Rule-based and data-driven active power distribution network dynamic optimization method and system

Abstract

The application discloses a kind of active power distribution network dynamic optimization method and system based on rule and data driving, belongs to power distribution network optimization technical field, comprising: obtaining active power distribution network data and pre-processing;Using hierarchical database architecture, data is classified and stored, integrity detection and abnormal screening and access permission management;Determine rule and be classified and stored to obtain rule base, optimize rule base based on multi-objective optimization algorithm;Data and rule are mutually optimized, and the optimization result of rule base and database is fed back to data acquisition acquisition step;System output optimization result, for self-growing type trainer, defense evaluation system etc. are used, and return feedback result to previous module.The method of the application can realize the efficient integration storage and optimization of data and the storage and adaptive optimization of rule, improve the security, stability and economy of active power distribution network in complex dynamic change, provide support for the optimized operation of smart grid.

Description

Technical Field

[0001] This invention relates to the field of distribution network optimization technology, and in particular to a rule- and data-driven method and system for dynamic optimization of active distribution networks. Background Technology

[0002] In the operation of active distribution networks, data from grid equipment and distributed energy sources comes from diverse sources, with significant differences in data formats and standards, and high real-time requirements, leading to low efficiency in data integration, optimization, and rule adjustment. Traditional active distribution network management methods mainly rely on static rules and experience-based judgments, lacking flexible adaptive adjustment capabilities and struggling to effectively cope with complex and ever-changing operating environments. Especially under conditions of load fluctuations and distributed energy integration, the safety, stability, and economy of the grid are difficult to guarantee. Existing active distribution network database management methods fail to provide refined classification and dynamic adjustment for different types of data, resulting in low data storage and retrieval efficiency, lagging rule base updates, and difficulty adapting to rapid changes in grid operation. The optimization processes of data and rules are disconnected, leading to low optimization efficiency and accuracy in active distribution networks. Summary of the Invention

[0003] The purpose of this application is to overcome the shortcomings of the prior art and provide a rule-based and data-driven method and system for dynamic optimization of active power distribution networks.

[0004] Firstly, this application provides a rule- and data-driven dynamic optimization method for active distribution networks, comprising the following steps: Data from the active power distribution network is acquired and preprocessed, including historical and real-time data. The database is designed using a hierarchical architecture. The preprocessed data undergoes integrity checks and anomaly filtering. The data after integrity checks and anomaly filtering is then categorized and stored to obtain the database. The rules for the active power distribution network are determined, and the rules are classified and stored in a hierarchical structure to obtain a rule base. The rule base is then optimized using a multi-objective optimization algorithm. The database sends real-time data to the rule base, optimizes and manages the rule base based on the real-time data to obtain an optimized rule base, provides optimization suggestions to the database based on the optimized rule base to obtain an optimized database, and adjusts the strategy for obtaining data from active power distribution network equipment based on the optimized database and the optimized rule base, and performs iterative updates. The system outputs optimized data from the database and optimized rules from the rule base for use by the self-growing trainer, the self-growing defense capability assessment system, and the active power distribution network scheduling decision-making. The analysis results of the self-growing trainer and the self-growing defense capability assessment system are fed back to the database and the rule base to guide their optimization and achieve optimization of the active power distribution network.

[0005] Optionally, the acquisition of data from the active power distribution network, and the preprocessing of the data, including historical data and real-time data, include: Acquire historical and real-time data of the active power distribution network; The real-time data and the historical data are formatted and standardized. The real-time data and the historical data are timestamped together; Data cleaning is performed on the real-time data and the historical data.

[0006] Optionally, the database is designed using a hierarchical architecture, performing integrity checks and anomaly filtering on the preprocessed data, and classifying and storing the data after integrity checks and anomaly filtering to obtain the database, which includes: The database is designed using a layered architecture; The preprocessed data is then classified. The preprocessed data undergoes integrity checks and anomaly screening. The data that has undergone integrity checks and anomaly screening is then stored in the corresponding database according to its category. Set database access permissions based on a role-based access control mechanism, and log all database access operations. The database is optimized based on a self-growing defense strategy.

[0007] Optionally, the database includes: a cache database, a distributed database, and a regular database.

[0008] Optionally, the preprocessed data is classified into: high-priority data, low-priority data, sensitive data, and non-sensitive data.

[0009] Optionally, the rules for determining the active distribution network are classified and stored in a hierarchical structure to obtain a rule base, and the rule base is optimized using a multi-objective optimization algorithm, including: Determine the rule categories for active power distribution networks; The rules are classified and stored using a hierarchical structure to obtain a hierarchical rule base. The rule base is optimized using a multi-objective optimization algorithm; A self-growing trainer is used to analyze historical data, and a self-growing defense capability assessment system is used to simulate the operation status of the active distribution network to obtain assessment results. Based on real-time data, the analysis results of historical data by the self-growing trainer, and the assessment results of the operation status of the active distribution network by the self-growing defense capability assessment system, the rule base is optimized and adjusted.

[0010] Optionally, the rule categories of the active distribution network include: stability rules, security rules, and economic rules.

[0011] Optionally, the multi-objective optimization algorithm The expression is: in, The objective function of the stability rule. The objective function of the security rules is... Let the objective function of the economic rules be... These are the weight coefficients of the objective function for the stability rule. These are the weight coefficients of the objective function for security rules. These are the weighting coefficients of the objective function of the economic rules. i The stability rule represents the first i Each secondary indicator j Indicating the security rule j Each secondary indicator k The first rule in the economic rules k Each secondary indicator x To optimize variables.

[0012] Optionally, the database transmits real-time data to a rule base, optimizes and manages the rule base based on the real-time data to obtain an optimized rule base, provides optimization suggestions to the database based on the optimized rule base to obtain an optimized database, and adjusts the strategy for acquiring active power distribution network equipment data based on the optimized database and the optimized rule base, performing iterative updates, including: The rule base is optimized based on real-time data in the database to obtain the optimized rule base; The database is then optimized based on the optimized rule base to obtain the optimized database. Adjust the strategy for acquiring data from active power distribution network equipment based on the optimized database and rule base; The evaluation results of the self-growing defense capability assessment system are fed back to the database construction and optimization steps and the rule base construction and optimization steps to achieve iterative updates.

[0013] Secondly, this application also provides a rule- and data-driven active power distribution network dynamic optimization system, comprising: a method for executing the rule- and data-driven active power distribution network dynamic optimization method as described in any of the first aspects, comprising: a data acquisition and preprocessing module, a database management and optimization module, a rule base management and optimization module, a rule and data interaction iterative update module, and a system output and feedback module, wherein the data acquisition and preprocessing module is used to acquire real-time data and historical data of active power distribution network equipment, preprocess the real-time data and historical data, receive feedback from the rule and data interaction iterative update module, and optimize the data acquisition strategy; The database management and optimization module is used to store, classify, manage, and optimize the real-time and historical data in layers, perform integrity checks and anomaly screening on the stored data, set access permissions, and dynamically adjust data priority and storage strategy based on the evaluation results of the self-growing defense capability assessment system. The rule base management and optimization module is used to determine the rules of the active distribution network, optimize the rules through a multi-objective optimization algorithm, and optimize and update the rules based on real-time data and the evaluation results of the self-growing defense capability assessment system. The rule and data interaction iterative update module is used to adjust the data and rules. The optimized database provides a basis for the dynamic adjustment of the rule base, and the optimized rule base provides a basis for the filtering and optimization of the database. The optimized rule base and database are fed back to the data acquisition and preprocessing module, the database management and optimization module, and the rule base management and optimization module to realize the iterative update of the entire system. The system output and feedback module is used to output optimized data and optimized rules for use by the self-growing trainer, the self-growing defense capability assessment system, and the active distribution network scheduling decision-making. It also feeds back the analysis results of the self-growing trainer and the self-growing defense capability assessment system to the rule and data interaction and iterative update module.

[0014] This application provides a rule- and data-driven dynamic optimization method and system for active distribution networks. It acquires and preprocesses data from the active distribution network to ensure data consistency and accuracy, laying the foundation for subsequent analysis. A hierarchical and categorized storage strategy improves data access efficiency and ensures rapid response to critical data. Adaptive rule priority adjustment and multi-objective optimization algorithms enable the active distribution network to adjust its operating strategies in real time under complex environments, reducing the impact of load fluctuations and enhancing its adaptability to abnormal operating conditions. An interactive iterative update mechanism between rules and data allows for database optimization of the rule base, and this optimization, in turn, guides database filtering and storage. The optimization results of the database and rule base can feed back into the data acquisition strategy; and through the feedback mechanism, iterative updates are achieved, improving the adaptability and dynamic optimization capabilities of the active distribution network; combined with the self-growing trainer and self-growing defense assessment system, the database and rule base are continuously optimized, improving the adaptive adjustment capability of the active distribution network; through closed-loop optimization among multiple modules, adaptive optimization of the data acquisition strategy, database, and rule base, as well as overall system collaborative optimization, are achieved, enabling efficient management of active distribution network operation data; through dynamic data management, rule adjustment, and closed-loop optimization mechanisms, the efficient use of key data and key rules is ensured, providing technical support for the optimized operation of the future smart grid.

[0015] To make the above-mentioned features and advantages of the invention more apparent and understandable, specific embodiments are described below, and detailed descriptions are provided in conjunction with the accompanying drawings. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a flowchart of a rule-based and data-driven active power distribution network dynamic optimization method provided in one embodiment of this application.

[0018] Figure 2 This is a flowchart of step S1 in a rule-based and data-driven active distribution network dynamic optimization method provided in one embodiment of this application.

[0019] Figure 3 This is a flowchart of step S2 in a rule- and data-driven active distribution network dynamic optimization method provided in one embodiment of this application.

[0020] Figure 4This is a flowchart of step S3 in a rule- and data-driven active distribution network dynamic optimization method provided in one embodiment of this application.

[0021] Figure 5 This is a flowchart of step S4 in a rule- and data-driven active distribution network dynamic optimization method provided in one embodiment of this application.

[0022] Figure 6 This is a schematic diagram of the structure of a rule-based and data-driven active power distribution network dynamic optimization system provided in another embodiment of this application. Detailed Implementation

[0023] To make the objectives and technical solutions of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the described embodiments of this application without creative effort are within the scope of protection of this application.

[0024] In one embodiment, see Figure 1 This application provides a rule-based and data-driven dynamic optimization method for active distribution networks, which may include the following steps: steps S1 to S5.

[0025] Step S1: Obtain data from the active power distribution network and preprocess the data, which includes historical data and real-time data.

[0026] Step S2: Design the database using a hierarchical architecture, perform integrity checks and anomaly filtering on the preprocessed data, and classify and store the data after integrity checks and anomaly filtering to obtain the database.

[0027] Step S3: Determine the rules of the active power distribution network, classify and store the rules using a hierarchical structure to obtain a rule base, and optimize the rule base using a multi-objective optimization algorithm.

[0028] Step S4: The database sends real-time data to the rule base, optimizes and manages the rule base based on the real-time data to obtain an optimized rule base, provides optimization suggestions to the database based on the optimized rule base to obtain an optimized database, and adjusts the strategy for obtaining active power distribution network equipment data in step S1 based on the optimized database and the optimized rule base, and performs iterative updates.

[0029] Step S5: Output the optimized data in the database and the optimized rules in the rule base for use by the self-growing trainer, the self-growing defense capability assessment system, and the active distribution network scheduling decision-making. Feed back the analysis results of the self-growing trainer and the self-growing defense capability assessment system to the database and the rule base to guide the optimization of the database and the rule base, thereby achieving the optimization of the active distribution network.

[0030] The rule- and data-driven dynamic optimization method for active distribution networks in this application constructs an efficient dynamic optimization system for active distribution networks through data classification and optimization, rule base management and optimization, interactive iterative updates of data and rules, and adaptive feedback mechanisms. This system can ensure that the power grid maintains safe, stable, and economical operation under conditions such as load fluctuations and distributed energy access.

[0031] In step S1, please refer to Figure 1 Step S1 involves acquiring data from the active power distribution network and preprocessing the data, which includes historical and real-time data. Specifically, this includes acquiring historical and real-time data from the active power distribution network; standardizing the format of the real-time and historical data; unifying the timestamps of the real-time and historical data; and cleaning the real-time and historical data.

[0032] For example, please refer to Figure 2 Step S1 may include the following steps: Step S11 to Step S12.

[0033] Step S11: Obtain data from the active power distribution network.

[0034] Step S12: Preprocess the data.

[0035] As an example, in step S11, during the operation of the active power distribution network, real-time data of the active power distribution network can be obtained through intelligent sensing units, data acquisition terminals, control centers, etc., of the equipment in the active power distribution network. The real-time data may include: operating data, equipment status data, control command data, system decision result data, and external environment data.

[0036] As an example, the real-time data includes, but is not limited to, voltage, current, frequency, power, power flow distribution constraints, load conditions, switch status, and output data of new energy wind power and photovoltaic power.

[0037] As an example, historical databases of devices in an active power distribution network can be accessed through API interfaces to obtain historical data of devices in the active power distribution network. Real-time data and historical data can be compared by feature to ensure data consistency.

[0038] As an example, the feature comparison between real-time data and historical data may include: extracting features of key indicators such as voltage, current, active power, and frequency regarding mean, extreme values, and standard deviation, and using a sliding window method to identify historical trends; comparing whether the changes in real-time data conform to historical distribution patterns and threshold ranges to determine whether the data is abnormal, thereby assisting in subsequent rule adjustments and data filtering.

[0039] As an example, in step S12, the real-time data and the historical data are preprocessed to unify the format and standardize the units, which facilitates data fusion and analysis.

[0040] As an example, the preprocessing may include: format standardization, timestamp unification, and data cleaning. Format standardization refers to the unified conversion of heterogeneous formats from different data sources into a standardized format to facilitate subsequent unified storage and management of the data. Timestamp unification converts timestamps of different precisions and time zones into a unified standard format and time zone, facilitating the sorting, analysis, and correlation of data according to time sequence. Data cleaning aims to identify and remove various noise and erroneous data in the original data, such as duplicate records, missing values, and outliers.

[0041] As an example, the industry-recognized CIM format standard can be adopted and the data units can be unified; GPS timing technology can be used to unify timestamps; the three-standard-deviation method can be used to remove outliers; and Kalman filtering can be used to remove noise.

[0042] In step S2, please refer to Figure 1 Step S2 involves designing the database using a layered architecture. Integrity checks and anomaly filtering are performed on the preprocessed data. The data after integrity checks and anomaly filtering is then categorized and stored to obtain the database. Specifically, a layered architecture is used to design the database; the preprocessed data is categorized; integrity checks and anomaly filtering are performed on the preprocessed data, and the data after integrity checks and anomaly filtering is stored in the corresponding database according to its category; database access permissions are set based on a role-based access control mechanism, and all database access operations are logged; the database is optimized based on a self-growing defense strategy to obtain the final database.

[0043] For example, please refer to Figure 3 Step S2 may include the following steps: Step S21 to Step S24.

[0044] Step S21: Database structure design and data storage management.

[0045] Step S22: Data integrity check and anomaly screening.

[0046] Step S23: Database access permission management.

[0047] Step S24: Optimize based on self-growing defense strategy.

[0048] As an example, in step S21, the preprocessed data is classified according to its real-time requirements, business importance, and security sensitivity. Specific categories may include: high-priority data, low-priority data, sensitive data, and non-sensitive data.

[0049] As an example, the high-priority data may include: real-time load, power grid operating status, etc.

[0050] As an example, the low-priority data may include: historical operational data, long-term statistical data, etc.

[0051] As an example, the sensitive data may include data related to user privacy, power grid security, etc. The sensitive data may be stored encrypted and subject to access control.

[0052] As an example, non-sensitive data may include: meteorological information, publicly available power grid data, etc.

[0053] Furthermore, a hierarchical architecture is adopted to classify and store the categorized data, resulting in a hierarchical database, with each layer storing different categories of data.

[0054] As an example, the layered architecture database may include: a cache database, a distributed database, and a regular database.

[0055] Furthermore, the preprocessed data is stored in the corresponding hierarchical database to achieve data storage.

[0056] As an example, the preprocessed data stored in the database may include: real-time operating data, equipment status data, control command data, system decision result data, external environment data, and historical data.

[0057] As an example, a cache database can be used to store high-priority data; a distributed database can be used to store low-priority data; and a regular database can be used to store non-sensitive data.

[0058] As an example, in step S22, before the data is stored in the database, a hash algorithm is used to calculate the corresponding hash value of the categorized data. This facilitates hash verification during subsequent data retrieval, ensuring data integrity and preventing unauthorized modification during storage or transmission. When data is retrieved, the hash value is recalculated and compared with the hash value from when the data was previously stored. If the two hash values ​​are the same, it indicates that the data has not been tampered with or damaged during storage and retrieval; if the two hash values ​​are different, it indicates that a problem may have occurred during storage and retrieval, requiring further investigation.

[0059] Furthermore, based on predefined security, stability, and economic rules, anomaly detection is performed on historical data in the database to identify problems such as data input errors, loss, or inconsistencies, and the data is marked or removed according to the severity of the anomaly.

[0060] As an example, marked abnormal data can be smoothed or manually reviewed to correct the anomalies, thereby completing the integrity check and anomaly screening of historical data and maintaining data integrity and accuracy.

[0061] As an example, in step S23, a Role-Based Access Control (RBAC) mechanism is used to assign different data access permissions to different roles. For critical and sensitive data, a multi-factor authentication mechanism based on digital certificates and dynamic passwords can be used to ensure security.

[0062] As an example, the different roles may include: dispatcher, administrator, and security monitoring system, etc.

[0063] As an example, all database access operations can be logged, and abnormal access can be detected by using methods such as time series analysis, access frequency statistics, and user behavior pattern matching to identify access requests with abnormal times, frequencies, or sources, thereby preventing potential security risks.

[0064] As an example, in step S24, based on the self-growing defense strategy, the operating status and key characteristics of the active power distribution network are obtained by analyzing historical data through the self-growing trainer, and data that may become important in the future are predicted, resulting in historical trend analysis, abnormal data distribution and prediction results, thereby raising the priority in advance. Specifically, the self-growing defense capability assessment system, based on the self-growing trainer, combines the operating data and attack logs collected by the Supervisory Control and Data Acquisition (SCADA) system, extracts key features after preprocessing, simulates typical attack behaviors and constructs state vectors, optimizes the defense strategy through deep reinforcement learning, and further introduces a transfer learning mechanism to achieve dynamic identification and risk assessment of different operating states and potential threats, thereby forming an adaptive defense strategy.

[0065] Furthermore, a self-growing defense capability assessment system is used to simulate the operating status of the active distribution network or analyze historical data to assess the security situation of the active distribution network, adjust defense strategies, and adjust the data storage structure and content according to the adjusted defense strategies. Specifically, the self-growing defense capability assessment system uses machine learning models to construct a power grid situation awareness and risk assessment model based on historical operating data, current state parameters, and the execution effect of defense strategies, so as to identify and assess potential risks and abnormal patterns.

[0066] As an example, in situations such as fluctuations in new energy sources, changes in load, or an increase in potential attack risks, the self-growing defense capability assessment system will feed the assessment results back to the database and adjust the classification and priority of the preprocessed data based on the feedback results.

[0067] As an example, under high load or abnormal operating conditions, an intelligent scheduling mechanism can be used to dynamically increase the sampling frequency of critical or related data and adjust storage priorities to support real-time defense requirements.

[0068] In step S3, please refer to Figure 1 Step S3 involves determining the rules for the active power distribution network, classifying and storing the rules using a hierarchical structure to obtain a rule base, and then optimizing the rule base using a multi-objective optimization algorithm. Specifically, this involves determining the rule categories for the active power distribution network; classifying and storing the rules using a hierarchical structure to obtain a layered rule base; optimizing the rule base using a multi-objective optimization algorithm; analyzing historical data using a self-growing trainer; simulating the operational status of the active power distribution network using a self-growing defense capability assessment system to obtain evaluation results; and optimizing and adjusting the rule base based on real-time data, the analysis results of historical data by the self-growing trainer, and the evaluation results of the active power distribution network's operational status by the self-growing defense capability assessment system.

[0069] For example, please refer to Figure 4 Step S3 may include the following steps: Step S31 to Step S34.

[0070] Step S31: Determine the rules and classify them.

[0071] Step S32: Standardize the storage and management of rules to obtain a rule base.

[0072] Step S33: Optimize the rule base using a multi-objective optimization algorithm.

[0073] Step S34: Optimize and adjust the rule base based on the self-growing defense strategy.

[0074] As an example, in step S31, the primary indicators of the rules are defined based on factors such as the parameters and operating environment of the active distribution network. The parameters of the active distribution network are then classified according to the primary indicators of the rules to determine the secondary indicators of the rules.

[0075] As an example, primary indicators of rules can be set based on power grid operation safety standards, national power industry regulations, and engineering experience knowledge, so that the defined rules can adapt to actual operation needs.

[0076] As an example, the primary indicators of a rule can include: stability rules, safety rules, and economic rules. Stability rules ensure that the output state of an active distribution network remains within a reasonable range under internal and external disturbances, maintaining normal operation; safety rules reduce risks and are crucial for ensuring the safety of the active distribution network and related personnel and equipment; economic rules aim to optimize resource utilization, reduce costs, improve output efficiency, and enhance competitiveness and sustainability.

[0077] As an example, secondary indicators of stability rules may include voltage stability, frequency stability, etc.

[0078] As an example, secondary indicators of safety rules may include: equipment capacity safety, load limitation, current overload protection, voltage over / under protection, and grid safety margin.

[0079] As an example, secondary indicators of economic rules could include: minimizing operating costs, optimizing energy dispatch, and reducing network losses.

[0080] As an example, in step S32, a hierarchical structure is used to standardize the storage of rules, resulting in a hierarchical rule base, with each layer of the rule base used to store rules of different categories.

[0081] As an example, the primary indicators of the rules can be used as framework rules, and the secondary indicators can be used as specific implementation details, and can be dynamically adjusted according to the real-time status of the active distribution network.

[0082] As an example, different management methods can be set according to the nature of the rules. Specifically, for rules that remain unchanged or relatively fixed in the long term, such as equipment capacity and load limits, static storage can be used; for rules that are affected by real-time data and environmental changes, such as frequency stability and load fluctuations, a dynamic update mechanism can be used, which can adjust the rules in a timely manner according to real-time data and the power grid operating status.

[0083] As an example, in step S33, a multi-objective optimization algorithm is used to optimize the rule base. The optimization objectives may include various secondary indicators from stability rules, security rules, and economic rules. Multi-objective optimization algorithm The expression is: in, Let the objective function be the stability rule. The objective function of the security rules is... Let the objective function of the economic rules be... These are the weight coefficients of the objective function for the stability rule. These are the weight coefficients of the objective function for security rules. These are the weighting coefficients of the objective function of the economic rules. i The stability rule represents the first i Each secondary indicator j Indicating the security rule j Each secondary indicator k The first rule in the economic rules k Each secondary indicator x To optimize variables.

[0084] As an example, the objective function of the stability rule The expression is: in, Indicates the first The actual values ​​of each stability parameter; Indicates the first Reference values ​​for each stability parameter; objective function of the stability rule. The goal is to minimize the degree of deviation in the system.

[0085] As an example, the objective function of security rules The expression is: in, Indicates the first The goal is to increase the success rate of detecting or defending against security incidents such as attacks and anomalies, and to maximize the success rate of security incident response.

[0086] As an example, the objective function of economic rules The expression is: in, Indicates the first The economic indicators include dispatch costs, power losses, load shifting costs, etc., with the goal of minimizing total costs.

[0087] As an example, in step S34, a self-growing trainer is used to analyze historical data, and a self-growing defense capability assessment system is used to simulate the operating status of the active distribution network to obtain assessment results. Based on real-time data, the analysis results of historical data by the self-growing trainer, and the assessment results of the active distribution network's operating status by the self-growing defense capability assessment system, optimization and adjustments are made to dynamically optimize and adjust the priority, execution conditions, and thresholds of the rule base. The priority can be the response order of each rule under conflict or resource constraints; the execution conditions are the prerequisites for triggering the operation of each rule, such as abnormal voltage, sudden current changes, or load exceeding a set proportion; the threshold can be the boundary value used to determine whether the execution conditions are met, for example, the voltage deviation threshold can be set to ±10%, and the line load rate can be 80%, etc.

[0088] As an example, the initial values ​​of execution conditions and thresholds can be set based on industry safety standards, power grid operation procedures and historical operation experience. For specific details, refer to the distribution network operation technical procedures and the setting of empirical parameters in typical scenarios.

[0089] Specifically, the priority of each rule in the rule base can be automatically adjusted based on the operating status and assessment results of the active power distribution network to ensure that the business or control corresponding to the key rules is executed with priority. For example, when the self-growing defense capability assessment system detects frequent voltage fluctuations, load anomalies, or potential attack behaviors in a certain area, it can increase the priority of rules related to voltage regulation, load reduction, and protection response in that area, while decreasing the priority of routine maintenance rules that do not currently involve risk areas.

[0090] As an example, the triggering conditions of rules can be optimized based on load fluctuations and equipment status changes in the active distribution network to ensure adaptation to the dynamic environment of multi-objective optimization in the active distribution network. For instance, when a rapid increase in load is detected, the triggering threshold of load control rules can be appropriately lowered to allow them to intervene and execute earlier; while when equipment enters a warning or fault mode, the triggering conditions of equipment protection or switching rules, such as thresholds for current, voltage, and frequency, can be dynamically adjusted to enhance the timeliness and accuracy of the response.

[0091] As an example, when an active distribution network experiences load fluctuations or equipment failures, the rule thresholds are adjusted based on feedback from the self-evolving defense capability assessment system to better reflect the current operating conditions. Specifically, the self-evolving defense capability assessment system identifies the changing trends of key indicators such as voltage, current, frequency, and load factor under the current operating conditions based on operational simulation results, compares and analyzes these trends with historical abnormal thresholds, and dynamically corrects the threshold settings for the corresponding parameters of each rule. For example, the starting current threshold for overcurrent protection can be adjusted to the upper limit of the safe range in recent assessments.

[0092] In step S4, please refer to Figure 1 In step S4, the database sends real-time data to the rule base. Based on the real-time data, the rule base is optimized and managed to obtain an optimized rule base. Optimization suggestions are then provided to the database based on the optimized rule base, resulting in an optimized database. The strategy for acquiring active power distribution network equipment data in step S1 is adjusted based on the optimized database and the optimized rule base, and iterative updates are performed. Specifically, the rule base is optimized based on real-time data in the database to obtain an optimized rule base; the database is optimized based on the optimized rule base to obtain an optimized database; the strategy for acquiring active power distribution network equipment data in step S1 is adjusted based on the optimized database and the optimized rule base; and the evaluation results of the self-growing defense capability assessment system are fed back to the database construction optimization in step S2 and the rule base construction optimization in step S3, achieving iterative updates.

[0093] For example, please refer to Figure 5 Step S4 may include the following steps: Step S41 to Step S45.

[0094] Step S41: Optimize the database rule base.

[0095] Step S42: Optimize the rule base database.

[0096] Step S43: Data acquisition optimization.

[0097] Step S44: Self-growth feedback.

[0098] As an example, in step S41, real-time data from the database is transmitted to the rule base. The rule base analyzes the operating status of the active distribution network based on the real-time data, dynamically adjusts the operating strategy of the active distribution network, and adjusts the rule base parameters to ensure the stability, security, and economy of the power grid.

[0099] As an example, the rule base can optimize the execution conditions, trigger thresholds, and priorities of rules based on historical trend analysis, abnormal data distribution, and prediction results obtained from self-growing trainers stored in the database.

[0100] As an example, in step S42, optimization suggestions are provided to the database based on the rules in the optimized rule base, and the data storage logic and filtering logic in the database are adjusted to obtain the optimized database.

[0101] As an example, in step S43, adjustment suggestions are provided for the strategy of obtaining active power distribution network equipment data in step S1 based on the optimized database and the optimized rule base, so as to adjust the data acquisition strategy.

[0102] As an example, adjusting the data acquisition strategy includes: dynamically adjusting the data sampling interval, increasing the data acquisition frequency for key scenarios, increasing data monitoring of key equipment and specific indicators in the active power distribution network, and optimizing the data cleaning strategy, so as to ensure that the acquired data can accurately support the determination and optimization of the rule base in step S3.

[0103] As an example, in step S44, the evaluation results of the self-growing defense capability assessment system are fed back to the database construction and optimization in step S1 and the determination and optimization of the rule base in step S3. Through iterative updates, continuous optimization and self-adjustment are achieved.

[0104] In step S5, please refer to Figure 1 In step S5, the optimized data in the database and the optimized rules in the rule base are output for use by the self-growing trainer, the self-growing defense capability assessment system, and the active distribution network scheduling decision. The analysis results of the self-growing trainer and the self-growing defense capability assessment system are fed back to the database and the rule base to guide the optimization of the database and the rule base, thereby achieving the optimization of the active distribution network.

[0105] As an example, the optimized data in the database and the optimized rules in the rule base are output for use by the self-growing trainer, the self-growing defense capability assessment system, and the active power distribution network scheduling decision-making.

[0106] As an example, the self-growing trainer and the self-growing defense capability assessment system can analyze the optimized output data and optimized rules, and feed the analysis results back to step S4 to guide the interaction between the database and the rule base to guide optimization.

[0107] In the rule- and data-driven dynamic optimization method for active distribution networks proposed in this application, preprocessing the acquired historical and real-time data ensures data consistency and accuracy, laying the foundation for subsequent analysis. By performing integrity checks and anomaly screening on the preprocessed data, and employing a hierarchical architecture for categorized storage and access control, data security and effective management are guaranteed, data availability is improved, and efficient access to critical data is achieved, ensuring data accuracy, security, and availability. By defining rules for active distribution network equipment, using a hierarchical storage structure, and optimizing the rule base using a multi-objective optimization algorithm, stability, security, and economy can be comprehensively considered to formulate reasonable rules. The application of multi-objective optimization algorithms and self-growing defense strategies balances stability, security, and economy, enabling rules to adapt to dynamic changes in the power grid and ensuring efficient, stable, and economical operation of the power grid under complex operating conditions. Through an interactive iterative update mechanism between data and rules, the rule base can be optimized based on real-time data, and then the database can be optimized based on the rule base, thereby adjusting the data acquisition strategy and forming a closed-loop optimization system. The method proposed in this application can effectively improve the efficiency of data access and rule adjustment, effectively cope with complex situations such as load fluctuations and distributed energy access, ensure the safe and stable operation of active distribution networks under complex conditions, improve resource utilization efficiency, reduce operating costs, enhance the intelligent management level and sustainable development capability of the power grid, and provide support for the optimized operation of smart grids.

[0108] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the accompanying drawings may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0109] In another embodiment, please refer to Figure 6This application also provides a rule- and data-driven active power distribution network dynamic optimization system, which may include: a data acquisition and preprocessing module 10, a database management and optimization module 20, a rule base management and optimization module 30, a rule and data interaction iterative update module 40, and a system output and feedback module 50. The data acquisition and preprocessing module 10 is connected to the database management and optimization module 20, the rule base management and optimization module 30, and the rule and data interaction iterative update module 40; the rule and data interaction iterative update module 40 is bidirectionally connected to the database management and optimization module 20, the rule base management and optimization module 30, and the system output and feedback module 50.

[0110] As an example, the data acquisition and preprocessing module 10 acquires real-time and historical data from active power distribution network equipment, preprocesses the real-time and historical data, receives feedback from the rule and data interaction iterative update module, and optimizes the data acquisition strategy.

[0111] As an example, the database management and optimization module 20 is used to store, classify, manage, and optimize the real-time and historical data in layers, perform integrity checks and anomaly screening on the stored data, set access permissions, and dynamically adjust data priority and storage strategy based on the evaluation results of the self-growing defense capability assessment system.

[0112] As an example, the rule base management and optimization module 30 is used to determine the rules of the active distribution network, optimize the rules through a multi-objective optimization algorithm, and optimize and update the rules based on real-time data and the evaluation results of the self-growing defense capability assessment system.

[0113] As an example, the rule and data interaction iterative update module 40 is used to adjust the data and rules. The optimized database provides a basis for the dynamic adjustment of the rule base, and the optimized rule base provides a basis for the filtering and optimization of the database. At the same time, the optimized rule base and database are fed back to the data acquisition and preprocessing module 10 to realize the iterative update of the entire system. That is, the interaction optimization results and the output results of the system output and feedback module 50 are continuously fed back to the database management and optimization module 20 and the rule base management and optimization module 30.

[0114] As an example, the system output and feedback module 50 is used to output optimized data and optimized rules for use by the self-growing trainer, the self-growing defense capability assessment system, and the active distribution network scheduling decision-making, and to feed back the analysis results of the self-growing trainer and the self-growing defense capability assessment system to the rule and data interaction and iterative update module 40.

[0115] In the aforementioned rule- and data-driven active distribution network dynamic optimization system, the data acquisition and preprocessing module 10 acquires and preprocesses data, and optimizes the data acquisition strategy through feedback from other modules, ensuring the accuracy and timeliness of the input data. The database management and optimization module 20 stores data hierarchically and categorizes it, checks its integrity, filters abnormal data, sets permissions, and dynamically adjusts strategies based on evaluations, achieving efficient data management and security. The rule base management and optimization module 30 determines and optimizes rules, updating them based on real-time data and evaluation results, allowing rules to adapt to complex changes in the power grid. The rule and data interaction iterative update module 40 optimizes data and rules, driving system iteration. The system output and feedback module 50 outputs optimization results and collects feedback, continuously promoting system improvement. This system can effectively improve the safety, stability, and economy of active distribution networks under complex conditions such as load fluctuations and new energy access, ensuring efficient power grid operation and providing strong technical support for the development of smart grids.

[0116] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0117] Although this application has been disclosed above with reference to embodiments, it is not intended to limit this application. Anyone skilled in the art may make some modifications and refinements without departing from the spirit and scope of this application. Therefore, the scope of protection of this application shall be determined by the appended claims.

Claims

1. A rule- and data-driven dynamic optimization method for active power distribution networks, characterized in that, Includes the following steps: Data from the active power distribution network is acquired and preprocessed, including historical and real-time data. The database is designed using a hierarchical architecture. The preprocessed data undergoes integrity checks and anomaly filtering. The data after integrity checks and anomaly filtering is then categorized and stored to obtain the database. The rules for the active power distribution network are determined, and the rules are classified and stored in a hierarchical structure to obtain a rule base. The rule base is then optimized using a multi-objective optimization algorithm. The database sends real-time data to the rule base, optimizes and manages the rule base based on the real-time data to obtain an optimized rule base, provides optimization suggestions to the database based on the optimized rule base to obtain an optimized database, and adjusts the strategy for obtaining data from active power distribution network equipment based on the optimized database and the optimized rule base, and performs iterative updates. The system outputs optimized data from the database and optimized rules from the rule base for use by the self-growing trainer, the self-growing defense capability assessment system, and the active power distribution network scheduling decision-making. The analysis results of the self-growing trainer and the self-growing defense capability assessment system are fed back to the database and the rule base to guide their optimization and achieve optimization of the active power distribution network.

2. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 1, characterized in that, The process of acquiring data from the active power distribution network and preprocessing the data, which includes historical and real-time data, includes: Acquire historical and real-time data of the active power distribution network; The real-time data and the historical data are formatted and standardized. The real-time data and the historical data are timestamped together; Data cleaning is performed on the real-time data and the historical data.

3. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 1, characterized in that, The database is designed using a hierarchical architecture. It performs integrity checks and anomaly filtering on the preprocessed data, and then categorizes and stores the data after integrity checks and anomaly filtering to obtain the database, which includes: The database is designed using a layered architecture. The preprocessed data is then classified. The preprocessed data undergoes integrity checks and anomaly screening. The data that has undergone integrity checks and anomaly screening is then stored in the corresponding database according to its category. Set database access permissions based on a role-based access control mechanism, and log all database access operations. The database is optimized based on a self-growing defense strategy.

4. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 3, characterized in that, The databases include: cache databases, distributed databases, and ordinary databases.

5. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 3, characterized in that, The preprocessed data is classified into: high-priority data, low-priority data, sensitive data, and non-sensitive data.

6. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 1, characterized in that, The rules for determining the active distribution network are classified and stored in a hierarchical structure to obtain a rule base. A multi-objective optimization algorithm is then used to optimize the rule base, including: Determine the rule categories for active power distribution networks; The rules are classified and stored using a hierarchical structure to obtain a hierarchical rule base. The rule base is optimized using a multi-objective optimization algorithm; A self-growing trainer is used to analyze historical data, and a self-growing defense capability assessment system is used to simulate the operation status of the active distribution network to obtain assessment results. Based on real-time data, the analysis results of historical data by the self-growing trainer, and the assessment results of the operation status of the active distribution network by the self-growing defense capability assessment system, the rule base is optimized and adjusted.

7. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 6, characterized in that, The rule categories of the active distribution network include: stability rules, security rules, and economic rules.

8. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 6, characterized in that, The multi-objective optimization algorithm The expression is: in, The objective function of the stability rule. The objective function of the security rules is... Let the objective function of the economic rules be... These are the weight coefficients of the objective function for the stability rule. These are the weight coefficients of the objective function for security rules. These are the weighting coefficients of the objective function of the economic rules. i The stability rule represents the first i Each secondary indicator j Indicating the security rule j Each secondary indicator k The first rule in the economic rules k Each secondary indicator x To optimize variables.

9. The rule-based and data-driven dynamic optimization method for active distribution networks according to claim 1, characterized in that, The database transmits real-time data to the rule base, optimizes and manages the rule base based on the real-time data to obtain an optimized rule base, provides optimization suggestions to the database based on the optimized rule base, and then adjusts the strategy for acquiring active power distribution network equipment data based on the optimized database and the optimized rule base, performing iterative updates, including: The rule base is optimized based on real-time data in the database to obtain the optimized rule base; The database is then optimized based on the optimized rule base to obtain the optimized database. Adjust the strategy for acquiring data from active power distribution network equipment based on the optimized database and rule base; The evaluation results of the self-growing defense capability assessment system are fed back to the database construction and optimization steps and the rule base construction and optimization steps to achieve iterative updates.

10. A rule- and data-driven dynamic optimization system for active power distribution networks, characterized in that, The method for executing the rule-based and data-driven dynamic optimization method for active distribution networks as described in any one of claims 1 to 9 includes: a data acquisition and preprocessing module, a database management and optimization module, a rule base management and optimization module, a rule and data interaction and iterative update module, and a system output and feedback module, wherein... The data acquisition and preprocessing module is used to acquire real-time and historical data from active power distribution network equipment, preprocess the real-time and historical data, receive feedback from the rule and data interaction iterative update module, and optimize the data acquisition strategy. The database management and optimization module is used to store, classify, manage, and optimize the real-time and historical data in layers, perform integrity checks and anomaly screening on the stored data, set access permissions, and dynamically adjust data priority and storage strategy based on the evaluation results of the self-growing defense capability assessment system. The rule base management and optimization module is used to determine the rules of the active distribution network, optimize the rules through a multi-objective optimization algorithm, and optimize and update the rules based on real-time data and the evaluation results of the self-growing defense capability assessment system. The rule and data interaction iterative update module is used to adjust the data and rules. The optimized database provides a basis for the dynamic adjustment of the rule base, and the optimized rule base provides a basis for the filtering and optimization of the database. The optimized rule base and database are fed back to the data acquisition and preprocessing module, the database management and optimization module, and the rule base management and optimization module to realize the iterative update of the entire system. The system output and feedback module is used to output optimized data and optimized rules for use by the self-growing trainer, the self-growing defense capability assessment system, and the active distribution network scheduling decision-making. It also feeds back the analysis results of the self-growing trainer and the self-growing defense capability assessment system to the rule and data interaction and iterative update module.

Images