CN-120611901-B - Power network data driving optimization method and system based on dynamic authority modeling

CN120611901BCN 120611901 BCN120611901 BCN 120611901BCN-120611901-B

Abstract

The application relates to the technical field of data analysis, and provides a power network data driving optimization method and system based on dynamic authority modeling, which are used for improving the self-adaptability and risk resistance of a power network in a complex operation environment. The method comprises the steps of obtaining an electric power network operation data set, carrying out dynamic authority modeling processing on the electric power network operation data set, generating a user authority feature set and a device authority feature set, generating an electric power resource dynamic allocation strategy according to the user authority feature set and the device authority feature set, and feeding back the electric power resource dynamic allocation strategy to an electric power network control system to activate authority configuration updating operation. Therefore, through the deep coupling of the authority model and the resource scheduling, a technical path which gives consideration to elasticity and reliability to the intelligent upgrading of the power system is provided, and thus the self-adaptability and the risk resistance of the power network in a complex operation environment can be improved.

Inventors

LIU LEI
HUO JIAHAO
YUAN CHAO
GUAN TI
ZHANG KUN
BAI YINGWEI
LIANG DONG
ZHOU JIANBO
CHAI YUAN
GAO KAIQIANG

Assignees

国网山东省电力公司莱芜供电公司
国网山东省电力公司
山东思极科技有限公司

Dates

Publication Date: 20260505
Application Date: 20250527

Claims (5)

1. A power network data driven optimization method based on dynamic rights modeling, the method comprising: Acquiring an electric power network operation data set, wherein the electric power network operation data set comprises real-time equipment state data, user operation behavior logs and network topology connection relations; Performing dynamic authority modeling processing on the power network operation data set to generate a user authority feature set and a device authority feature set, wherein the user authority feature set represents access control parameters of different user roles to power resources, and the device authority feature set represents resource allocation authorities of different power devices in an operation environment; Generating a power resource dynamic allocation strategy according to the user authority characteristic set and the equipment authority characteristic set, wherein the power resource dynamic allocation strategy is used for adjusting a load balancing path and equipment control priority in a power network; Feeding back the dynamic power resource allocation strategy to a power network control system to activate authority configuration updating operation, wherein the authority configuration updating operation comprises adjustment of a user operation authority range and a resource response rule of power equipment; The dynamic authority modeling processing is performed on the power network operation data set to generate a user authority feature set and a device authority feature set, and the method comprises the following steps: Extracting an operation type sequence and an operation time stamp sequence in the user operation behavior log, wherein the operation type sequence comprises a control instruction type of a user role on the power equipment and corresponding operation times; generating a user authority dynamic evolution map based on the operation type sequence and the operation time stamp sequence, wherein the user authority dynamic evolution map comprises operation authority change tracks of user roles in a preset time window; analyzing the device relevance of the network topology connection relation to generate a device authority dependency graph, wherein the device authority dependency graph reflects the resource calling relation and the authority sharing constraint condition between the power devices; Constructing a user authority feature set and a device authority feature set according to the user authority dynamic evolution map and the device authority dependency map, wherein the user authority feature set comprises access frequency features and operation legality labels of a user role on target power devices, and the device authority feature set comprises resource occupancy rate features and authority conflict detection parameters of the power devices; the generating a user authority dynamic evolution map based on the operation type sequence and the operation time stamp sequence comprises the following steps: Performing authority influence degree evaluation on the operation type sequence, and determining influence weight of each operation type on user authority change based on probability calculation of triggering authority adjustment of different operation types in the history authority change record; Generating permission change time interval distribution based on the operation time stamp sequence, wherein the time interval distribution is used for representing a dynamic association relation between the time intensity of user operation behaviors and the permission adjustment frequency; performing association analysis on the influence weight and the time interval distribution, and calculating authority state association strength of the user role in a continuous time window to generate node connection strength parameters in a user authority dynamic evolution map, wherein the node connection strength parameters quantify probability and time sequence dependency relationship of authority state transition of adjacent time windows; Constructing a dynamic authority state transition matrix based on the node connection strength parameters, wherein a row dimension of the matrix represents an authority state set of a current time window, a column dimension represents an authority state set of a next time window, and matrix element values are obtained by carrying out normalization processing on the corresponding node connection strength parameters and are used for reflecting the transition weight of the authority state across the time window; Mapping the authority state of each time window into time sequence nodes with time stamp attribute, connecting adjacent time sequence nodes according to time evolution sequence to form directed edges, wherein each directed edge carries a normalized connection strength weight value; Embedding a multi-dimensional authority feature vector in the time sequence node, wherein the multi-dimensional authority feature vector is formed by an operation type distribution histogram in a corresponding time window, and statistical features of authority influence degree accumulated values and time interval distribution; Generating the user authority dynamic evolution map by fusing the space-time topological structure and the multidimensional authority feature vector, wherein the topological edge weight of the user authority dynamic evolution map represents authority state transition probability, the node feature vector represents the dynamic attribute of the authority state, and the user authority dynamic evolution map is a visual map model comprising a time dimension evolution path and a space dimension authority association rule; The generating a dynamic allocation strategy of the power resource according to the user authority characteristic set and the device authority characteristic set comprises the following steps: performing matching degree calculation on the access frequency characteristics in the user authority characteristic set and the resource occupancy rate characteristics in the equipment authority characteristic set to generate a user-equipment authority matching matrix; determining a target power equipment set with resource conflict according to the user-equipment permission matching matrix, wherein the resource conflict comprises that a user access request exceeds equipment resource capacity or permission sharing rules do not meet security constraint; Generating a dynamic load balancing path and a permission priority adjustment instruction based on the resource conflict type of the target power equipment set, and determining a power resource dynamic allocation strategy by combining the dynamic load balancing path and the permission priority adjustment instruction; The step of generating a dynamic load balancing path includes: Acquiring real-time load rate data and topology connection states of all devices in an electric power network; constructing a device load distribution thermodynamic diagram according to the real-time load rate data, wherein the device load distribution thermodynamic diagram reflects the resource residual capacity and overload risk level of the power devices in different areas; generating a plurality of candidate load balancing paths based on the equipment load distribution thermodynamic diagram and the topological connection state, wherein each candidate load balancing path comprises a target equipment set with resources being redistributed and data transmission delay parameters; according to the data transmission delay parameter and the overload risk level, selecting an optimal path from the candidate load balancing paths as a dynamic load balancing path; The generating a plurality of candidate load balancing paths based on the device load distribution thermodynamic diagram and the topological connection state comprises the following steps: Identifying a standby device node having redundant resource capacity in the network topology connection relationship; Generating a standby resource calling link according to the resource residual capacity of the standby equipment node and the physical distance between the standby equipment node and the key load equipment; combining the standby resource calling link with the current resource calling path to form a candidate load balancing path comprising N-level resource allocation levels, wherein the N-level resource allocation levels correspond to load migration schemes with different response speeds; The feedback of the dynamic allocation strategy of the power resource to the power network control system to activate the authority configuration updating operation comprises the following steps: Receiving a dynamic load balancing path and a right priority adjustment instruction through a strategy analysis interface of the power network control system; generating a path configuration acknowledgement signal based on the dynamic load balancing path and transmitting the path configuration acknowledgement signal to a target power device controller to activate a link switching operation; capturing a device response state code when the link switching operation is completed, and synchronously checking the device response state code and the permission priority adjustment instruction; Triggering the updating process of the user role access control list according to the verified equipment response status code, and generating a user permission configuration table containing a new permission range; Writing the user authority configuration table into an access authorization database of the power network control system, and updating priority ordering parameters of the equipment resource response rule; acquiring a writing completion identification of the access authorization database in real time, and activating a full network authority policy validation instruction based on the writing completion identification; Driving an electric power network control system to execute authority configuration synchronous broadcasting through the whole network authority strategy validation instruction, and generating an authority update broadcasting confirmation queue; and after all the power equipment in the authority update broadcast confirmation queue returns confirmation response, generating an authority configuration update completion mark and storing the mark in a system log.
2. The method of claim 1, wherein the performing device association resolution on the network topology connection relationship to generate a device permission dependency graph comprises: analyzing the device hierarchy structure in the network topology connection relationship, and determining a resource calling path between the main control device and the controlled device; Calculating a resource competition coefficient between the main control equipment and the controlled equipment according to the resource occupancy rate data in the real-time equipment state data, wherein the resource competition coefficient reflects the conflict probability when the main control equipment allocates resources for the controlled equipment; Generating equipment node weights and edge connection rules corresponding to the equipment authority dependency graphs based on the resource calling paths and the resource competition coefficients, wherein the equipment node weights are used for identifying the priority of the power equipment in resource allocation, and the edge connection rules are used for restricting the maximum resource threshold value of authority sharing among the equipment; and generating a device authority dependency graph based on the device node weights and the edge connection rules.
3. The method of claim 1, further comprising monitoring the updated power network operating state for the rights configuration in real time: periodically polling an equipment resource occupancy rate acquisition interface of the power network control system, and extracting an updated real-time resource occupancy rate data set; Synchronously scanning a storage partition of the user operation behavior log to obtain a user operation response log record after authority configuration updating; comparing the real-time resource occupancy rate data set with a reference resource allocation template of an expected load balancing path item by item to generate a device-level resource deviation index set; Synchronously verifying the validity of the user operation response log record and the user authority configuration table to generate a user operation violation event list; aggregating the equipment-level resource deviation index set and the user operation violation event list to generate a global resource allocation abnormality detection report; Generating a permission conflict alarm trigger signal when a device-level resource deviation index exceeding a preset threshold or a non-empty user operation violation event exists in the global resource allocation abnormality detection report; invoking a restarting interface of dynamic authority modeling processing according to the authority conflict alarm triggering signal, and injecting a latest equipment state snapshot and a user operation record into the power network operation data set; After the injection is completed, a regeneration process of the user authority dynamic evolution spectrum and the device authority dependency spectrum is activated to trigger new allocation strategy calculation.
4. The method of claim 1, wherein the method further comprises a training process of a target dynamic rights model: Acquiring a historical power network operation data set and corresponding optimization strategy execution effect data; Performing authority characteristic extraction and conflict event labeling on the historical power network operation data set to generate a model training sample set; training an initial dynamic authority model by adopting a reinforcement learning algorithm to obtain a target dynamic authority model, wherein the target dynamic authority model outputs a user authority characteristic set and a device authority characteristic set which are matched with the execution effect data of the optimization strategy according to the input power network operation data set; And deploying the target dynamic authority model to a power network control system to realize real-time authority modeling and resource allocation strategy generation.
5. An electrical network data driven optimisation system comprising a processor and a memory, wherein the memory stores a computer program which when executed by the processor causes the processor to perform the steps of the method of any one of claims 1 to 4.

Description

Power network data driving optimization method and system based on dynamic authority modeling Technical Field The application belongs to the technical field of data analysis, and particularly relates to a power network data driving optimization method and system based on dynamic authority modeling. Background Along with the expansion of the scale of the power network and the wide access of distributed energy sources, the real-time identification and accurate regulation of the running state of the power system become core requirements for guaranteeing the safe and stable running of the power grid. The traditional power network state identification technology mainly relies on equipment sensor data acquisition and static topology analysis, and abnormality detection and resource scheduling are realized through threshold alarming and preset rules. However, in the environment of a novel power system, the interaction complexity of a user side is increased, the dynamic access of equipment is frequent, and the real-time association analysis and dynamic decision-making requirements of multidimensional heterogeneous data are difficult to effectively respond to in the prior art. The prior art has the following limitations that firstly, the authority management mechanism and the resource scheduling logic are mutually cut, the user operation authority is usually set based on a fixed role template, and cannot be dynamically adjusted according to the real-time network state (such as load fluctuation and equipment failure), so that authority overload or resource allocation conflict frequently occurs. Secondly, the existing resource allocation strategy is mostly dependent on offline simulation or historical experience configuration, lacks collaborative modeling capability for a user behavior mode and a device running state, and is difficult to quickly generate an optimal scheduling path when a topological structure is dynamically changed. Thirdly, authority configuration update is delayed from power grid running state change, and a traditional batch update mechanism causes overlong policy validation period and cannot meet dual requirements of safety protection and resource optimization in a high-real-time scene. Therefore, the prior art is difficult to realize the dynamic adaptation of the user permission, the equipment resource and the network running state, and particularly when dealing with sudden load migration or security threat, the problems of policy stiffness and response delay are easy to occur. Therefore, a technical solution capable of improving the adaptability and the risk resistance of the power network in a complex operation environment is needed. Disclosure of Invention The application provides a power network data driving optimization method and system based on dynamic authority modeling, which are used for improving the self-adaptability and the risk resistance of a power network in a complex operation environment. In a first aspect, an embodiment of the present application provides a power network data driving optimization method based on dynamic authority modeling, which is applied to a power network data driving optimization system, and the method includes acquiring a power network operation data set, where the power network operation data set includes real-time equipment state data, a user operation behavior log, and a network topology connection relationship; the method comprises the steps of carrying out dynamic authority modeling processing on an electric power network operation data set to generate a user authority characteristic set and a device authority characteristic set, wherein the user authority characteristic set represents access control parameters of different user roles to electric power resources, the device authority characteristic set represents resource allocation authorities of different electric power devices in an operation environment, generating an electric power resource dynamic allocation strategy according to the user authority characteristic set and the device authority characteristic set, wherein the electric power resource dynamic allocation strategy is used for adjusting a load balancing path and device control priority in the electric power network, and feeding back the electric power resource dynamic allocation strategy to an electric power network control system to activate authority configuration updating operation, and the authority configuration updating operation comprises adjustment of a user operation authority range and resource response rules of the electric power devices. In a second aspect, an embodiment of the present application provides an electrical network data driven optimization system comprising a processor and a memory, wherein the memory stores a computer program which, when executed by the processor, causes the processor to perform the steps of the above method. In a third aspect, embodiments of the present application provide a computer readable st