust the proportion of recent volatility in the state vector in the comprehensive disturbance index; for A multidimensional state vector composed of measurement data at any given time; In the time window The statistical standard deviation of the internal multidimensional state vector; In the time window The statistical mean of the internal multidimensional state vector; The length of the time window; It is a very small positive number.

[0027] In one possible implementation, identifying anomalous data in the target measurement data based on a dynamic reference baseline includes:

[0028] For each type of electrical quantity data corresponding to each of the multiple measurement nodes, determine the normalized anomaly score of the electrical quantity data relative to the dynamic reference benchmark corresponding to the electrical quantity data;

[0029] If the normalized anomaly score is greater than or equal to the set anomaly judgment threshold, the measurement node is identified as an abnormal measurement node, and the electrical quantity data is identified as abnormal data in the target measurement data.

[0030] In one possible implementation, acquiring measurement data from multiple measurement nodes includes:

[0031] The system acquires raw measurement data from multiple measurement nodes and pre-built engineering ledger information corresponding to the substation. The engineering ledger information stores the mapping relationship between measurement nodes and current transformer ratio, voltage transformer ratio, measurement device range, transformer vector group, current transformer installation direction, and measurement data type.

[0032] For each of the multiple measurement nodes, the original measurement data of the measurement node is normalized according to the current transformer ratio, voltage transformer ratio, measurement device range and measurement data type corresponding to the measurement node, so as to obtain normalized measurement data.

[0033] Based on the transformer vector group and current transformer installation direction corresponding to the measurement node, the normalized measurement data is phase aligned to obtain the measurement data corresponding to the measurement node.

[0034] In one possible implementation, the measurement data carries mapping metadata, which is used to mark the source of each data point in the measurement data.

[0035] Correspondingly, substation abnormal data identification methods also include:

[0036] By integrating the abnormal data in the target measurement data and the corresponding mapping metadata, an abnormal data evidence package for the substation is obtained.

[0037] Secondly, this application provides a substation abnormal data identification device, wherein multiple measurement nodes are installed in the substation, and the substation abnormal data identification device includes:

[0038] The acquisition module is used to acquire the power grid topology corresponding to the substation and the measurement data of multiple measurement nodes. The measurement data includes various electrical quantity data.

[0039] The first determining module is used to determine the target operating scenario of the power grid where the substation is located based on the measurement data.

[0040] The second determining module is used to determine the target data processing rules for the measurement data based on the correspondence between the preset operating scenarios and data processing rules, according to the target operating scenario.

[0041] The processing module is used to process the measurement data by applying target data processing rules to obtain the target measurement data required for anomaly data identification in the target operating scenario.

[0042] The third determination module is used to determine the dynamic reference benchmark corresponding to each type of electrical quantity data based on the power grid topology for each type of electrical quantity data.

[0043] The fourth determination module is used to determine abnormal data in the target measurement data based on the dynamic reference benchmark.

[0044] In one possible implementation, the third determining module is specifically used to: determine the electrical distance between any two measuring nodes among multiple measuring nodes based on the power grid topology; determine the normalized topology weight of each measuring node among multiple measuring nodes according to the electrical distance, wherein the normalized topology weight reflects the degree of influence of the measuring node on the electrical quantity data of other measuring nodes; and for each type of electrical quantity data among multiple electrical quantity data, use the weighted M estimation method to determine the dynamic reference benchmark corresponding to the electrical quantity data according to the normalized topology weight.

[0045] In one possible implementation, the dynamic reference datum satisfies the following formula:

[0046]

[0047] In the formula, For the first The dynamic reference benchmark corresponding to the electrical quantity data =1,2,… , A pre-defined constant; These are candidate reference values ​​during the optimization process; For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Various electrical quantity data; The number of measurement nodes in the substation; For the first Normalized topological weights of each measurement node; This is the loss function.

[0048] In one possible implementation, the first determining module is specifically used to: determine the comprehensive disturbance index of the measurement data, the comprehensive disturbance index reflecting the comprehensive disturbance degree of the electrical quantity of the substation at a specific moment; if the comprehensive disturbance index is less than or equal to a first set threshold, then the target operating scenario is determined to be steady state; if the comprehensive disturbance index is greater than the first set threshold, and the comprehensive disturbance index is less than or equal to a second set threshold, then the target operating scenario is determined to be transient state; if the comprehensive disturbance index is greater than the second set threshold, then the target operating scenario is determined to be topology switching state.

[0049] In one possible implementation, the combined disturbance index satisfies the following formula:

[0050]

[0051] In the formula, For measurement data in The overall disturbance index at any given time; This is the instantaneous rate of change weighting coefficient, used to adjust the proportion of instantaneous changes in the state vector in the comprehensive disturbance index; This is the recent volatility weighting coefficient, used to adjust the proportion of recent volatility in the state vector in the comprehensive disturbance index; for A multidimensional state vector composed of measurement data at any given time; In the time window The statistical standard deviation of the internal multidimensional state vector; In the time window The statistical mean of the internal multidimensional state vector; The length of the time window; It is a very small positive number.

[0052] In one possible implementation, the fourth determining module is specifically used to: determine the normalized anomaly score of the electrical quantity data relative to the dynamic reference benchmark corresponding to each type of electrical quantity data for each of the multiple measurement nodes; if the normalized anomaly score is greater than or equal to a set anomaly judgment threshold, then the measurement node is determined as an abnormal measurement node, and the electrical quantity data is determined as abnormal data in the target measurement data.

[0053] In one possible implementation, the acquisition module is specifically used to: acquire raw measurement data from multiple measurement nodes and pre-built engineering ledger information corresponding to the substation. The engineering ledger information stores the mapping relationship between measurement nodes and current transformer ratios, voltage transformer ratios, measurement device ranges, transformer vector groups, current transformer installation directions, and measurement data types. For each measurement node among the multiple measurement nodes, the module performs amplitude normalization processing on the raw measurement data of the measurement node according to the current transformer ratio, voltage transformer ratio, measurement device range, and measurement data type corresponding to the measurement node, to obtain normalized measurement data. The module then performs phase alignment processing on the normalized measurement data according to the transformer vector group and current transformer installation direction corresponding to the measurement node, to obtain the measurement data corresponding to the measurement node.

[0054] In one possible implementation, the measurement data carries mapping metadata, which is used to mark the source of each data in the measurement data; correspondingly, the processing module is also used to: fuse the abnormal data in the target measurement data and the mapping metadata corresponding to the abnormal data to obtain the abnormal data evidence package of the substation.

[0055] Thirdly, this application provides an electronic device, including: a memory and a processor;

[0056] The memory stores instructions that the computer executes;

[0057] The processor executes computer execution instructions stored in memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0058] Fourthly, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible embodiments of the first aspect.

[0059] Fifthly, this application provides a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0060] The substation abnormal data identification method, device, electronic equipment, and storage medium provided in this application acquire measurement data from the power grid topology and multiple measurement nodes corresponding to the substation. The measurement data includes various electrical quantities. Based on the measurement data, the target operating scenario of the power grid where the substation is located is determined. Based on a pre-defined correspondence between the operating scenario and data processing rules, target data processing rules for the measurement data are determined according to the target operating scenario. The target data processing rules are applied to process the measurement data to obtain the target measurement data required for abnormal data identification under the target operating scenario. For each electrical quantity included in the target measurement data, a dynamic reference benchmark is determined based on the power grid topology. Abnormal data in the target measurement data is identified based on the dynamic reference benchmark. In this process, considering that the distribution characteristics of measurement data may differ under different power grid operating scenarios, data processing rules matching the operating scenario are used to process the measurement data, specifically generating target measurement data that meets the requirements for abnormal data identification under the operating scenario, improving data quality, and facilitating the determination of the reference benchmark and the identification of abnormal data. Furthermore, when determining the dynamic reference benchmark, the impact of local faults or topology changes on the distribution of measurement data is fully considered. By combining the real-time power grid topology and target measurement data to determine the reference benchmark, timely responses to power grid changes are made, the accuracy and reliability of the dynamic reference benchmark are improved, thereby improving the accuracy of abnormal data identification, avoiding false or missed reports during substation operation and maintenance, improving substation operation and maintenance efficiency, and ensuring the safe and stable operation of substations. Attached Figure Description

[0061] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0062] Figure 1 A flowchart illustrating the substation abnormal data identification method provided in this application embodiment;

[0063] Figure 2 This is a schematic diagram of the substation abnormal data identification device provided in the embodiments of this application;

[0064] Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0065] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0066] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0067] The terms “first,” “second,” etc., used in the specification and claims of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, products, or apparatus.

[0068] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with relevant laws, regulations and standards, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0069] In related technologies, setting fixed thresholds based on human experience fails to consider the dynamic changes in the power grid during actual operation. These dynamic changes lead to significant variations in the distribution characteristics of measurement data. For example, measurement data fluctuates greatly during load surges or topology switching, and fixed thresholds cannot adapt to dynamic scenarios. Establishing reference benchmarks based on the mean or median of historical measurement data is susceptible to outliers, resulting in inaccuracies. Using fixed thresholds to filter out abnormal data, or using reference benchmarks built from historical data, both methods struggle to accurately identify anomalies, leading to false positives or false negatives.

[0070] To address the aforementioned technical issues, the substation abnormal data identification method provided in this application identifies power grid operation scenarios in real time, dynamically matches the data processing rules corresponding to the operation scenarios, and specifically generates target measurement data that meets the abnormal data identification requirements under the operation scenarios. By combining the real-time power grid topology and target measurement data, a reference benchmark is dynamically determined, improving the accuracy and reliability of the reference benchmark, thereby enhancing the accuracy of abnormal data identification.

[0071] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0072] Figure 1 A flowchart illustrating the substation abnormal data identification method provided in this application embodiment. Figure 1 ,like Figure 1 As shown, the method includes:

[0073] S101. Obtain the power grid topology and measurement data of multiple measurement nodes corresponding to the substation. The measurement data includes various electrical quantity data.

[0074] The power grid topology is an abstract model describing the connection relationships and layout of various electrical components (such as generators, transformers, lines, and loads) within the power grid where a substation is located. It reflects the physical architecture and electrical connection characteristics of the power grid. It should be understood that the power grid topology also specifically includes the connection methods and layout relationships of electrical equipment and lines within the substation, reflecting the physical architecture and electrical connection relationships of the substation. This typically includes the connection forms of key components such as busbars, transformers, circuit breakers, disconnect switches, and transmission lines. Multiple measurement nodes are usually set up within the substation to collect the operating status of different locations and / or electrical equipment. These measurement nodes include sensors, instrument transformers, protection devices, PMUs, etc., located at key locations within the substation. The measured data is not limited to electrical quantities such as voltage, current, active power, reactive power, and harmonic content.

[0075] For example, the real-time grid topology of the power grid where the substation is located can be obtained from the power grid dispatching platform, the power grid geographic information system, and the power grid online monitoring system. The timing of obtaining the grid topology can be based on topology change events, such as subscribing to topology change events like switch status changes, and promptly obtaining the latest grid topology when a topology change is detected. Measurement data can be obtained directly from the measurement devices at each measurement node; or through a communication interface from the substation automation system, which integrates the monitoring and control of various equipment within the substation, including data acquisition from measurement devices; or from a cloud platform, which can integrate and store data from different systems and measurement nodes.

[0076] S102. Based on the measurement data, determine the target operating scenario of the power grid where the substation is located.

[0077] It should be understood that the distribution characteristics of measurement data at various measurement nodes within a substation differ under different operating scenarios. For example, when a short-circuit fault or switching operation occurs in the power grid, electrical quantities will change drastically.

[0078] For example, in one implementation, a machine learning-based classification algorithm (such as a pre-trained runtime scene recognition model) identifies features in the measurement data and outputs the target runtime scene corresponding to the measurement data. The runtime scene recognition model is trained using a large amount of historical measurement data and its corresponding runtime scenes, enabling the model to accurately learn the mapping relationship between data and scenes, thereby accurately outputting the runtime scene corresponding to the measurement data.

[0079] Another implementation method is a rule-based classification method. For example, based on the power grid's operating standards and specifications, combined with actual operating experience, a series of classification rules are formulated. Based on the characteristics of the measurement data, the data is matched against the pre-defined classification rules, and the operating scenario of the power grid is determined based on the matching results.

[0080] S103. Based on the pre-defined correspondence between operating scenarios and data processing rules, determine the target data processing rules for the measurement data according to the target operating scenario.

[0081] The operating scenarios of a power grid include transient, steady-state, and topology switching states. A transient state is a brief, unstable state experienced by the power grid during the transition from one stable state to another, where electrical quantities change drastically; for example, voltage may fluctuate significantly, even momentarily dropping to zero or rising far above its rated value. A steady state is a state in which, under normal operating conditions, electrical quantities (such as voltage, current, frequency, and power) remain relatively stable and do not change significantly over time; the fluctuation range of electrical quantities is small, typically within an allowable deviation range near the rated value. A topology switching state is the process of changes in the power grid's topology, such as structural changes like switching operations or equipment commissioning / decommissioning. Topology switching is usually caused by power grid maintenance, fault handling, or adjustments to operating modes.

[0082] The correspondence between the operating scenario and the data processing rules is predefined, for example, based on expert experience, theoretical knowledge, historical data, etc. The data processing rules are used to process the measurement data into data that better matches the target operating scenario for anomaly data identification.

[0083] In one specific implementation, a search is performed in the corresponding relationship based on the target operating scenario, and the data processing rule corresponding to the target operating scenario is used as the target data processing rule for the measurement data.

[0084] S104. Apply the target data processing rules to process the measurement data to obtain the target measurement data required for anomaly data identification in the target operating scenario.

[0085] For example, if the target operating scenario is steady state, it indicates that the power grid is operating smoothly, the electrical quantities change slowly and the fluctuation range is small, and the instantaneous values ​​of the measurement data are highly comparable. In this case, the corresponding target data processing rule is not to process the measurement data, that is, to directly use the measurement data obtained in step S101 as the target measurement data.

[0086] If the target operating scenario is transient, it indicates that the power grid has dynamic processes such as load changes and fault transients. Electrical quantities change rapidly over time, but the network topology remains unchanged. The waveform trend of the measurement data is more meaningful than the instantaneous value. The corresponding target data processing rule is to use high-pass filtering to remove low-frequency noise components from the measurement data at the current moment and the measurement data within the short-term historical time window, retain the high-frequency dynamic change information, and then perform trend extraction on the filtered data (e.g., using moving average or least squares fitting) to form target measurement data that characterizes dynamic features.

[0087] If the target operating scenario is in a topology switching state, the current abnormal data identification process will be skipped, and the measurement data after the topology switching will be reacquired after completion. In other words, if the target measurement data is in steady-state or transient condition, the system will not perform abnormal data identification during topology switching. Typically, the power grid will enter a steady state after a topology switching.

[0088] S105. For each type of electrical quantity data contained in the target measurement data, determine the corresponding dynamic reference benchmark based on the power grid topology.

[0089] It should be understood that for each electrical quantity data included in the measurement data, such as voltage, current, and active power, there is a corresponding dynamic reference benchmark (voltage reference benchmark, current reference benchmark, and active power reference benchmark). The dynamic reference benchmark can reflect the expected values ​​of electrical quantities of the power grid under normal operation or specific operating modes, providing a comparison standard for actual measurement data, thereby enabling timely detection of abnormal changes in the power grid.

[0090] The changes in electrical quantity data in substations are constrained by the power grid topology. One approach is to combine the power grid topology with topology-weighted aggregation of the target measurement data to construct a dynamic reference benchmark. Specifically, this can be achieved using the following steps 1.1 to 1.3:

[0091] Step 1.1: Based on the power grid topology, determine the electrical distance between any two measurement nodes among multiple measurement nodes.

[0092] Electrical distance is an important indicator for measuring the degree of electrical coupling between two measurement nodes in a power grid. It reflects the strength of the mutual influence of electrical quantities such as voltage and power between measurement nodes.

[0093] For example, the electrical distance between any two measurement nodes can be calculated using the impedance matrix method, admittance matrix method, or graph theory method. Taking the graph theory method as an example, the power grid topology is converted into a graph model. Based on a defined graph theory distance (such as shortest path length or resistance distance), the electrical distance between any two measurement nodes can be calculated (e.g., if the graph theory distance is defined as resistance distance, the electrical distance can be obtained by constructing the Laplace matrix of the graph and calculating its pseudo-inverse). Taking the admittance matrix method as an example, the admittance matrix reflects the connection relationship and electrical parameters of the measurement nodes. The mutual admittance magnitude between nodes can approximate the electrical distance; the larger the magnitude, the tighter the electrical connection and the closer the distance.

[0094] It should be noted that the embodiments of this application do not impose specific limitations on the calculation method of the electrical distance between two measurement nodes, as long as the calculation of the electrical distance can be achieved.

[0095] Step 1.2: Based on the electrical distance, determine the normalized topology weight of each of the multiple measurement nodes. The normalized topology weight reflects the degree of influence of the measurement node on the electrical quantity data of other measurement nodes.

[0096] For example, the closer the electrical distance, the stronger the correlation between the two measurement nodes. The formula for calculating the normalized topology weight is as follows:

[0097]

[0098] In the formula, For the first Normalized topological weights of each measurement node; For the first The electrical distance from each measurement node to itself; N is the number of measurement nodes in the substation; For the first Electrical distance between one measurement node and other measurement nodes.

[0099] It should be understood that the electrical distance from a measurement node to itself is theoretically 0. However, in engineering calculations, the electrical distance from a measurement node to itself is defined as a uniform, extremely small positive number. The purpose is to generate a huge self-weight through mathematical calculations to ensure that the data of each measurement node occupies a core position in anomaly detection.

[0100] Step 1.3: For each type of electrical quantity data, the weighted M-estimation method is used to determine the dynamic reference benchmark corresponding to the electrical quantity data based on the normalized topology weights.

[0101] The weighted M-estimation method is a robust statistical method that combines weight allocation and the M-estimation concept. M-estimation estimates parameters by minimizing a predefined loss function. By introducing weights based on M-estimation, the contribution of different measurement nodes to parameter estimation can be further adjusted. In other words, the core of the weighted M-estimation method is to find a set of parameter estimates that minimizes the weighted loss function. In one possible implementation, the dynamic reference benchmark satisfies the following formula:

[0102]

[0103] In the formula, For the first The dynamic reference benchmark corresponding to the electrical quantity data =1,2,… , A pre-defined constant; For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Various electrical quantity data; The number of measurement nodes in the substation; For the first Normalized topological weights of each measurement node; The loss function; The candidate reference value is an intermediate variable in the mathematical optimization process, and it is a value that is continuously updated and tested during the iteration process.

[0104] This implementation method uses the weighted M-estimation method to determine the dynamic reference benchmark corresponding to the electrical quantity data based on the normalized topology weights. This effectively avoids the interference of outliers on the reference benchmark and achieves a combination of statistical estimation and power grid topology information, thereby providing a more accurate and reasonable dynamic reference benchmark for electrical quantity data and providing strong support for the identification of abnormal data in substations.

[0105] S106. Based on the dynamic reference benchmark, identify abnormal data in the target measurement data.

[0106] The dynamic reference benchmark reflects the electrical quantity distribution characteristics of a substation under normal operating conditions.

[0107] For example, for each type of electrical quantity data in the target measurement data, each data point in the electrical quantity data is compared with the dynamic reference benchmark corresponding to that electrical quantity data, and the deviation between each data point and the dynamic reference benchmark is calculated. This deviation can be an absolute value, a relative value, or some form of distance measurement. If the deviation between the data point and the dynamic reference benchmark exceeds a set anomaly judgment threshold, the data point is considered as an abnormal data.

[0108] Optionally, in practical applications, alarms can also be triggered for abnormal data. Alarms can be sent to maintenance personnel by pushing corresponding alarm information. For example, alarms can be pushed to the maintenance personnel's mobile terminals (such as mobile phones) or to the monitoring screen in the central control room. Specifically, alarm information may include the abnormal data and the time when the abnormal data occurred.

[0109] In this embodiment, considering that the distribution characteristics of measurement data may differ under different power grid operating scenarios, data processing rules matching the operating scenario are adopted to process the measurement data. Target measurement data that meets the requirements for anomaly data identification under the operating scenario is generated specifically, improving data quality and facilitating the determination of reference benchmarks and the identification of anomalies. Furthermore, when determining the dynamic reference benchmark, the impact of local faults or topology changes on the distribution of measurement data is fully considered. By combining the real-time power grid topology and target measurement data to determine the reference benchmark, timely responses to power grid changes are made, improving the accuracy and reliability of the dynamic reference benchmark. This, in turn, improves the accuracy of anomaly data identification, avoids false or missed reports during substation operation and maintenance, improves substation operation and maintenance efficiency, and ensures the safe and stable operation of substations.

[0110] Furthermore, in one possible implementation, the target operating scenario of the power grid where the substation is located is determined based on measurement data, including:

[0111] Step 2.1: Determine the comprehensive disturbance index of the measurement data. The comprehensive disturbance index reflects the degree of comprehensive disturbance of the electrical quantities of the substation at a specific moment.

[0112] The comprehensive disturbance index quantifies the degree of disturbance experienced by the power grid system at a specific moment. For example, the comprehensive disturbance index satisfies the following formula:

[0113]

[0114] In the formula, For measurement data in The overall disturbance index at any given time; This is the instantaneous rate of change weighting coefficient, used to adjust the proportion of instantaneous changes in the state vector in the comprehensive disturbance index; This is the recent volatility weighting coefficient, used to adjust the proportion of recent volatility in the state vector in the comprehensive disturbance index; for A multidimensional state vector composed of measurement data at any given time; In the time window The statistical standard deviation of the internal multidimensional state vector; In the time window The statistical mean of the internal multidimensional state vector; The length of the time window; It is a very small positive number.

[0115] It should be understood that express A multidimensional state vector composed of various electrical quantity data from all measurement nodes at any given time, where the dimension equals the number of types of electrical quantity data. For example, if the measurement data includes voltage, current, and active power, then... The state vector is a 3-dimensional state vector, and each dimension of the state vector can be decomposed into multiple (the number of which is the same as the number of measurement nodes) independent univariate time series (e.g., voltage series, current series, or active power series).

[0116] For each independent univariate time series, its value within that time window is calculated separately. The standard deviation and mean within the range. The arithmetic mean of the standard deviations of all individual electrical quantities is calculated to obtain a single scalar value that comprehensively represents the fluctuation level of all electrical quantities, namely the statistical standard deviation. The larger the standard deviation, the more unstable the fluctuation of the electrical quantity.

[0117] Similarly, the arithmetic mean of the mean results of all individual electrical quantities is calculated to obtain a single scalar value that comprehensively represents the average magnitude of all electrical quantities, namely the statistical mean. The statistical mean is used to normalize the standard deviation to eliminate the influence of dimensions.

[0118] The 2-norm operation enables the dimensionality reduction, aggregation, and compression of a multidimensional state vector into a scalar representing the combined magnitude of the rate of change of all electrical quantities or the instantaneous relative change intensity of the state.

[0119] Step 2.2: If the comprehensive disturbance index is less than or equal to the first set threshold, the target operating scenario is determined to be in steady state.

[0120] The first set threshold represents the steady-state-transient critical threshold. When the comprehensive disturbance index is less than or equal to the steady-state-transient critical threshold, it indicates that the power grid is operating smoothly, the electrical quantity changes slowly and the fluctuation range is small, and the target operating scenario is determined to be steady state.

[0121] Step 2.3: If the comprehensive disturbance index is greater than the first set threshold and the comprehensive disturbance index is less than or equal to the second set threshold, then the target operating scenario is determined to be transient.

[0122] The second threshold represents the critical threshold for transient-topology switching. When the comprehensive disturbance index is greater than the steady-state-transient critical threshold and less than or equal to the transient-topology switching critical threshold, it indicates that the power grid has dynamic processes such as load changes and fault transients. The electrical quantities change rapidly over time, but the network topology remains unchanged. The target operating scenario is determined to be transient.

[0123] Step 2.4: If the comprehensive disturbance index is greater than the second set threshold, the target operating scenario is determined to be in the topology switching state.

[0124] When the overall disturbance index is greater than the critical threshold of the transient-topology switching state, it indicates that the network topology has changed, and the target operating scenario is determined to be in the topology switching state.

[0125] In this embodiment of the application, the calculation of the comprehensive disturbance index takes into account the measurement data of multiple measurement nodes, which can more accurately reflect the operating status of the power grid. The application of the comprehensive disturbance index to determine the operating scenario of the power grid improves the accuracy of operating scenario identification.

[0126] In some embodiments, determining abnormal data in the target measurement data based on a dynamic reference benchmark includes: for each type of electrical quantity data corresponding to each of the multiple measurement nodes, determining the normalized abnormal score of the electrical quantity data relative to the dynamic reference benchmark corresponding to the electrical quantity data; if the normalized abnormal score is greater than or equal to a set abnormal judgment threshold, then the measurement node is determined as an abnormal measurement node, and the electrical quantity data is determined as abnormal data in the target measurement data.

[0127] It should be understood that the normalized anomaly score measures the degree of deviation of electrical quantity data from the dynamic group reference baseline, and is an important indicator for judging whether electrical quantity data is abnormal. The higher the normalized anomaly score, the greater the deviation of the data point from the normal range, and the higher the probability of an anomaly. When identifying abnormal data in the target measurement data, anomaly identification is performed separately for each type of electrical quantity data corresponding to each measurement node.

[0128] For example, the normalized outlier score is calculated according to the following formula:

[0129]

[0130] In the formula, For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Normalized anomaly scores for various electrical quantity data; For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Various electrical quantity data; For the first The dynamic reference benchmark corresponding to the electrical quantity data =1,2,… , A pre-defined constant; As a scenario impact factor, the criteria for anomaly judgment can be adjusted according to the specific operating scenario (such as steady state, transient state, topology switching state) to make the calculation results more consistent with the actual situation; The weighted median absolute deviation is used to dynamically assess the normal fluctuation range of current electrical quantity data. The value can be obtained through the following methods: calculation Each data point is relative The residuals are calculated, resulting in multiple residuals. The weighted median of these residuals is then calculated. The absolute deviation of each residual from this weighted median is then calculated. Finally, the absolute deviations are weighted and aggregated based on the normalized topological weights of each measurement point to obtain the final result. .

[0131] When the calculated normalized anomaly score is greater than or equal to the set anomaly judgment threshold, the measurement node is identified as an abnormal measurement node, and the electrical quantity data is identified as abnormal data in the target measurement data.

[0132] This application embodiment comprehensively considers the actual measurement data of the measurement node, the dynamic reference benchmark, the scene influence factor, and the normal fluctuation range of the data, so as to more accurately quantify the degree of deviation of electrical quantity data from the dynamic group reference benchmark, thereby improving the accuracy of abnormal data identification and providing an effective solution for abnormal data detection in substation data analysis.

[0133] In practical applications, considering the differences in the attributes of the measuring devices corresponding to different measuring nodes in a substation, such as current transformer ratio, voltage transformer ratio, range, transformer vector group, and installation direction, these attribute differences may lead to inequivalence issues (such as dimensions and phase) in the measurement data collected from different measuring nodes, which can easily be misjudged as abnormal data. Therefore, it is necessary to perform a unified mapping on the acquired measurement data to eliminate differences in dimensions, phase, etc., and form a comparable dataset. One feasible method for acquiring measurement data from multiple measuring nodes specifically includes:

[0134] Step 3.1: Obtain the original measurement data of multiple measurement nodes and the pre-built engineering ledger information corresponding to the substation. The engineering ledger information stores the mapping relationship between measurement nodes and current transformer ratio, voltage transformer ratio, measurement device range, transformer vector group, current transformer installation direction, and measurement data type.

[0135] Among them, the project ledger information is constructed based on the actual physical environment information of the substation, recording the mapping relationship between each measurement node and key attributes. Key attributes can be collected from sources such as substation design documents, equipment manuals, and installation records.

[0136] For example, the raw measurement data can be obtained directly from the measurement devices at each measurement node; or from the substation automation system via a communication interface, which integrates the monitoring and control of various devices within the substation, including data acquisition from measurement devices; or from a cloud platform, which can integrate and store data from different systems and measurement nodes.

[0137] Project ledger information can be obtained from substation automation systems, cloud platforms, local databases, etc.

[0138] This application does not restrict the method of obtaining original measurement data and engineering ledger information.

[0139] Step 3.2: For each of the multiple measurement nodes, perform amplitude normalization processing on the original measurement data of the measurement node according to the current transformer ratio, voltage transformer ratio, measurement device range and measurement data type corresponding to the measurement node, to obtain normalized measurement data.

[0140] The types of measured data refer to the types of electrical quantity data, including current, voltage, active power, reactive power, etc. The current transformer ratio converts the secondary current of a current transformer into a primary current value. The voltage transformer ratio converts the secondary voltage of a voltage transformer into a primary voltage value. The measuring device range refers to the maximum and minimum values ​​that can be accurately measured.

[0141] For example, the secondary side data is restored to the primary side data by applying the current transformer ratio and the voltage transformer ratio, respectively; the primary side data within the range is normalized by scaling the data proportionally to fall into a specific interval, such as [0,1] or [-1,1].

[0142] Step 3.3: Based on the transformer vector group and current transformer installation direction corresponding to the measurement node, perform phase alignment processing on the normalized measurement data to obtain the measurement data corresponding to the measurement node.

[0143] The transformer vector group describes the phase relationship between transformer windings; different transformer vector groups have different phases. The installation direction of the current transformer affects the phase of the current measurement.

[0144] For example, based on the transformer vector group and the current transformer installation direction, the phase compensation angle of each measurement node relative to the system standard reference phase (usually the specified bus voltage phase) is calculated. For instance, based on the electromagnetic transformation relationship established by Faraday's law of electromagnetic induction and Kirchhoff's laws, and the phasor analysis method for AC circuits, the fixed phase compensation angle of each measurement node relative to the specified bus voltage standard reference phase is calculated using the transformer vector group and the current transformer installation direction. The phase compensation angle is then input into the phase alignment correction formula to rotate the normalized measurement data by the corresponding phase compensation angle, making it consistent with the system standard reference phase. The phase alignment correction formula satisfies the following form:

[0145]

[0146] In the formula, For the first Measurement data from each measurement node after amplitude normalization and phase alignment processing; For the first Normalized measurement data of each measurement node; For the first The comprehensive transformation ratio and caliber conversion factor of each measurement node, which is equal to the product of the current transformer ratio, the voltage transformer ratio and the range conversion ratio of the measuring device. For the first Phase compensation angle at the location of each measurement node; It is a rotation operator used to rotate normalized measurement data by a specified angle in the complex plane to achieve alignment with the system reference phase.

[0147] It should be noted that the original measurement data can be phase aligned first, and then the phase aligned data can be normalized to obtain the measurement data corresponding to the measurement node.

[0148] In this embodiment, engineering ledger information is used to perform amplitude normalization and phase alignment correction on multi-source measurement data, unifying the original data of different measurement nodes into the same reference system, eliminating dimensional and phase differences, improving data quality, and avoiding problems such as inaccurate identification of abnormal data and false alarms due to data inequivalence.

[0149] Furthermore, in some embodiments, the measurement data carries mapping metadata, which is used to mark the source of each data in the measurement data; correspondingly, the substation abnormal data identification method further includes: fusing the abnormal data in the target measurement data and the mapping metadata corresponding to the abnormal data to obtain the abnormal data evidence package of the substation.

[0150] For example, the mapping metadata indicates that the data in the measurement data originates from a certain measurement node, a certain measurement device, or a certain point in time. It should be understood that the data carries mapping metadata throughout the entire process of acquiring raw measurement data, determining target measurement data, and identifying abnormal data.

[0151] Mapping metadata allows for the tracking of the specific source of anomalous data, enhancing traceability, facilitating rapid problem identification, reducing troubleshooting time, and improving operational efficiency. Anomaly data evidence packages provide strong support for anomaly analysis, fault prediction and health management, and decision support.

[0152] Figure 2 This is a schematic diagram of the substation abnormal data identification device provided in the embodiments of this application. The substation is equipped with multiple measurement nodes, such as... Figure 2 As shown, the substation abnormal data identification device 20 provided in this embodiment includes:

[0153] The acquisition module 21 is used to acquire the power grid topology corresponding to the substation and the measurement data of the multiple measurement nodes, wherein the measurement data includes a variety of electrical quantity data;

[0154] The first determining module 22 is used to determine the target operating scenario of the power grid where the substation is located based on the measurement data.

[0155] The second determining module 23 is used to determine the target data processing rules for the measurement data based on the correspondence between the preset operating scenarios and data processing rules, according to the target operating scenario.

[0156] Processing module 24 is used to process the measurement data by applying the target data processing rules to obtain the target measurement data required for abnormal data identification in the target operating scenario;

[0157] The third determining module 25 is used to determine the dynamic reference benchmark corresponding to each type of electrical quantity data based on the power grid topology for each type of electrical quantity data.

[0158] The fourth determining module 26 is used to determine abnormal data in the target measurement data based on the dynamic reference benchmark.

[0159] In one possible implementation, the third determining module 25 is specifically used for: determining the electrical distance between any two measuring nodes among multiple measuring nodes based on the power grid topology; determining the normalized topology weight of each measuring node among multiple measuring nodes according to the electrical distance, wherein the normalized topology weight reflects the degree of influence of the measuring node on the electrical quantity data of other measuring nodes; and determining the dynamic reference benchmark corresponding to the electrical quantity data for each type of electrical quantity data using the weighted M estimation method according to the normalized topology weight.

[0160] In one possible implementation, the dynamic reference datum satisfies the following formula:

[0161]

[0162] In the formula, For the first The dynamic reference benchmark corresponding to the electrical quantity data =1,2,… , A pre-defined constant; These are candidate reference values ​​during the optimization process; For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Various electrical quantity data; The number of measurement nodes in the substation; For the first Normalized topological weights of each measurement node; This is the loss function.

[0163] In one possible implementation, the first determining module 22 is specifically used to: determine the comprehensive disturbance index of the measurement data, the comprehensive disturbance index reflecting the comprehensive disturbance degree of the electrical quantity of the substation at a specific time; if the comprehensive disturbance index is less than or equal to a first set threshold, then the target operating scenario is determined to be steady state; if the comprehensive disturbance index is greater than the first set threshold, and the comprehensive disturbance index is less than or equal to a second set threshold, then the target operating scenario is determined to be transient state; if the comprehensive disturbance index is greater than the second set threshold, then the target operating scenario is determined to be topology switching state.

[0164] In one possible implementation, the combined disturbance index satisfies the following formula:

[0165]

[0166] In the formula, For measurement data in The overall disturbance index at any given time; This is the instantaneous rate of change weighting coefficient, used to adjust the proportion of instantaneous changes in the state vector in the comprehensive disturbance index; This is the recent volatility weighting coefficient, used to adjust the proportion of recent volatility in the state vector in the comprehensive disturbance index; for A multidimensional state vector composed of measurement data at any given time; In the time window The statistical standard deviation of the internal multidimensional state vector; In the time window The statistical mean of the internal multidimensional state vector; The length of the time window; It is a very small positive number.

[0167] In one possible implementation, the fourth determining module 26 is specifically used to: determine the normalized anomaly score of the electrical quantity data relative to the dynamic reference benchmark corresponding to each type of electrical quantity data for each of the multiple measuring nodes; if the normalized anomaly score is greater than or equal to a set anomaly judgment threshold, then the measuring node is determined as an abnormal measuring node, and the electrical quantity data is determined as abnormal data in the target measuring data.

[0168] In one possible implementation, the acquisition module 21 is specifically used to: acquire the original measurement data of multiple measurement nodes and the pre-built engineering ledger information corresponding to the substation. The engineering ledger information stores the mapping relationship between the measurement nodes and the current transformer ratio, voltage transformer ratio, measurement device range, transformer vector group, current transformer installation direction, and measurement data type. For each of the multiple measurement nodes, the original measurement data of the measurement node is normalized according to the current transformer ratio, voltage transformer ratio, measurement device range, and measurement data type corresponding to the measurement node to obtain normalized measurement data. The normalized measurement data is phase aligned according to the transformer vector group and current transformer installation direction corresponding to the measurement node to obtain the measurement data corresponding to the measurement node.

[0169] In one possible implementation, the measurement data carries mapping metadata, which is used to mark the source of each data in the measurement data; correspondingly, the processing module 24 is also used to: fuse the abnormal data in the target measurement data and the mapping metadata corresponding to the abnormal data to obtain the abnormal data evidence package of the substation.

[0170] The substation abnormal data identification device provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0171] Figure 3 A schematic diagram of the structure of the electronic device provided in this application. Figure 3 As shown, the electronic device 30 provided in this embodiment includes at least one processor 301 and a memory 302. Optionally, the electronic device 30 further includes a communication component 303. The processor 301, memory 302, and communication component 303 are connected via a bus 304.

[0172] In a specific implementation, at least one processor 301 executes computer execution instructions stored in memory 302, causing at least one processor 301 to perform the above-described method.

[0173] The specific implementation process of processor 301 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0174] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0175] The memory may include random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk storage device.

[0176] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0177] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0178] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0179] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0180] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in the device.

[0181] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0182] The units described as separate components may or may not be physically separate. The components shown as units 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0183] In addition, the functional units 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.

[0184] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0185] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0186] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for identifying abnormal data in a substation, characterized in that, The substation is equipped with multiple measurement nodes, and the substation abnormal data identification method includes: Obtain the power grid topology corresponding to the substation and the measurement data of the multiple measurement nodes, wherein the measurement data includes various electrical quantity data; Based on the measurement data, the target operating scenario of the power grid where the substation is located is determined; Based on the pre-defined correspondence between operating scenarios and data processing rules, the target data processing rules for the measurement data are determined according to the target operating scenario. The measurement data is processed using the target data processing rules to obtain the target measurement data required for anomaly data identification in the target operating scenario. For each type of electrical quantity data included in the target measurement data, a dynamic reference benchmark corresponding to the electrical quantity data is determined based on the power grid topology. Based on the dynamic reference benchmark, abnormal data in the target measurement data are identified.

2. The substation abnormal data identification method according to claim 1, characterized in that, The step of determining a dynamic reference benchmark corresponding to each electrical quantity data among the various electrical quantity data included in the target measurement data, based on the power grid topology, includes: Based on the power grid topology, determine the electrical distance between any two measurement nodes among the plurality of measurement nodes; Based on the electrical distance, the normalized topology weight of each of the plurality of measurement nodes is determined, and the normalized topology weight reflects the degree of influence of the measurement node on the electrical quantity data of other measurement nodes. For each type of electrical quantity data, the weighted M-estimation method is used to determine the dynamic reference benchmark corresponding to the electrical quantity data based on the normalized topology weights.

3. The substation abnormal data identification method according to claim 2, characterized in that, The dynamic reference datum satisfies the following formula: In the formula, For the first The dynamic reference benchmark corresponding to the electrical quantity data =1,2,… , A pre-defined constant; These are candidate reference values ​​during the optimization process; For the target measurement data, the first The first of the various electrical quantity data corresponding to the measurement node Various electrical quantity data; The number of measurement nodes in the substation; For the first Normalized topological weights of each measurement node; This is the loss function.

4. The substation abnormal data identification method according to any one of claims 1 to 3, characterized in that, The step of determining the target operating scenario of the power grid where the substation is located based on the measurement data includes: Determine the comprehensive disturbance index of the measurement data, wherein the comprehensive disturbance index reflects the comprehensive disturbance degree of the electrical quantities of the substation at a specific time; If the comprehensive disturbance index is less than or equal to the first preset threshold, then the target operating scenario is determined to be in a steady state. If the comprehensive disturbance index is greater than the first preset threshold, and the comprehensive disturbance index is less than or equal to the second preset threshold, then the target operating scenario is determined to be transient. If the comprehensive disturbance index is greater than the second set threshold, then the target operating scenario is determined to be in a topology switching state.

5. The substation abnormal data identification method according to claim 4, characterized in that, The comprehensive disturbance index satisfies the following formula: In the formula, For measurement data in The overall disturbance index at any given time; This is the instantaneous rate of change weighting coefficient, used to adjust the proportion of instantaneous changes in the state vector in the comprehensive disturbance index; This is the recent volatility weighting coefficient, used to adjust the proportion of recent volatility in the state vector in the comprehensive disturbance index; for A multidimensional state vector composed of measurement data at any given time; In the time window The statistical standard deviation of the internal multidimensional state vector; In the time window The statistical mean of the internal multidimensional state vector; The length of the time window; It is a very small positive number.

6. The substation abnormal data identification method according to any one of claims 1 to 3, characterized in that, The step of determining abnormal data in the target measurement data based on the dynamic reference benchmark includes: For each type of electrical quantity data corresponding to each of the plurality of measurement nodes, determine the normalized anomaly score of the electrical quantity data relative to the dynamic reference benchmark corresponding to the electrical quantity data; If the normalized anomaly score is greater than or equal to the set anomaly determination threshold, then the measurement node is determined as an abnormal measurement node, and the electrical quantity data is determined as abnormal data in the target measurement data.

7. The substation abnormal data identification method according to any one of claims 1 to 3, characterized in that, Acquiring measurement data from the multiple measurement nodes includes: The original measurement data of the multiple measurement nodes and the pre-built engineering ledger information corresponding to the substation are obtained. The engineering ledger information stores the mapping relationship between the measurement nodes and the current transformer ratio, voltage transformer ratio, measurement device range, transformer vector group, current transformer installation direction, and measurement data type. For each of the plurality of measurement nodes, the original measurement data of the measurement node is normalized according to the current transformer ratio, voltage transformer ratio, measurement device range and measurement data type corresponding to the measurement node, so as to obtain normalized measurement data. Based on the transformer vector group and current transformer installation direction corresponding to the measurement node, the normalized measurement data is phase aligned to obtain the measurement data corresponding to the measurement node.

8. The substation abnormal data identification method according to claim 7, characterized in that, The measurement data carries mapping metadata, which is used to mark the source of each data point in the measurement data. Correspondingly, the substation abnormal data identification method further includes: By integrating the abnormal data in the target measurement data and the corresponding mapping metadata, an abnormal data evidence package of the substation is obtained.

9. A substation abnormal data identification device, characterized in that, The substation is equipped with multiple measurement nodes, and the substation abnormal data identification device includes: The acquisition module is used to acquire the power grid topology corresponding to the substation and the measurement data of the multiple measurement nodes, wherein the measurement data includes various electrical quantity data; The first determining module is used to determine the target operating scenario of the power grid where the substation is located based on the measurement data. The second determining module is used to determine the target data processing rules for the measurement data based on the correspondence between the preset operating scenarios and data processing rules, according to the target operating scenario. The processing module is used to process the measurement data by applying the target data processing rules to obtain the target measurement data required for abnormal data identification in the target operating scenario. The third determining module is used to determine the dynamic reference benchmark corresponding to each of the various electrical quantity data based on the power grid topology. The fourth determination module is used to determine abnormal data in the target measurement data based on the dynamic reference benchmark.

10. An electronic device, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed, are used to implement the method as described in any one of claims 1 to 8.

Images