Online reconstruction control method and system for secondary equipment of intelligent substation

Abstract

The invention discloses an intelligent substation secondary equipment online reconstruction control method and system, and relates to the technical field of power system automation, and the method comprises the following steps: broadcasting fault information containing digital fingerprints generated by fault secondary equipment to adjacent equipment; enabling the adjacent equipment to verify the digital fingerprint and carry out lightweight consensus confirmation, and obtaining a dynamic reputation score of the secondary equipment based on the operation data of the secondary equipment; screening out a candidate takeover equipment set from the adjacent equipment according to the dynamic reputation score; determining target takeover equipment from the candidate takeover equipment set through a multi-target optimization algorithm; and causing the target takeover device to execute the reconstruction logic. The method is used for solving the problems of insufficient authenticity of fault information of existing secondary equipment, low equipment cooperation efficiency, lack of intelligence of takeover decision and poor cohesion of a reconstruction process.

Description

Technical Field

[0001] This invention relates to the field of power system automation technology, and more specifically, to a method and system for online reconfiguration control of secondary equipment in intelligent substations. Background Technology

[0002] With the rapid development of smart grids and digital substations, secondary equipment, as a core component for protection, monitoring, and control in power systems, directly impacts the safety and stability of the power grid through its operational reliability and intelligence level. In recent years, although the industry has widely adopted technologies such as online monitoring, remote communication, and data analysis to support the operation, maintenance, and reconfiguration of secondary equipment, significant technical bottlenecks remain in areas such as ensuring the authenticity of fault information, confirming collaborative consistency, and making dynamic takeover decisions, hindering further improvements in the intelligence level of the power grid.

[0003] While existing methods for reconfiguring secondary equipment in smart substations can achieve fault detection and equipment takeover to a certain extent, they generally suffer from the following problems: fault information lacks reliable unique identifiers and anti-tampering mechanisms, resulting in insufficient authenticity and traceability; consensus mechanisms between adjacent devices are inefficient, relying on centralized or manual confirmation methods, which are difficult to meet the real-time requirements under complex topologies; the selection mechanism for candidate takeover equipment relies too heavily on static thresholds or single indicators, failing to achieve comprehensive consideration of multiple factors and dynamic adaptation; the selection and verification of target takeover equipment lack systematic optimization and forward-looking testing, which can easily lead to the takeover scheme deviating from the optimal state and even triggering new risks; and the reconfiguration process lacks sufficient continuity in state loading and logic switching, often resulting in incomplete data or control link interruptions, affecting the continuity and stability of power grid operation.

[0004] To address the above problems, this invention proposes an improved solution. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an online reconfiguration control method and system for intelligent substation secondary equipment. By introducing digital fingerprints and a lightweight consensus mechanism, the method ensures the authenticity and consistency of fault information. It also combines multi-objective optimization algorithms, deep reinforcement learning, and digital twin verification to dynamically select and verify target takeover equipment. Finally, it loads the operating status of the faulty equipment to achieve seamless functional switching. This solves the problems of insufficient authenticity of fault information in existing secondary equipment, low equipment coordination efficiency, lack of intelligence in takeover decisions, and poor continuity of the reconfiguration process.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for online reconfiguration control of secondary equipment in an intelligent substation includes the following steps: broadcasting fault information containing digital fingerprints generated by faulty secondary equipment to neighboring equipment; having neighboring equipment verify the digital fingerprints and perform lightweight consensus confirmation, and obtaining a dynamic reputation score for the secondary equipment based on its operating data; selecting a set of candidate takeover equipment from the neighboring equipment based on the dynamic reputation score; determining the target takeover equipment from the set of candidate takeover equipment using a multi-objective optimization algorithm; and having the target takeover equipment execute reconfiguration logic.

[0007] In a preferred embodiment, broadcasting the fault information containing a digital fingerprint generated by the faulty secondary device to adjacent devices specifically involves: collecting the unique identifier and real-time fault code of the faulty secondary device; combining the unique identifier, fault code, and current timestamp, and generating a digital fingerprint using a preset cryptographic hash function; combining the fault code with the digital fingerprint to generate fault information and broadcasting it.

[0008] In a preferred embodiment, the lightweight consensus confirmation specifically involves: after receiving fault information, a neighboring device acts as a proposal node; it encapsulates the verification result of the digital fingerprint and the fault information into a consensus request and broadcasts it to the neighboring devices; the remaining neighboring devices respond to the consensus request, independently verify the digital fingerprint, and provide feedback on the verification result; when the proposal node receives more than a first preset threshold of valid verification results, it determines that consensus has been reached and generates a consensus confirmation signal.

[0009] In a preferred embodiment, obtaining the dynamic reputation score of the secondary equipment based on the secondary equipment's operating data specifically involves: acquiring historical and real-time operating status data of the secondary equipment and performing data preprocessing to obtain first data; applying time-series weighting to the first data based on an adaptively adjusted sliding time window and a forgetting factor to obtain second data; acquiring the statistical features of the second data and constructing a multimodal reputation feature vector by combining it with the equipment's multimodal features; inputting the feature vector into a scoring model based on an attention mechanism to calculate and output the dynamic reputation score, wherein the attention mechanism is used to adaptively assign weights to different features in the multimodal reputation feature vector.

[0010] In a preferred embodiment, the step of selecting a set of candidate takeover devices from neighboring devices based on dynamic reputation scores specifically involves: adding devices with reputation scores higher than a dynamic threshold to the initial set of candidate takeover devices; performing a secondary selection on the initial set of candidate takeover devices based on topology location and load balancing factor; and outputting the set of candidate takeover devices.

[0011] In a preferred embodiment, determining the target takeover device from the candidate takeover device set using a multi-objective optimization algorithm specifically involves: constructing a multi-objective optimization function and dynamically generating multi-objective weight coefficients that match the current power grid state based on a deep reinforcement learning model; inputting the candidate takeover device set as the solution space into a non-dominated sorting algorithm, and using the multi-objective optimization function and its weight coefficients as evaluation criteria, obtaining a Pareto optimal solution set through iterative solving; sorting the Pareto optimal solution set, selecting the candidate device corresponding to the highest-ranked solution as the preliminary target takeover device; performing prospective verification based on digital twins on the preliminary target takeover device, and finally determining it as the target takeover device after successful verification.

[0012] In a preferred embodiment, the deep reinforcement learning model includes: constructing a reinforcement learning model based on the Actor-Critic framework; using real-time power grid state information as the input state of the model; mapping the output actions of the model to the weight coefficients of a multi-objective optimization function; wherein the model is trained using a reward function based on historical takeover success rate and system stability.

[0013] In a preferred embodiment, the specific process of sorting the Pareto optimal solution set includes: using the multi-objective weight coefficients as the evaluation criterion weights of the TOPSIS method; determining the positive and negative ideal values ​​of each optimization objective, and calculating the weighted Euclidean distance between each candidate solution in the Pareto optimal solution set and the positive and negative ideal values ​​in each optimization objective dimension; calculating the relative proximity of each candidate solution based on the weighted Euclidean distance, and sorting them according to the magnitude of the relative proximity to obtain the sorting result.

[0014] In a preferred embodiment, the prospective verification of the initial target takeover equipment based on digital twins specifically involves: inputting the topology parameters and operating status data of the target takeover equipment into a power grid digital twin simulation platform; importing future load forecast data and renewable energy output forecast data into the simulation platform to construct a takeover simulation scenario; performing power grid dynamic power flow calculations under the takeover simulation scenario to obtain node voltage, line power, and power balance data; and generating verification conclusions based on the results of the power grid dynamic power flow calculations.

[0015] Secondly, this application provides an online reconfiguration control system for secondary equipment in an intelligent substation, comprising: a fingerprint broadcasting module for broadcasting fault information containing a digital fingerprint generated by a faulty secondary equipment to adjacent equipment; a verification consensus and scoring module for enabling adjacent equipment to verify the digital fingerprint and perform lightweight consensus confirmation, and to obtain a dynamic reputation score for the secondary equipment based on the secondary equipment's operating data; a candidate screening module for selecting a set of candidate takeover equipment from adjacent equipment based on the dynamic reputation score; an optimization selection module for determining a target takeover equipment from the set of candidate takeover equipment through a multi-objective optimization algorithm; and a reconfiguration execution module for instructing the target takeover equipment to execute reconfiguration logic.

[0016] As can be seen from the above technical solutions, this invention solves the problems of difficulty in ensuring the authenticity of fault information and low verification efficiency in existing systems by using digital fingerprints and lightweight consensus mechanisms to verify fault information; by combining multi-objective optimization algorithms, deep reinforcement learning weight adaptation, and digital twin forward-looking verification, it solves the problems of lack of intelligence in candidate takeover device selection and insufficient reliability of takeover schemes. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating an online reconfiguration control method for secondary equipment in an intelligent substation according to the present invention.

[0018] Figure 2 This is a schematic diagram of the structure of an online reconfiguration control system for secondary equipment in an intelligent substation according to the present invention.

[0019] Figure 3 Flowchart for verifying the target takeover equipment. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1, Figure 1 This invention provides an online reconfiguration control method for secondary equipment in an intelligent substation, comprising the following steps: S1 broadcasts the fault information containing digital fingerprints generated by the faulty secondary device to adjacent devices.

[0022] In this embodiment, broadcasting the fault information containing digital fingerprints generated by the faulty secondary device to adjacent devices specifically involves: Collect the unique identifier and real-time fault code of the faulty secondary equipment; The unique identifier, fault code, and current timestamp are combined, and a digital fingerprint is generated using a preset cryptographic hash function. The fault code is combined with the digital fingerprint to generate fault information, which is then broadcast.

[0023] Specifically: When a secondary device malfunctions, its unique identifier is collected first. and real-time generated fault codes And simultaneously obtain the current standard timestamp. ; The three elements mentioned above are combined and input into a preset cryptographic hash function to calculate the digital fingerprint. :

[0024] in: This indicates a string or data concatenation operation. Cryptographic hash functions, such as SHA-256, are used to map combinations of input elements to a fixed-length digest value; The digital fingerprint is used to uniquely identify the fault event.

[0025] By introducing a triplet of device identifier, fault code, and timestamp, and then calculating it using a hash function, the digital fingerprint possesses uniqueness, immutability, and time sensitivity, thereby ensuring the authenticity and traceability of fault information.

[0026] After obtaining digital fingerprints Then, compare it with the fault code. Combine the data to generate a complete fault information message. :

[0027] Finally, the faulty secondary device broadcasts the fault information to adjacent devices via a communication link.

[0028] S2 enables adjacent devices to verify digital fingerprints and perform lightweight consensus confirmation, and obtains dynamic reputation scores for secondary devices based on their operating data.

[0029] After receiving a fault information message, the adjacent device re-acquires the unique identifier from the message. Fault codes and timestamp And recalculate the digital fingerprint using the hash function used when generating the digital fingerprint:

[0030] like If the error is detected, the fault information is determined to be unaltered and the verification is successful; otherwise, the message is discarded.

[0031] Lightweight consensus confirmation is performed after verifying the digital fingerprint: In this embodiment, the lightweight consensus confirmation specifically refers to: Upon receiving a fault message, a neighboring device acts as a proposing node. The verification result of the digital fingerprint and the fault information are encapsulated into a consensus request and broadcast to neighboring devices; The remaining adjacent devices respond to the consensus request, independently verify the digital fingerprint, and report the verification results; When the proposing node receives more than a first preset threshold of valid verification results, it determines that a consensus has been reached and generates a consensus confirmation signal.

[0032] Specifically: When a neighboring device receives a fault message and completes local verification of the digital fingerprint, the device, as a proposing node, encapsulates its verification result and the fault message together into a consensus request message and broadcasts it to the other neighboring devices. The proposing device is the device that first receives the fault information and completes digital fingerprint verification in the set of neighboring devices. This device is responsible for broadcasting the consensus request. In the consensus process of the same fault event, the system usually designates a single proposing device.

[0033] If multiple devices meet the proposal conditions almost simultaneously, a unique proposing device is determined according to preset rules to avoid duplicate broadcasts, such as comparing the size of the device's unique identifier or the order of timestamps.

[0034] After receiving the consensus request, each adjacent device independently executes the same hash function verification process and obtains its own verification result. ,in: This indicates that the verification has passed; This indicates that the verification failed.

[0035] Each device then feeds back its own verification results to the proposing node; After receiving and collecting the verification results from all neighboring devices, the proposing node calculates the total number of valid verifications:

[0036] in, Represents a set of adjacent devices. This indicates the total number of devices.

[0037] The proposing node will compare the number of valid verifications with the first preset threshold. A comparison is made, and when the number of valid verifications is greater than the first preset threshold... The system determines that a consensus has been reached, generates a consensus confirmation signal, and broadcasts it to all adjacent devices.

[0038] After completing the lightweight consensus confirmation, the dynamic reputation score of each candidate takeover device is calculated based on the operating data of the secondary equipment. In this embodiment, obtaining the dynamic reputation score of the secondary equipment based on the secondary equipment's operating data specifically involves: The historical and real-time operating status data of the secondary equipment are acquired and preprocessed to obtain the first data. Based on an adaptively adjusted sliding time window and a forgetting factor, the first data is time-weighted to obtain the second data; Obtain the statistical features of the second data and combine them with the multimodal features of the device to construct a multimodal reputation feature vector; The feature vector is input into the attention-based scoring model to calculate and output a dynamic reputation score. The attention mechanism is used to adaptively assign weights to different features in the multimodal reputation feature vector.

[0039] Specifically: First, within a preset time period, historical and current operating status data of the secondary equipment are collected. The data includes electrical parameter data, communication link status data, and operation and maintenance log information. Historical operational status data can be retrieved from the monitoring center database, while current operational status data is collected in real-time by the online monitoring device. Timestamp alignment and missing value imputation are performed on historical and current operational status data to obtain the first data. Timestamp alignment is used to map different modal data to a unified time axis to ensure that the data have a corresponding relationship at the same time. Missing value imputation can be achieved through interpolation, preserving previous values, or based on prediction models, thereby ensuring the integrity of time series data; Then, the first data is time-weighted based on the sliding time window that is adaptively adjusted according to the operating conditions and the forgetting factor to obtain the second data; The weighted calculation formula is as follows:

[0040] in: Indicates time The second data obtained through weighted calculation; Indicates time index The first data below; Indicates the forgetting factor, The smaller the value, the faster the historical data decays; Indicates the current time A sliding window centered on the user; This indicates the length of the window, which is adaptively adjusted according to the operating conditions. When the operating conditions are in a steady state, the value is taken as... When the operating condition is a disturbance or fault, take ; Indicates the sliding window The time index is the value point at a specific moment within the window; during the weighting process, all values ​​within the window are traversed. Calculate the corresponding weighted contribution.

[0041] Statistical features are extracted from the second data and concatenated with the device's multimodal feature vector to obtain a multimodal reputation feature vector. The statistical features include trend slope, fluctuation amplitude, and stability index. The multimodal feature vector includes device type, environmental factors, communication status, etc. Furthermore, a reference information representing the global state is generated based on the current operating conditions; The reference information is compared with the components in the multimodal reputation feature vector to determine the degree of matching between different features and the current working conditions. During the comparison process, the multimodal reputation feature vector with higher relevance is assigned a larger weight, while the feature component with lower relevance is assigned a smaller weight, thereby realizing the dynamic ranking of feature importance. After weighting, all multimodal reputation feature vectors are weighted and combined to obtain a score that comprehensively reflects the reputation status of the device at that moment.

[0042] This rating result is the dynamic reputation score, which can be updated in real time according to changes in working conditions, making the model both adaptive and robust.

[0043] S3, selects a set of candidate takeover devices from neighboring devices based on dynamic reputation scores.

[0044] In this embodiment, the step of selecting a set of candidate takeover devices from neighboring devices based on dynamic reputation scores specifically involves: Devices with a credit score higher than the dynamic threshold are added to the initial candidate takeover device set; The initial set of candidate takeover devices is further filtered based on topology location and load balancing factor, and the final set of candidate takeover devices is output.

[0045] Specifically: First, a preliminary screening is performed on the reputation scores calculated from all adjacent devices, and a dynamic threshold is set. The dynamic threshold is obtained based on the average and standard deviation of the reputation scores of all devices. When a device's reputation score is higher than the dynamic threshold, the corresponding device is added to the initial candidate takeover device set. After obtaining the initial set of candidate takeover devices, the candidate set is further screened based on topology location and load balancing factor; The formula for calculating the topological location factor is as follows:

[0046] in, For candidate devices Network hop count with the faulty device; The formula for calculating the load balancing factor is as follows:

[0047] in, For candidate devices The current load percentage is: The load balancing factor and topology location factor are combined into a secondary screening criterion. :

[0048] in, This is an adjustment factor used to balance the importance of topology priority and load balancing; Finally, secondary screening indicators Higher than the secondary screening threshold The corresponding equipment is added to the candidate takeover equipment set; Output a set of candidate takeover devices.

[0049] S4, the target takeover device is determined from the candidate takeover device set using a multi-objective optimization algorithm, specifically as follows: A multi-objective optimization function is constructed, and multi-objective weight coefficients that match the current power grid state are dynamically generated based on a deep reinforcement learning model; The set of candidate takeover devices is used as the input to the solution space of the non-dominated sorting algorithm, and the multi-objective optimization function and its weight coefficients are used as the evaluation criteria. After iterative solution, the Pareto optimal solution set is obtained. The Pareto optimal solution set is sorted, and the candidate device corresponding to the highest-ranked solution is selected as the initial target takeover device; The initial target takeover equipment was prospectively verified based on digital twins, and after the verification was passed, it was finally determined as the target takeover equipment.

[0050] Specifically: Based on the power grid operation requirements, multiple optimization objectives are defined for the candidate equipment to be connected, and a multi-objective optimization function is constructed based on these objectives. The optimization objectives include minimizing takeover latency, maximizing operational reliability, optimizing load balancing, and maximizing communication timeliness. in, Indicates candidate device Takeover delay Indicates candidate device Operational reliability Indicates candidate device Load balancing Indicates candidate device Communication timeliness , , , The weights corresponding to takeover delay, operational reliability, load balancing, and communication timeliness can be set according to the actual needs of the power grid operation scenario.

[0051] The importance of the optimization objective varies under different operating conditions. To address this, this invention introduces a deep reinforcement learning model that dynamically outputs a set of multi-objective weight coefficients by sensing the power grid's operating state. This allows the optimization objective to adaptively adjust according to the power grid's condition.

[0052] The deep reinforcement learning model is built on the Actor-Critic framework; The input states of the deep reinforcement learning model include fault attributes, power grid load levels, and network topology information acquired in real time. The output action of the deep reinforcement learning model is used to dynamically generate the multi-objective weight coefficients; The deep reinforcement learning model is trained using a reward function, the value of which is positively correlated with the success rate of historical takeover operations and system stability indicators.

[0053] After obtaining the multi-objective optimization function and the multi-objective weight coefficients, the set of candidate takeover devices is used as the input to the non-dominated sorting algorithm in the solution space; In the initial phase, each device in the candidate takeover device set is coded and identified as an individual; Based on the constructed multi-objective optimization function, the original evaluation values ​​of the equipment on multiple objectives such as takeover delay, operational reliability, load balancing, and communication timeliness are calculated. By combining the multi-objective weight coefficients, these original evaluation values ​​are weighted or normalized to obtain a comprehensive evaluation vector that reflects the current working condition preferences. Each candidate device and its corresponding comprehensive evaluation vector are treated as an individual and aggregated to form the initial solution set of the algorithm.

[0054] During the iteration process, the algorithm compares candidate devices based on the comprehensive evaluation vector and the non-dominated ranking principle. Specifically: If a device's evaluation score is no lower than another device's across all weighted objectives, and it outperforms that device in at least one objective, then it is determined to be superior in multi-objective performance. In this case, the weaker device is marked as a suboptimal solution and will not proceed to the next iteration; while the stronger device is retained as a potential candidate solution to continue participating in the evolution. Through this step-by-step comparison and selection, it can be ensured that the solutions retained after each iteration have a relative advantage in multiple objective dimensions, thereby gradually approaching the Pareto optimal solution set.

[0055] After multiple iterations, the Pareto optimal solution set was finally obtained. The devices in this solution set are all compromise optimal solutions based on multi-objective optimization function calculation and dynamic adjustment of weight coefficients, representing candidate takeover device combinations that achieve a reasonable balance among different optimization objectives.

[0056] In this embodiment, the specific process of sorting the Pareto optimal solution set includes: The multi-objective weight coefficients are used as the evaluation criteria weights for the TOPSIS method; Determine the positive and negative ideal values ​​for each optimization objective, and calculate the weighted Euclidean distance between each candidate solution in the Pareto optimal solution set and the positive and negative ideal values ​​in each optimization objective dimension; The relative proximity of each candidate solution is calculated based on the weighted Euclidean distance, and the solutions are sorted according to the magnitude of the relative proximity to obtain the sorting result.

[0057] Specifically: Using the multi-objective weight coefficients dynamically generated by the deep reinforcement learning model as the weight input of the TOPSIS method, we can ensure that the ranking process is consistent with the current power grid operating conditions, and the influence of different objectives in the ranking can be adjusted in real time according to the operating status. The Pareto optimal solution set candidate solutions In its first The evaluation value under each optimization objective is denoted as . ; When the dimensions of different objectives differ significantly, normalization can be performed on each objective first to obtain normalized values. If the target dimensions are consistent, the original evaluation values ​​can be used directly.

[0058] Based on this, define: Positive ideal value : indicates the first The optimal performance under each objective is determined by taking the maximum value for benefit-oriented objectives and the minimum value for cost-oriented objectives. Positive ideal value : indicates the first The worst-case performance under each objective is determined by taking the minimum value for benefit-oriented objectives and the maximum value for cost-oriented objectives.

[0059] Combining multi-objective weighting coefficients We can calculate the weighted Euclidean distance between the candidate solution and the ideal solution:

[0060]

[0061] in: Indicates candidate solutions The weighted Euclidean distance between the ideal solution and the positive ideal solution; Indicates candidate solutions The weighted Euclidean distance between the solution and the negative ideal solution; Indicates the first The weighting coefficients of each objective.

[0062] The above calculations can quantify the relative differences between each candidate solution and the optimal and worst states, providing a basis for subsequent proximity calculations and ranking.

[0063] After obtaining the distance to the ideal solution, the relative closeness of the candidate solutions is further calculated. :

[0064] in, The value of is between 0 and 1, and the larger the value, the closer the candidate solution is to the ideal solution; Finally, according to The size of the algorithm is used to sort the candidate solutions in the Pareto optimal solution set, and the candidate device corresponding to the highest-ranked solution is taken as the initial target takeover device.

[0065] In this embodiment, the prospective verification of the initial target takeover equipment based on digital twins specifically includes: Input the topology parameters and operating status data of the target takeover equipment into the power grid digital twin simulation platform; Import future load forecast data and new energy output forecast data into the simulation platform to construct a takeover simulation scenario; Under the aforementioned takeover simulation scenario, dynamic power flow calculations of the power grid are performed to obtain node voltage, line power, and power balance data. Verification conclusions are generated based on the results of the power grid dynamic power flow calculation.

[0066] like Figure 3 As shown, the present invention provides a flowchart for the verification of target takeover equipment; Specifically: Input the topology parameters and operating status data of the initial target takeover equipment into the power grid digital twin simulation platform; The topology parameters are used to describe the node connection relationship and line attributes of the device in the power grid, and the operating status data are used to reflect the real-time operating status of the device, such as voltage, current, and power, thereby forming a corresponding virtual model in the simulation platform; The power grid digital twin simulation platform is used to build a model corresponding to the actual power grid operation status in a virtual environment. By importing power grid topology parameters, equipment operation data and prediction data, it performs dynamic power flow calculation and outputs operation constraint comparison results, thereby performing forward-looking verification of the target takeover equipment. Import load forecast data and renewable energy output forecast data for future periods into the simulation platform; the load forecast data is used to set the power demand curves of each node in the future period, and the renewable energy output forecast data is used to set the output curves of renewable energy sources such as wind power and photovoltaics. Based on the above data, the simulation platform introduces future operational disturbances into the original power grid digital twin model and embeds the takeover logic of the initial target takeover equipment into the simulation process, such as switching control or replacing faulty equipment at specific times.

[0067] In this way, the simulation platform can form a takeover simulation scenario that includes equipment takeover actions and future load and output changes, so that the power grid operation status can dynamically reflect possible future situations. In the aforementioned takeover simulation scenario, dynamic power flow calculations are performed to obtain the voltage distribution of each node, the power transmission status of each line, and the overall power balance of the system. After obtaining the results of the dynamic power flow calculation, they need to be compared with the preset operating constraints to form a verification conclusion. The constraints mainly include node voltage constraints, line power constraints, and system power balance constraints, as detailed below: Node voltage constraints:

[0068] Line power constraints:

[0069] System power balance constraints:

[0070] in: Indicates the first Voltage amplitude at each node; and These represent the lower and upper limits of the node voltage, respectively. Indicates the first The apparent power of the line; This indicates the maximum allowable power of the line; This indicates the system power balance error; This indicates the allowable error threshold.

[0071] If all three conditions are met, a conclusion that the verification passed is generated; if any condition is not met, a conclusion that the verification failed is generated. If the verification passes, the preliminary target takeover device is determined as the target takeover device; if the verification fails, the preliminary target takeover device is removed from the candidate set, and a new preliminary target takeover device is selected for iterative verification. After multiple rounds of iteration, the target takeover device is finally obtained.

[0072] S5 instructs the target takeover device to execute the reconfiguration logic.

[0073] After identifying the target equipment to be taken over, it is necessary to ensure that it has the same operating conditions as the faulty secondary equipment in order to complete the replacement of subsequent control functions; First, the operating status data of the faulty secondary equipment is loaded into the target takeover equipment. This operating status data includes the equipment's real-time monitoring data, control command cache, and communication interface parameters.

[0074] Through the loading process, the target takeover device can inherit the operating context of the faulty device at the data level, thereby avoiding data loss or control logic interruption during the switchover process. Subsequently, the target takeover device executes a preset reconstruction logic based on the loaded operating status. The reconstruction logic includes steps such as restoring protection and control strategies, rebuilding the communication link with the upper-level scheduling system, and re-establishing the collaborative control relationship with adjacent devices. Through this process, the target takeover equipment can seamlessly replace the faulty equipment in terms of function, realize the rapid reconfiguration of the control link, and ensure the continuity and stability of power grid operation.

[0075] Example 2, Figure 2 This invention discloses an online reconfiguration control system for secondary equipment in an intelligent substation, comprising: The fingerprint broadcasting module is used to broadcast fault information containing digital fingerprints generated by the faulty secondary device to adjacent devices; The consensus verification and scoring module is used to enable adjacent devices to verify digital fingerprints and perform lightweight consensus confirmation, and to obtain dynamic reputation scores for secondary devices based on the operating data of the secondary devices. The candidate screening module is used to select a set of candidate takeover devices from neighboring devices based on dynamic reputation scores; The optimization selection module is used to determine the target takeover device from the set of candidate takeover devices through a multi-objective optimization algorithm; The refactoring execution module is used to instruct the target takeover device to execute refactoring logic.

[0076] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0077] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0078] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0079] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0080] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0081] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for online reconfiguration control of secondary equipment in an intelligent substation, characterized in that, Includes the following steps: The fault information containing digital fingerprints generated by the faulty secondary device is broadcast to adjacent devices; The system enables adjacent devices to verify digital fingerprints and perform lightweight consensus confirmation, and obtains dynamic reputation scores for secondary devices based on their operational data. A set of candidate takeover devices is selected from neighboring devices based on dynamic reputation scores; The target takeover device is determined from the set of candidate takeover devices using a multi-objective optimization algorithm; Instruct the target takeover device to execute the reconfiguration logic.

2. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 1, characterized in that, The step of broadcasting the fault information containing digital fingerprints generated by the faulty secondary device to adjacent devices specifically involves: Collect the unique identifier and real-time fault code of the faulty secondary equipment; The unique identifier, fault code, and current timestamp are combined, and a digital fingerprint is generated using a preset cryptographic hash function. The fault code is combined with the digital fingerprint to generate fault information, which is then broadcast.

3. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 2, characterized in that, The lightweight consensus confirmation process specifically involves: Upon receiving a fault message, a neighboring device acts as a proposing node. The verification result of the digital fingerprint and the fault information are encapsulated into a consensus request and broadcast to neighboring devices; The remaining adjacent devices respond to the consensus request, independently verify the digital fingerprint, and report the verification results; When the proposing node receives more than a first preset threshold of valid verification results, it determines that a consensus has been reached and generates a consensus confirmation signal.

4. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 3, characterized in that, The dynamic reputation score of secondary equipment obtained based on the operating data of secondary equipment is as follows: The historical and real-time operating status data of the secondary equipment are acquired and preprocessed to obtain the first data. Based on an adaptively adjusted sliding time window and a forgetting factor, the first data is time-weighted to obtain the second data; Obtain the statistical features of the second data and combine them with the multimodal features of the device to construct a multimodal reputation feature vector; The feature vector is input into the attention-based scoring model to calculate and output a dynamic reputation score. The attention mechanism is used to adaptively assign weights to different features in the multimodal reputation feature vector.

5. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 4, characterized in that, The process of selecting a set of candidate takeover devices from neighboring devices based on dynamic reputation scores specifically involves: Devices with a credit score higher than the dynamic threshold are added to the initial candidate takeover device set; The initial set of candidate takeover devices is further filtered based on topology location and load balancing factor, and the final set of candidate takeover devices is output.

6. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 5, characterized in that, The step of determining the target takeover device from the candidate takeover device set using a multi-objective optimization algorithm specifically involves: A multi-objective optimization function is constructed, and multi-objective weight coefficients that match the current power grid state are dynamically generated based on a deep reinforcement learning model; The set of candidate takeover devices is used as the input to the solution space of the non-dominated sorting algorithm, and the multi-objective optimization function and its weight coefficients are used as the evaluation criteria. After iterative solution, the Pareto optimal solution set is obtained. The Pareto optimal solution set is sorted, and the candidate device corresponding to the highest-ranked solution is selected as the initial target takeover device; The initial target takeover equipment was prospectively verified based on digital twins, and after the verification was passed, it was finally determined as the target takeover equipment.

7. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 6, characterized in that, The deep reinforcement learning model includes: Construct a reinforcement learning model based on the Actor-Critic framework; Real-time power grid status information is used as the input status of the model; The output actions of the model are mapped to the weight coefficients of a multi-objective optimization function; The model is trained using a reward function based on historical takeover success rate and system stability.

8. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 7, characterized in that, The specific process of sorting the Pareto optimal solution set includes: The multi-objective weight coefficients are used as the evaluation criteria weights for the TOPSIS method; Determine the positive and negative ideal values ​​for each optimization objective, and calculate the weighted Euclidean distance between each candidate solution in the Pareto optimal solution set and the positive and negative ideal values ​​in each optimization objective dimension; The relative proximity of each candidate solution is calculated based on the weighted Euclidean distance, and the solutions are sorted according to the magnitude of the relative proximity to obtain the sorting result.

9. The online reconfiguration control method for secondary equipment in an intelligent substation according to claim 8, characterized in that, The aforementioned prospective verification of the initial target takeover equipment based on digital twins specifically includes: Input the topology parameters and operating status data of the target takeover equipment into the power grid digital twin simulation platform; Import future load forecast data and new energy output forecast data into the simulation platform to construct a takeover simulation scenario; Under the aforementioned takeover simulation scenario, dynamic power flow calculations of the power grid are performed to obtain node voltage, line power, and power balance data. Verification conclusions are generated based on the results of the power grid dynamic power flow calculation.

10. A system using the online reconfiguration control method for secondary equipment in an intelligent substation as described in any one of claims 1-9, comprising: The fingerprint broadcasting module is used to broadcast fault information containing digital fingerprints generated by the faulty secondary device to adjacent devices; The consensus verification and scoring module is used to enable adjacent devices to verify digital fingerprints and perform lightweight consensus confirmation, and to obtain dynamic reputation scores for secondary devices based on the operating data of the secondary devices. The candidate screening module is used to select a set of candidate takeover devices from neighboring devices based on dynamic reputation scores; The optimization selection module is used to determine the target takeover device from the set of candidate takeover devices through a multi-objective optimization algorithm; The refactoring execution module is used to instruct the target takeover device to execute refactoring logic.

Images