Power dispatching intelligent operation ticket generation method based on multi-source data and related equipment
By constructing multi-source data influence relationships and topological data influence relationships, and combining clustering algorithms to generate power operation master tickets and auxiliary tickets, the problems of accuracy and adaptability in generating power dispatch operation tickets are solved, and efficient dispatch operations are realized in complex power grid environments.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- STATE GRID ZHEJIANG ELECTRIC POWER CO LTD
- Filing Date
- 2026-04-23
- Publication Date
- 2026-05-26
AI Technical Summary
Existing methods for generating power dispatch operation tickets suffer from low accuracy and poor adaptability. They are unable to quickly adapt to dynamic changes in power grid topology and complex fault scenarios, and are easily affected by different operating conditions, leading to deviations in matching results.
Based on multi-source data, the relationship between operational data and topology data is constructed. Basic operation tickets are generated through fault simulation, and power operation master tickets and auxiliary tickets are generated using clustering algorithms. Finally, the operation tickets are generated by combining current scheduling requirements and preset operation rules.
It improves the accuracy and adaptability of power dispatch operation ticket generation, reduces the risk of redundant operations, enhances the accuracy and efficiency of dispatch operations, and ensures adaptability under different fault scenarios and real-time power grid conditions.
Smart Images
Figure CN122089006A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of power system dispatching technology, and in particular to a method and related equipment for generating intelligent operation tickets for power dispatching based on multi-source data. Background Technology
[0002] Power dispatch operation tickets are core technical documents that guide power grid operation and ensure the safe and stable operation of the power grid during the dispatching and operation of the power system. The accuracy and timeliness of their preparation are directly related to the safety of power grid equipment operation and the reliability of power grid operation.
[0003] Traditional power dispatch operation tickets mainly rely on manual compilation by dispatchers. This method not only consumes a lot of manpower and time and has low compilation efficiency, but is also prone to operational omissions and logical errors due to differences in the experience of dispatchers and human negligence. It is difficult to adapt to the complex and ever-changing power grid operating conditions.
[0004] With the intelligent development of power systems, intelligent operation ticket generation methods have emerged in existing technologies. These methods are mostly based on preset rule bases for ticket generation, but they have obvious limitations: on the one hand, the construction and updating of the rule bases rely on manual sorting, making it difficult to quickly adapt to the dynamic changes in the power grid topology and complex fault scenarios; on the other hand, the verification based on whether the operation behavior matches the historical operation behavior is easily affected by different operating conditions, which can lead to deviations in the matching results. Summary of the Invention
[0005] This application addresses the technical problems of low accuracy and poor adaptability in existing methods for generating dispatch operation tickets. It proposes a power dispatch intelligent operation ticket generation method and related equipment based on multi-source data. By constructing the influence relationship between operation data and topology data, the matching of operation behaviors is transformed into the matching of dispatch results, avoiding the misleading influence of different operation behaviors under different operating conditions. Furthermore, it combines common operations with different operations with the same dispatch results to generate power operation master tickets and power operation auxiliary tickets, simultaneously retaining core dispatch operations and flexible dispatch operations, thereby improving the accuracy and adaptability of power dispatch operation ticket generation.
[0006] To achieve the above technical objectives, this application provides a technical solution: a method for generating intelligent power dispatch operation tickets based on multi-source data, comprising the following steps: The method includes: acquiring historical compliant dispatch operation tickets and the changes in multi-source data before and after their execution, and constructing operation data influence relationships; acquiring historical fault information and the changes in multi-source data before and after the fault, and constructing topological data influence relationships; performing fault simulation based on the current power grid topology, predicting multi-source data changes based on the topological data influence relationships, and generating basic operation tickets based on the predicted values of multi-source data changes and the operation data influence relationships; clustering the basic operation tickets according to the predicted values of multi-source data changes, obtaining a secondary verification set, generating a main power operation ticket based on common operations within the secondary verification set, and generating auxiliary power operation tickets based on differing operations within the secondary verification set; retrieving the main power operation ticket and auxiliary power operation ticket based on the matching results of the predicted values of multi-source data changes for the current dispatch requirements and the secondary verification set; and merging the main power operation ticket and auxiliary power operation ticket based on preset operation rules to generate a power dispatch operation ticket.
[0007] Furthermore, the step of simulating faults based on the current power grid topology, predicting changes in multi-source data by combining topology data influence relationships, and generating basic operation tickets based on the predicted values of multi-source data changes and the influence relationships of operation data includes: obtaining the current power grid topology, matching fault characteristics in the topology data influence relationships, predicting changes in multi-source data based on the matched fault characteristics; performing reverse matching of the predicted values of multi-source data changes and the influence relationships of operation data, obtaining the corresponding scheduling operations, and generating basic operation tickets.
[0008] Furthermore, the step of clustering basic operation tickets according to the predicted values of changes in multi-source data to obtain a secondary verification set, generating a main power operation ticket based on common operations within the secondary verification set, and generating auxiliary power operation tickets based on differential operations within the secondary verification set includes: grouping basic operation tickets with the same predicted values of changes in multi-source data into the same secondary verification set using a clustering algorithm; extracting common operations that exist in all basic operation tickets in the secondary verification set and have the same operation order to generate a main power operation ticket; and taking the remaining operations in the secondary verification set excluding the common operations as differential operations, and generating auxiliary power operation tickets based on the operation order of the differential operations.
[0009] Furthermore, the step of clustering basic operation tickets according to multi-source data change prediction values to obtain a secondary verification set, generating a power operation master ticket based on common operations within the secondary verification set, and generating a power operation auxiliary ticket based on differential operations within the secondary verification set includes: using a clustering algorithm to cluster basic operation tickets with the same state-type multi-source data change prediction values and the same trend-type numerical multi-source data change prediction values into the same secondary verification set; extracting common operations present in all basic operation tickets in the secondary verification set; generating operation behaviors in the power operation master ticket based on common operations; generating the operation order in the power operation master ticket based on the unique operation order among common operations; taking the remaining operations in the secondary verification set excluding common operations as differential operations, and generating a power operation auxiliary ticket based on the operation order of differential operations relative to common operations.
[0010] Furthermore, the step of using clustering algorithms to cluster basic operation tickets with the same state-type multi-source data change prediction values and the same trend of numerical multi-source data change prediction values into the same secondary verification set includes: performing a first clustering of basic operation tickets based on the same state using a spectral clustering algorithm to obtain an initial secondary verification set containing the same state-type multi-source data change prediction values; and performing a second clustering of basic operation tickets in the initial secondary verification set based on the same numerical change trend using a mean drift algorithm to obtain a secondary verification set.
[0011] Furthermore, the matching process between the current scheduling requirement and the secondary verification set includes: if the current scheduling requirement is a fault scheduling requirement, then the predicted value of the multi-source data change of the current scheduling requirement is obtained based on the difference between the multi-source data before and after the current fault; if the current scheduling requirement is an operational scheduling requirement, then the predicted value of the multi-source data change of the current scheduling requirement is obtained based on the difference between the operational scheduling requirement and the current multi-source data; the cosine similarity between the predicted value of the multi-source data change of the current scheduling requirement and the central predicted value of the secondary verification set is calculated; if the cosine similarity is greater than a preset threshold, then the current scheduling requirement is considered to match the secondary verification set.
[0012] Furthermore, the step of generating a power dispatch operation ticket by fusing power operation master tickets and power operation auxiliary tickets based on preset operation rules includes: following the topological connection order of the current power grid topology, taking the power equipment corresponding to the power operation master ticket as the traversal benchmark, sequentially traversing the remaining power equipment, and retrieving the power operation auxiliary tickets corresponding to the power equipment according to the consistency between the traversal order and the operation order; for each power operation auxiliary ticket retrieved, fusing the power operation master ticket and the power operation auxiliary ticket according to the operation order, and performing basic logic verification on the fusion result, until all power operation auxiliary tickets are fused, and outputting the final verified fusion result as a power dispatch operation ticket.
[0013] Furthermore, the step of outputting the final verified fusion result as a power dispatch operation ticket includes: constructing a first failure rule based on the conflict relationship of the same operation sequence among the final verified fusion results; constructing a second failure rule based on the redundancy relationship of secondary equipment in the final verified fusion results; and generating a power dispatch operation ticket based on the first failure rule, the second failure rule, and the fusion result.
[0014] Furthermore, the first failure rule is that only one operation corresponding to the fusion result is valid under the same operation sequence; the second failure rule is that if the failure operation is a secondary equipment operation and there is no redundancy relationship between the current secondary equipment, then the power dispatch operation ticket corresponding to the failure operation is invalid.
[0015] Another technical solution provided in this application is a power dispatching intelligent operation ticket generation device based on multi-source data, used to implement the method described above, including: a data acquisition module for acquiring historical compliant dispatching operation tickets, historical fault information, multi-source power grid data, and current power grid topology data; a relationship construction module, including an operation relationship construction unit for constructing operation data influence relationships and a topology relationship construction unit for constructing topology data influence relationships; a basic generation module for performing fault simulation based on the current power grid topology, predicting multi-source data changes based on topology data influence relationships, and generating basic operation tickets based on the predicted values of multi-source data changes and operation data influence relationships; a main and auxiliary generation module for clustering the basic operation tickets according to the predicted values of multi-source data changes, obtaining a secondary verification set, generating a power operation main ticket based on common operations within the secondary verification set, and generating a power operation auxiliary ticket based on the differential operations within the secondary verification set; and a real-time generation module for retrieving the power operation main ticket and power operation auxiliary ticket based on the matching results of the predicted values of multi-source data changes for the current dispatching needs and the secondary verification set, and fusing the power operation main ticket and power operation auxiliary ticket based on preset operation rules to generate a power dispatching operation ticket.
[0016] The beneficial effects of this application are as follows: 1. By constructing the impact relationship of operational behavior on multi-source data of the power grid through changes in multi-source data before and after the execution of the operation ticket, and by constructing the impact relationship of different faults on multi-source data of the power grid through changes in multi-source data before and after a fault, the operational behavior and fault situation under different operating conditions are reflected through unified multi-source data changes, enabling the generation of adapted basic operation tickets based on the current power grid topology. Furthermore, the basic operation tickets are clustered based on the predicted values of multi-source data changes, and power dispatch master tickets are generated based on the common operations corresponding to fault situations with the same multi-source data changes to ensure core dispatch. Then, the power dispatch master tickets and power operation auxiliary tickets are integrated based on the matching degree between the current actual operating conditions of the power grid and the operating conditions of historical fault scenarios to form the final power dispatch operation ticket. This ensures that the generated operation ticket has a unified core execution logic and can dynamically adapt to different fault scenarios and real-time power grid status, improving the accuracy, timeliness, and adaptability of power dispatch operation ticket generation.
[0017] 2. By grouping predicted values of changes from the same multi-source data into a secondary verification set, operation tickets for all heterogeneous faults and operations under the same operation result are aggregated. This ensures that the secondary verification set covers all fault types and operation scenarios that achieve the desired operation result. The core mandatory operations, i.e., common operations, are extracted to adapt to all heterogeneous faults and operations under the same operation result. Since these common operations are necessary to achieve the predetermined predicted values of changes from the multi-source data, power operation master tickets are generated based on these common operations. This ensures that dispatching can be performed according to the corresponding power operation master ticket regardless of the fault type. Furthermore, because the operation steps in the power operation master ticket are common operations under heterogeneous fault and operation scenarios, personalized unnecessary operations in various faults or operations are eliminated. This solves the problem of redundant steps added to traditional operation tickets to accommodate multiple faults and operations, significantly reducing the cognitive load on operators, reducing the risk of misoperation caused by redundant operations, and improving the accuracy and efficiency of dispatching operations. Attached Figure Description
[0018] Figure 1 This is a flowchart illustrating the intelligent operation ticket generation method for power dispatching based on multi-source data in this application.
[0019] Figure 2 A schematic diagram of the sub-processes that construct the operational data impact relationships for this application.
[0020] Figure 3 This is a schematic diagram of the sub-process for constructing the topological data influence relationship in this application. Detailed Implementation
[0021] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description of this application is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of this application and are only used to explain this application. They do not limit the scope of protection 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.
[0022] like Figure 1 As shown in the first embodiment of this application, the method for generating intelligent operation tickets for power dispatch based on multi-source data includes the following steps: Obtain historical compliant scheduling operation tickets and the changes in multi-source data before and after their execution, and construct the impact relationships of operation data; Acquire historical fault information and changes in multi-source data before and after the fault, and construct the topological data influence relationship; Fault simulation is performed based on the current power grid topology, and the changes in multi-source data are predicted by combining the influence relationship of topology data. Basic operation tickets are generated based on the predicted values of multi-source data changes and the influence relationship of operation data. The basic operation tickets are clustered according to the predicted values of changes in multi-source data to obtain a secondary verification set. Power operation master tickets are generated based on the common operations in the secondary verification set, and power operation auxiliary tickets are generated based on the different operations in the secondary verification set. Based on the matching results of the multi-source data change prediction value of the current scheduling requirements and the secondary verification set, retrieve the main power operation ticket and the auxiliary power operation ticket; Based on preset operation rules, the main power operation ticket and auxiliary power operation ticket are integrated to generate a power dispatch operation ticket.
[0023] In this embodiment, the impact of operational behavior on multi-source data of the power grid is constructed by analyzing changes in multi-source data before and after the execution of the operation ticket. Similarly, the impact of different faults on multi-source data of the power grid is constructed by analyzing changes in multi-source data before and after a fault. Operational behavior and fault conditions under different operating conditions are reflected through unified multi-source data changes, enabling the generation of adapted basic operation tickets based on the current power grid topology. Furthermore, the basic operation tickets are clustered based on predicted values of multi-source data changes. Power dispatch master tickets are generated based on common operations corresponding to fault conditions with the same multi-source data changes to ensure core dispatch. Finally, the power dispatch master ticket and power operation auxiliary tickets are merged based on the matching degree between the current actual power grid operating conditions and historical fault scenarios to form the final power dispatch operation ticket. This ensures that the generated operation ticket possesses a unified core execution logic and can dynamically adapt to different fault scenarios and real-time power grid states, improving the accuracy, timeliness, and adaptability of power dispatch operation ticket generation.
[0024] like Figure 2As shown, historical compliant scheduling operation tickets and their multi-source data changes before and after execution are obtained, and the operation data impact relationships are constructed, including: Extract the first multi-source data before the execution of historical compliant scheduling operation tickets and the second multi-source data after the execution based on the preset time window; The difference between the same dimension data in the first multi-source data and the second multi-source data is taken as the change in the first multi-source data; A correlation analysis was conducted between the differences in operational behavior of different historical compliant scheduling operation tickets and the differences in changes in the first multi-source data. A mapping relationship between operations and changes in multi-source data was established, and the impact relationship of operation data was constructed.
[0025] Retrieve executed and uncancelled scheduling operation tickets from the SCADA system or operation ticket management system as historical compliant scheduling operation tickets. Simultaneously, retrieve multi-source data from preset time windows before and after the operation ticket execution.
[0026] In this embodiment, the multi-source data includes at least power grid topology data, equipment status data, relay protection data, telemetry data, and remote signaling data.
[0027] In some cases, the preset time window for data retrieval is the same before and after execution. For example, if the preset time window is 1 minute, multi-source data is retrieved for the minute before and the minute after the execution of historical compliant scheduling operation tickets.
[0028] In other cases, dual-time-window sampling is employed. The preset time window includes at least a preset pre-execution time window and a preset post-execution time window to ensure that the multi-source data collected is steady-state data before and after execution. In this embodiment, the preset pre-execution time window is 1 minute, and the preset post-execution time window is 3 minutes. Multi-source data from 1 minute before the execution of historical compliant scheduling operation tickets and multi-source data from 3 minutes after the execution are retrieved.
[0029] Since the sampling dimensions are the same before and after the execution of historical compliant dispatch operation tickets, the differences between the first and second multi-source data are calculated based on the same dimensions. In this embodiment, for numerical multi-source data such as voltage, current, and power, the difference or rate of change between the first and second multi-source data is calculated; for state-type multi-source data such as switch position and protection activation / deactivation status, 0-1 encoding is used to represent state changes, and the state difference is used as the change in the first multi-source data before and after the operation ticket execution. Furthermore, for different historical compliant operation tickets, operation behavior features are extracted using feature extraction algorithms, including operation objects, operation steps, operation sequence, and operation parameters. Correlation analysis methods are used to associate the operation behavior features with the changes in the first multi-source data, and the correlation between the differences in different operation behaviors and the differences in data changes is calculated to form the operation data influence relationship.
[0030] In this embodiment, a combination of frequent item mining and cosine similarity is used to construct the operational data influence relationship: First, the Apriori frequent item mining algorithm is used to mine operational behavior features and multi-source data change features to obtain high-frequency co-occurrence rules between operational behavior and multi-source data changes, and to filter out combinations of operations and data changes with strong correlation. Then, the cosine similarity algorithm is used to calculate the similarity of multi-source data changes corresponding to different operational behaviors, quantify the degree of correlation between operational behavior and data changes, and finally establish the operational data influence relationship based on high-frequency co-occurrence and similarity quantification.
[0031] like Figure 3 As shown, obtaining historical fault information and changes in multi-source data before and after the fault, and constructing the topological data influence relationships includes: Extract the third multi-source data before the occurrence of historical faults and the fourth multi-source data after the occurrence of faults according to the preset time window; The differences in the same dimension of data between the third and fourth multi-source data are used as the changes in the second multi-source data. A correlation analysis was conducted between the differences in fault characteristics of different historical faults and the differences in changes in the second multi-source data to establish a mapping relationship between fault characteristics and changes in multi-source data, and to construct the topological data influence relationship.
[0032] Historical fault information includes at least the fault type, fault topology, fault duration, fault handling measures, and fault handling results. In this embodiment, the preset time window for extracting the third multi-source data is set to 5 minutes, and the preset time window for extracting the third multi-source data is set to 2 minutes to ensure the collection of steady-state data. The difference between the third and fourth multi-source data in the same dimension is calculated as the change in the second multi-source data. Fault features of historical faults are extracted, and a combination of frequent term mining and Pearson correlation coefficient is used to associate fault features with changes in the second multi-source data, calculate the correlation between fault features and data changes, and construct the topological data influence relationship.
[0033] Fault simulation is performed based on the current power grid topology. Multi-source data changes are predicted by combining topological data influence relationships. Basic operation tickets are generated based on the predicted multi-source data changes and the influence relationships of operational data, including: Obtain the current power grid topology, match fault characteristics in the topology data influence relationships, and predict changes in multi-source data based on the matched fault characteristics; Perform opposite matching of the predicted changes in multi-source data and the impact relationship of operational data to obtain the corresponding scheduling operations and generate basic operation tickets.
[0034] By matching fault characteristics in the current power grid topology and the influence relationship of topology data, it is determined whether there are fault characteristics of historically occurring faults in the current power grid topology, and the changes in multi-source data are predicted based on the matched fault characteristics. Since fault handling usually involves restoring the power grid operating data to the state before the fault occurred, the data change implemented by the dispatch ticket used for fault handling is to adjust the fourth multi-source data to the third multi-source data, and to map the fourth multi-source data to the first multi-source data, and the third multi-source data to the second multi-source data, reflecting the data changes before, during (dispatch before) the fault, and after the dispatch. By matching the data changes throughout the fault process with the data changes before and after the dispatch of the operation ticket, the operation behavior that can adapt to the fault repair can be obtained. This can avoid misleading differences in operation behavior due to inconsistent operating conditions, and ensure that the operation behavior is adapted to the current power grid, thereby improving the accuracy and adaptability of automatically generated operation tickets.
[0035] In this embodiment, the specific process of fault simulation based on the current power grid topology is as follows: First, the current power grid topology data is obtained through the scheduling SCADA system, a power grid simulation model consistent with the actual power grid is constructed, and the model is imported into the power system simulation platform. The current power grid topology is traversed according to the historical fault topology, and the changes in multi-source data corresponding to the matched fault topology are used as the multi-source data change values.
[0036] In some cases, basic operation tickets are clustered according to the predicted values of changes in multi-source data to obtain a secondary verification set. Power operation master tickets are generated based on common operations within the secondary verification set, and power operation auxiliary tickets are generated based on differing operations within the secondary verification set. These include: Basic operation tickets with the same multi-source data change prediction values are grouped into the same secondary verification set using a clustering algorithm; Extract common operations from the secondary verification set that exist in all basic operation tickets and have the same operation order to generate power operation master tickets; The remaining operations in the secondary verification set, excluding common operations, are taken as differential operations, and power operation auxiliary tickets are generated based on the operation order of the differential operations.
[0037] In this scenario, predicted values of changes from the same multi-source data are grouped into a secondary verification set. This enables the aggregation of operation tickets for all heterogeneous faults and operations under the same operational result, ensuring that the secondary verification set covers all fault types and operational scenarios that achieve the operational result. The core mandatory operations, i.e., common operations, are extracted to adapt to all heterogeneous faults and operations under the same operational result. Since these common operations are necessary to achieve the predetermined predicted values of changes from the multi-source data, power operation master tickets are generated based on these common operations. This ensures that dispatching can be performed according to the corresponding power operation master ticket regardless of the fault type. Furthermore, because the operation steps in the power operation master ticket are common operations under heterogeneous fault and operational scenario scenarios, personalized unnecessary operations in various faults or operations are filtered out. This solves the problem of redundant steps added to traditional operation tickets to accommodate multiple faults and operations, significantly reducing the cognitive load on operators, reducing the risk of misoperation caused by redundant operations, and improving the accuracy and efficiency of dispatching operations.
[0038] Meanwhile, power operation auxiliary tickets are generated based on the operation sequence of the remaining operations excluding common operations in the secondary verification set. This preserves the flexibility of actual dispatching. By pre-generating auxiliary tickets instead of the traditional mode of confirming the working conditions on-site before triggering the generation of auxiliary tickets, redundant steps such as additional auxiliary ticket generation and confirmation on-site are eliminated. After verifying the actual working conditions on-site, operators can directly select the appropriate operation sequence from the pre-generated auxiliary tickets without the need for temporary design and generation of auxiliary operation steps on-site. This avoids dispatching interruptions caused by on-site ticketing processes and ensures the efficiency of overall dispatching operations.
[0039] By constructing a hierarchical system of main power operation tickets and auxiliary power operation tickets, the system ensures core dispatching while retaining the option of flexible dispatching, thus balancing the security and flexibility of power operation dispatching.
[0040] Understandably, clustering algorithms can be one or more of K-Means, hierarchical clustering, or DBSCAN.
[0041] In other cases, basic operation tickets are clustered according to the predicted values of changes in multi-source data to obtain a secondary verification set. Power operation master tickets are generated based on common operations within the secondary verification set, and power operation auxiliary tickets are generated based on differing operations within the secondary verification set. Clustering algorithms are used to cluster basic operation tickets with the same state-type multi-source data change prediction values and the same trend of numerical multi-source data change prediction values into the same secondary verification set. Extract the common operations that exist in all basic operation tickets in the secondary verification set; The operation behaviors in the power operation master ticket are generated based on common operations, and the operation sequence in the power operation master ticket is generated based on the unique operation sequence among common operations. The remaining operations in the secondary verification set, excluding common operations, are taken as differential operations. Power operation auxiliary tickets are generated according to the operation order of differential operations relative to common operations.
[0042] In this scenario, power dispatch multi-source data is divided into state-based multi-source data and numerical multi-source data. State-based data includes discrete binary data (0-1) such as switch positions and protection activation / deactivation status, while numerical multi-source data includes data such as voltage, current, and power. Since state-based data only contains 0-1 values, while numerical multi-source data exhibits continuous fluctuations due to grid load and topology changes, it is impossible to generate unique, fixed predicted values. Therefore, dual clustering is performed based on the predicted changes in both state-based and numerical multi-source data to ensure that operation tickets within the secondary verification set correspond to the same basic equipment status and the same continuous changes in grid operation characteristics.
[0043] Specifically, clustering algorithms are used to group basic operational tickets with the same state-type multi-source data change prediction values and the same trend of numerical multi-source data change prediction values into the same secondary validation set, including: Based on the spectral clustering algorithm, the basic operation tickets are first clustered according to the same state to obtain an initial secondary verification set containing the predicted values of changes in multi-source data of the same state. Based on the mean shift algorithm, the basic operation tickets in the initial secondary verification set are clustered a second time according to the same numerical change trend to obtain the secondary verification set.
[0044] In this context, a global coarse clustering of state-type multi-source data changes is performed based on the spectral clustering algorithm. The spectral clustering algorithm uses graph theory to reduce the dimensionality of high-dimensional data and is suitable for high-dimensional clustering involving multiple equipment state changes in power dispatching scenarios. It matches the predicted values of all state-type multi-source data changes in the basic operation ticket one by one to cluster the global power grid state changes brought about by the operation ticket, rather than clustering based on a single state, thus ensuring the consistency of the initial secondary verification set at the global level.
[0045] Meanwhile, the mean-shift algorithm is used to perform local fine clustering of numerical multi-source data changes. The mean-shift algorithm is a density clustering algorithm without a preset number of clusters. It does not require pre-determining the number of clusters based on numerical trends. It can automatically calculate sample density and identify density peaks through kernel functions to complete cluster division. It is suitable for power dispatching scenarios with multiple numerical trend changes that make it impossible to predict the number of trend clusters, ensuring the accuracy of flexible clustering based on numerical change trends.
[0046] Basic operation tickets are dual-clustered using spectral clustering and mean-shift algorithms. Spectral clustering ensures consistency in hard-matched states globally, preventing operation tickets with heterogeneous states from being mixed into the same initial secondary verification set. Mean-shifting performs trend subdivision within a small range of the initial secondary verification set, preserving local data differences and achieving a balance between global accuracy and local flexibility. Furthermore, because coarse clustering is performed beforehand using spectral clustering, the computational cost of clustering by the mean-shift algorithm is minimized, reducing computational burden.
[0047] In this scenario, the operation behavior of the master power operation ticket is generated based on the common operations present in all basic operation tickets within the secondary verification set. This ensures that the power equipment corresponding to the fixed operation behavior can only be executed according to the master power operation ticket. The order of operations between power equipment in the master power operation ticket is determined by the order of common operations. The master power operation ticket is used to determine the necessary operation objects and necessary operation order for the scheduling target, reducing unnecessary and redundant operations. Furthermore, based on the operation order of differentiated operations relative to common operations, auxiliary power operation tickets are generated. These auxiliary power operation tickets serve as personalized supplements to the master power operation tickets. Operators can directly embed the appropriate master power operation ticket operations into the execution flow in a relative order according to the actual on-site conditions, without needing to re-examine the overall operation logic. This maintains the flexibility of on-site scheduling while ensuring the continuity of the operation sequence, adapting to the complex and ever-changing subdivided operating conditions required in power scheduling.
[0048] In this embodiment, retrieving the main power operation ticket and auxiliary power operation ticket based on the matching result of the multi-source data change prediction value of the current scheduling demand and the secondary verification set includes: If the current scheduling requirement is a fault scheduling requirement, then the predicted value of the multi-source data change of the current scheduling requirement is obtained based on the difference of multi-source data before and after the current fault. If the current scheduling requirement is an operational scheduling requirement, then the predicted value of the multi-source data change for the current scheduling requirement is obtained based on the difference between the operational scheduling requirement and the current multi-source data. Calculate the cosine similarity between the predicted value of the multi-source data change of the current scheduling demand and the central predicted value of the secondary verification set. If the cosine similarity is greater than the preset threshold, the current scheduling demand is considered to match the secondary verification set, and the main power operation ticket and auxiliary power operation ticket that match the secondary verification set are retrieved.
[0049] To address the different needs of fault dispatching and operational dispatching, a differentiated approach is adopted to generate corresponding multi-source data change prediction values. This approach can automatically generate operation tickets based on real-time power grid fault data, as well as based on the dispatching data needs of operators' daily operations. This improves the efficiency of fault dispatching, reduces the requirement for operators to have a global understanding, and eliminates the need for operators to remember all dispatching paths. Various dispatching paths that meet the operational dispatching needs of operators can be generated through different combinations of main power operation tickets and auxiliary power operation tickets.
[0050] In this embodiment, the predicted value of the center of the secondary verification set is the predicted value of the multi-source data change corresponding to the cluster center during the second clustering. The preset threshold is 0.85, which ensures that the predicted value of the multi-source data change for the current scheduling requirement can only match one secondary verification set while allowing for a similarity margin.
[0051] Based on preset operation rules, the power operation master ticket and power operation auxiliary ticket are integrated to generate power dispatch operation tickets, including: According to the topological connection order of the current power grid topology, the power equipment corresponding to the main power operation ticket is used as the traversal reference, and the remaining power equipment is traversed in turn. Based on the consistency between the traversal order and the operation order, the auxiliary power operation ticket corresponding to the power equipment is retrieved. For each power operation auxiliary ticket retrieved, the power operation master ticket and power operation auxiliary ticket are merged according to the operation sequence, and the merging result is checked by basic logic until all power operation auxiliary tickets are merged. The final verified merging result is then output as a power dispatch operation ticket.
[0052] By ensuring consistency between the traversal order and the operation order, intermittent operations are avoided. The fusion process is based on the fundamental topological order, followed by a basic logic check for each fusion, reducing unnecessary checks and improving the efficiency of power dispatch operation ticket generation. It's understandable that each fusion of power operation auxiliary tickets is based on the fusion of the previous power operation auxiliary ticket, ensuring that the overall power operation conforms to the fundamental logic.
[0053] The basic logic verification includes at least the five-prevention logic verification for power dispatching and the verification of standard operating procedures. It is understandable that a complete power dispatching operation sequence is considered a power dispatching operation ticket. Multiple power dispatching operation tickets can exist, but in practical applications, to avoid unnecessary duplication of operation tickets, only one master power dispatching ticket is generated to prevent redundant ticket generation.
[0054] In the remaining embodiments, outputting the final verified fusion result as a power dispatch operation ticket includes: The first failure rule is constructed based on the conflict relationship of the same operation sequence among the fusion results that have passed the final verification; A second failure rule is constructed based on the redundancy relationship of secondary equipment in the final verified fusion results; Power dispatch operation tickets are generated based on the first failure rule, the second failure rule, and the fusion result.
[0055] Furthermore, during the execution of power dispatch operation tickets, based on whether there is a first failure rule in the current operation sequence, the operator corresponding to different operations is prompted to determine multi-directional failure. Based on whether there is a second failure rule in the currently failed power operation, the operator is prompted to decide whether to prompt the operator that the corresponding power operation auxiliary ticket is invalid.
[0056] Specifically, the first failure rule is that only one operation corresponding to the fusion result is valid under the same operation sequence. The second failure rule is that if the failure operation is a secondary equipment operation and there is no redundancy relationship between the current secondary equipment, then the power dispatch operation ticket corresponding to the failure operation is invalid.
[0057] The first failure rule resolves potential conflicts and duplicate operations during the integration of primary and secondary operation tickets by retaining only one valid operation under the same sequence. This ensures a clear and non-redundant operation sequence, avoids execution chaos caused by conflicts between multiple sequential operation instructions, and guarantees the continuity of the operation process. The second failure rule prevents grid operation risks caused by secondary equipment operation failures by invalidating the corresponding operation ticket when there is no redundancy in the secondary equipment, balancing the compliance and security of the operation ticket.
[0058] As a second embodiment of this application, a power dispatching intelligent operation ticket generation device based on multi-source data includes: The data acquisition module is used to acquire historical compliant dispatch operation tickets, historical fault information, multi-source power grid data, and current power grid topology data; The relationship building module includes an operational relationship building unit for building operational data influence relationships and a topology relationship building unit for building topology data influence relationships; The basic generation module is used to simulate faults based on the current power grid topology, predict changes in multi-source data by combining the influence relationship of topology data, and generate basic operation tickets based on the predicted values of multi-source data changes and the influence relationship of operation data. The main and auxiliary generation modules are used to cluster basic operation tickets according to the predicted values of changes in multi-source data, obtain a secondary verification set, generate power operation main tickets based on common operations in the secondary verification set, and generate power operation auxiliary tickets based on differential operations in the secondary verification set. The real-time generation module is used to retrieve the main power operation ticket and auxiliary power operation ticket based on the matching result of the multi-source data change prediction value and the secondary verification set of the current scheduling requirements, and to generate a power dispatching operation ticket by merging the main power operation ticket and auxiliary power operation ticket based on preset operation rules.
[0059] In this embodiment, the data acquisition module is connected to the SCADA system or the operation ticket management system; the relationship building module is connected to the data acquisition module; the basic generation module is connected to both the relationship building module and the data acquisition module; the primary and secondary generation modules are connected to the basic generation module; and the real-time generation module is connected to both the data acquisition module and the primary and secondary generation modules. All functional modules interact and transmit commands via a bus. This module can be integrated into the server of the power dispatch master station or deployed in a distributed manner.
[0060] As a third embodiment of this application, a computer-readable storage medium is provided, on which a computer program is stored. When executed by a processor, the computer program implements the power dispatching intelligent operation ticket generation method based on multi-source data described in Embodiment 1. The storage medium is a non-volatile storage medium, including a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a solid-state drive (SSD), an optical disc, etc. The computer program can be compiled into an executable file, adapted to mainstream operating systems such as Windows and Linux, and can be directly deployed on the server of the power dispatching system.
[0061] The specific embodiments described above are preferred embodiments of the power dispatch intelligent operation ticket generation method and related equipment based on multi-source data in this application, and are not intended to limit the specific implementation scope of this application. The scope of this application includes but is not limited to the specific embodiments described above. All equivalent changes made in accordance with the shape and structure of this application are within the protection scope of this application.
Claims
1. A method for generating intelligent operation tickets for power dispatching based on multi-source data, characterized in that: Includes the following steps: Obtain historical compliant scheduling operation tickets and the changes in multi-source data before and after their execution, and construct the impact relationships of operation data; Acquire historical fault information and changes in multi-source data before and after the fault, and construct the topological data influence relationship; Fault simulation is performed based on the current power grid topology, and the changes in multi-source data are predicted by combining the influence relationship of topology data. Basic operation tickets are generated based on the predicted values of multi-source data changes and the influence relationship of operation data. The basic operation tickets are clustered according to the predicted values of changes in multi-source data to obtain a secondary verification set. Power operation master tickets are generated based on the common operations in the secondary verification set, and power operation auxiliary tickets are generated based on the different operations in the secondary verification set. Based on the matching results of the multi-source data change prediction value of the current scheduling requirements and the secondary verification set, retrieve the main power operation ticket and the auxiliary power operation ticket; Based on preset operation rules, the main power operation ticket and auxiliary power operation ticket are integrated to generate a power dispatch operation ticket.
2. The method for generating intelligent operation tickets for power dispatch based on multi-source data as described in claim 1, characterized in that: The process of simulating faults based on the current power grid topology, predicting changes in multi-source data by combining the influence relationships of topology data, and generating basic operation tickets based on the predicted values of multi-source data changes and the influence relationships of operational data includes: Obtain the current power grid topology, match fault characteristics in the topology data influence relationships, and predict changes in multi-source data based on the matched fault characteristics; Perform opposite matching of the predicted changes in multi-source data and the impact relationship of operational data to obtain the corresponding scheduling operations and generate basic operation tickets.
3. The method for generating intelligent operation tickets for power dispatch based on multi-source data as described in claim 1, characterized in that: The process of clustering basic operation tickets according to the predicted values of changes in multi-source data to obtain a secondary verification set, generating a main power operation ticket based on common operations within the secondary verification set, and generating auxiliary power operation tickets based on differing operations within the secondary verification set includes: Basic operation tickets with the same multi-source data change prediction values are grouped into the same secondary verification set using a clustering algorithm; Extract common operations from the secondary verification set that exist in all basic operation tickets and have the same operation order to generate power operation master tickets; The remaining operations in the secondary verification set, excluding common operations, are taken as differential operations, and power operation auxiliary tickets are generated based on the operation order of the differential operations.
4. The method for generating intelligent operation tickets for power dispatch based on multi-source data as described in claim 1, characterized in that: The process of clustering basic operation tickets according to the predicted values of changes in multi-source data to obtain a secondary verification set, generating a main power operation ticket based on common operations within the secondary verification set, and generating auxiliary power operation tickets based on differing operations within the secondary verification set includes: Clustering algorithms are used to cluster basic operation tickets with the same state-type multi-source data change prediction values and the same trend of numerical multi-source data change prediction values into the same secondary verification set. Extract the common operations that exist in all basic operation tickets in the secondary verification set; The operation behaviors in the power operation master ticket are generated based on common operations, and the operation sequence in the power operation master ticket is generated based on the unique operation sequence among common operations. The remaining operations in the secondary verification set, excluding common operations, are taken as differential operations. Power operation auxiliary tickets are generated according to the operation order of differential operations relative to common operations.
5. The method for generating intelligent operation tickets for power dispatching based on multi-source data as described in claim 4, characterized in that: The method of using clustering algorithms to cluster basic operation tickets with the same state-type multi-source data change prediction values and the same trend of numerical multi-source data change prediction values into the same secondary verification set includes: Based on the spectral clustering algorithm, the basic operation tickets are first clustered according to the same state to obtain an initial secondary verification set containing the predicted values of changes in multi-source data of the same state. Based on the mean shift algorithm, the basic operation tickets in the initial secondary verification set are clustered a second time according to the same numerical change trend to obtain the secondary verification set.
6. The method for generating intelligent operation tickets for power dispatch based on multi-source data as described in claim 1, characterized in that: The matching process between the current scheduling requirement and the secondary verification set includes: If the current scheduling requirement is a fault scheduling requirement, then the predicted value of the multi-source data change of the current scheduling requirement is obtained based on the difference of multi-source data before and after the current fault. If the current scheduling requirement is an operational scheduling requirement, then the predicted value of the multi-source data change for the current scheduling requirement is obtained based on the difference between the operational scheduling requirement and the current multi-source data. Calculate the cosine similarity between the predicted value of the multi-source data change of the current scheduling requirement and the central predicted value of the secondary verification set. If the cosine similarity is greater than a preset threshold, the current scheduling requirement is considered to match the secondary verification set.
7. The method for generating intelligent operation tickets for power dispatching based on multi-source data as described in claim 1, characterized in that: The process of generating a power dispatch operation ticket by integrating the main power operation ticket and auxiliary power operation ticket based on preset operation rules includes: According to the topological connection order of the current power grid topology, the power equipment corresponding to the main power operation ticket is used as the traversal reference, and the remaining power equipment is traversed in turn. Based on the consistency between the traversal order and the operation order, the auxiliary power operation ticket corresponding to the power equipment is retrieved. For each power operation auxiliary ticket retrieved, the power operation master ticket and power operation auxiliary ticket are merged according to the operation sequence, and the merging result is checked by basic logic until all power operation auxiliary tickets are merged. The final verified merging result is then output as a power dispatch operation ticket.
8. The method for generating intelligent operation tickets for power dispatching based on multi-source data as described in claim 7, characterized in that: The step of outputting the final verified fusion result as a power dispatch operation ticket includes: The first failure rule is constructed based on the conflict relationship of the same operation sequence among the fusion results that have passed the final verification; A second failure rule is constructed based on the redundancy relationship of secondary equipment in the final verified fusion results; Power dispatch operation tickets are generated based on the first failure rule, the second failure rule, and the fusion result.
9. The method for generating intelligent operation tickets for power dispatching based on multi-source data as described in claim 8, characterized in that: The first failure rule is that under the same operation sequence, only one operation corresponding to the fusion result is valid; The second failure rule is that if the failure operation is a secondary equipment operation and there is no redundancy relationship between the current secondary equipment, then the power dispatch operation ticket corresponding to the failure operation is invalid.
10. A power dispatching intelligent operation ticket generation device based on multi-source data, used to implement the method as described in any one of claims 1 to 9, characterized in that: include: The data acquisition module is used to acquire historical compliant dispatch operation tickets, historical fault information, multi-source power grid data, and current power grid topology data; The relationship building module includes an operational relationship building unit for building operational data influence relationships and a topology relationship building unit for building topology data influence relationships; The basic generation module is used to simulate faults based on the current power grid topology, predict changes in multi-source data by combining the influence relationship of topology data, and generate basic operation tickets based on the predicted values of multi-source data changes and the influence relationship of operation data. The main and auxiliary generation modules are used to cluster basic operation tickets according to the predicted values of changes in multi-source data, obtain a secondary verification set, generate power operation main tickets based on common operations in the secondary verification set, and generate power operation auxiliary tickets based on differential operations in the secondary verification set. The real-time generation module is used to retrieve the main power operation ticket and auxiliary power operation ticket based on the matching result of the multi-source data change prediction value and the secondary verification set of the current scheduling requirements, and to generate a power dispatching operation ticket by merging the main power operation ticket and auxiliary power operation ticket based on preset operation rules.