Full-process intelligent checking and control method for power grid dispatching operation ticket

Abstract

The invention relates to the technical field of intelligent management and control, in particular to a power grid dispatching operation ticket whole-process intelligent checking management and control method. The method comprises the following steps: S1, a dispatcher selects operated equipment in a dispatching management system and sets a target running state of the operated equipment; s2, acquiring the current running state of the operated equipment from the D5000 system, intelligently compiling a scheduling operation ticket in combination with a historical scheduling operation ticket, identifying a risk point through an anomaly detection integration algorithm, and generating a real-time scheduling operation ticket; s3, performing simulation operation rehearsal on the scheduling operation ticket through the operation verification model, and checking the correctness of the operation ticket; and S4, after the operation order is transferred to a pre-notice link, related personnel are notified, and after all personnel complete signature, the operation is automatically transferred to an execution link. Potential risk points are identified through an anomaly detection algorithm, early warning information is provided in advance, and operation risks are reduced. Through simulation operation rehearsal, correctness of the operation order is verified before actual operation, and power grid accidents caused by misoperation are avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and more specifically, to a full-process intelligent verification and control method for power grid dispatching operation tickets. Background Art

[0002] The full-process intelligent verification and control method for power grid dispatching operation tickets is an advanced technology developed in the operation and maintenance of power systems to ensure the safe and stable operation of the power grid and improve the accuracy and efficiency of dispatching operations. Traditional power grid dispatching operations mainly rely on manual review and execution, which have problems such as low efficiency and high risk of human errors. With the increasing complexity of power systems and the continuous improvement of requirements for safety and reliability, it has become necessary to use modern information technologies such as artificial intelligence and big data to achieve automatic verification, process control, and risk warning of dispatching operation tickets. By constructing an intelligent operation ticket management system, this method can effectively reduce the influence of human factors, improve the processing speed and quality of operation tickets, and ensure the safety and efficiency of power grid dispatching work.

[0003] In the existing full-process intelligent verification and control method for power grid dispatching operation tickets, for the power outage and power-on operations of power grid transmission and transformation equipment, the deputy dispatcher prepares a dispatching operation ticket in advance and notifies the on-site operator after the positive dispatcher has verified it without error. In the execution link, the deputy dispatcher gives orders and receives operation reports, and the positive dispatcher monitors the operation process. In the entire process of the execution of the above dispatching operation ticket, the deputy dispatcher prepares the operation ticket based on work experience, which is prone to compilation errors, and the positive dispatcher also has problems of oversight during the ticket review process. At the same time, in the operation ticket execution link, the positive and deputy dispatchers analyze the on-site operation process based on the telemetry and telecontrol changes of the intelligent dispatching control system (such as the OPEN3000 system or the D5000 system), which is prone to situations of inadequate monitoring, thus increasing the safety risks of misdispatching and misoperation to personnel, the power grid, and equipment. Therefore, a full-process intelligent verification and control method for power grid dispatching operation tickets is provided. Summary of the Invention

[0004] The purpose of the present invention is to provide a full-process intelligent verification and control method for power grid dispatching operation tickets to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides a full-process intelligent verification and control method for power grid dispatching operation tickets, including the following steps: S1. The dispatcher selects the equipment to be operated in the dispatching management system and sets its target operating state; S2. Obtain the current operating state of the equipment to be operated from the D5000 system, combine it with historical dispatching operation tickets, intelligently prepare a dispatching operation ticket, identify risk points through an anomaly detection integration algorithm, and generate a real-time dispatching operation ticket; S3. Perform a simulation operation rehearsal on the dispatching operation ticket through the operation verification model to check the correctness of the operation ticket. S4. After the operation ticket is transferred to the notice link, the operation confirmation reminder is carried out through the reminder module. After the operation is confirmed, the operation automatically flows to the execution link.

[0006] As a further improvement of this technical solution, in S1, the devices to be operated mainly include circuit breakers, lines, busbars, transformers, and grid devices; the target operating states are running, hot standby, cold standby, and maintenance states.

[0007] As a further improvement of this technical solution, in S2, the current operating state of the device to be operated is obtained from the D5000 system, combined with the historical dispatching operation ticket, the dispatching operation ticket is intelligently compiled, and the risk points are identified through the anomaly detection integration algorithm. The specific steps for generating the real-time dispatching operation ticket are as follows: S2.1. The dispatching management system obtains the current operating state of the selected device from the D5000 system. S2.2. The dispatching management system intelligently generates a draft operation ticket according to the historical dispatching operation ticket. Among them, the historical dispatching operation ticket is automatically retrieved from the dispatching management system. S2.3. The dispatching management system identifies potential risk points in the draft operation ticket through the anomaly detection integration algorithm and provides early warning information. S2.4. Combine the anomaly detection results to generate a real-time dispatching operation ticket.

[0008] As a further improvement of this technical solution, in S2.3, the specific steps for the dispatching management system to identify potential risk points in the draft operation ticket through the anomaly detection integration algorithm are as follows: S2.31. Collect historical operation ticket data and preprocess the collected historical operation ticket data. S2.32. According to the characteristics of the dispatching operation, select the features that can reflect the operation safety. Among them, the features reflecting the operation safety include operation type, operation time, device state, and environmental conditions. S2.33. Use the anomaly detection integration algorithm to make a judgment. S2.34. Output the judgment result.

[0009] As a further improvement of this technical solution, in S2.31, the specific steps for preprocessing the collected historical operation ticket data are as follows: Delete and correct missing values, outliers, and duplicate records: ; Among them, represents the cleaned data set. represents the original dataset; represents the th record in the original dataset; represents the th record is valid; Standardize the historical operation ticket data to make it have the same scale: ; Among them, represents the standardized value; represents a certain value in the original data; represents the mean of the data; represents the standard deviation of the data.

[0010] As a further improvement of this technical solution, in the S2.33, the specific steps for using the anomaly detection integration algorithm for judgment are as follows: Obtain the feature matrix and the label vector through the preprocessing in S2.31; Construct a K-NN model, a DBSCAN model, and an autoencoder model based on the historical operation ticket data, and form an anomaly detection integration model with these three models; Among them, the specific expression of the anomaly detection integration model is: ; Among them, represents the probability that the given feature vector , and the data point belongs to the anomaly class; represents the new feature vector of the th data point, ; represents the th data point; represents the nearest neighbor distance calculated by the K-NN model; represents the density calculated by the DBSCAN model; represents the reconstruction error calculated by the autoencoder model; represents the intercept term of the model; represents the influence degree of the feature on the model output; represents the influence degree of the feature on the model output; represents the influence degree of the feature on the model output; Considering the influence of the device health level on the probability of the anomaly class, introduce the health level For the anomaly detection integrated model Further optimization: ; Among them, represents the anomaly detection integrated model after introducing the health level ; represents the device health level; represents the feature 's influence degree on the model output; Among them, 's expression is: ; In the formula, represents the health level score of the device at time ; represents the weight of the temperature feature; represents the weight of the vibration feature; represents the weight of the insulation resistance feature; represents the maximum allowable operating temperature; represents the minimum allowable operating temperature; represents the actual temperature value measured at time ; represents the maximum allowable vibration level; represents the minimum allowable vibration level; represents the actual vibration value measured at time ; represents the maximum allowable insulation resistance; represents the minimum allowable insulation resistance; represents the actual insulation resistance value measured at time ; And considering the cumulative maintenance function to optimize the device health level , obtaining the device health level , substituting the optimized health level into the anomaly detection integrated model , obtaining the new anomaly detection integrated model ; Initialize the model parameters ; Use the logarithmic loss function: ; Among them, represents the logarithmic loss function; represents the number of samples in the dataset; represents the th sample's true label; Considering that the historical operation ticket data in different time periods has different abnormal tendencies, a timestamp is introduced to optimize the logarithmic loss function: ; Among them, represents the logarithmic loss function optimized by introducing a timestamp; represents the hyperparameter that controls the influence intensity of the timestamp; represents the th data timestamp; represents the timestamp weight function; Use the new anomaly detection integration model to evaluate the new operation ticket draft and calculate its anomaly probability : ; Among them, represents the new anomaly detection integration model; According to the anomaly probability and the preset threshold judge whether there is a potential risk in the operation ticket draft: When , it is considered that the draft has a potential risk; Otherwise, it is considered that the draft is normal.

[0011] As a further improvement of this technical solution, in S3, the specific steps of using the operation verification model to perform a simulated operation rehearsal on the dispatching operation ticket and checking the correctness of the operation ticket are as follows: S3.1. Train the operation verification model using historical operation ticket data; S3.2. Optimize the model parameters by minimizing the loss function; S3.3. Input the content of the operation ticket to be verified into the trained model and output the simulated operation result; S3.4. Check the correctness of the operation ticket by comparing the simulated operation result with the expected result; S3.5. When the check is correct, send the operation ticket to manual review; S3.6. When inconsistencies are found, return the problematic operation ticket to the operation ticket preparation link, re-prepare the operation ticket, and automatically perform simulated rehearsal verification.

[0012] As a further improvement of this technical solution, in S3.2, the specific expression for optimizing the model parameters by minimizing the loss function is: ; Among them, represents the model parameters; represents the minimum loss function; represents the content of the operation ticket input; represents the operation result output; represents the neural network function.

[0013] As a further improvement of this technical solution, in the S3.4, the specific expression for checking the correctness of the operation ticket by comparing the simulated operation result with the expected result is: ; where represents the overall evaluation function; represents the expected state of the th operation step; represents the actual state after the simulated operation of the th operation step; represents the total number of steps; When , it means that the simulated operation results of all steps are consistent with the expected results, and the operation ticket is correct; When , it means that the simulated operation result of at least one step is inconsistent with the expected result, and there is an error in the operation ticket.

[0014] As a further improvement of this technical solution, in the S4, after the operation ticket is transferred to the preview link, the operation confirmation reminder is carried out through the reminder module. After the operation is confirmed, the specific steps for the operation to automatically transfer to the execution link are: S4.1. Update the status of the operation ticket to "preview", and generate a preview reminder through the reminder module; S4.2. Group all monitoring station personnel, operation and maintenance personnel, and user station duty personnel and their mobile phone numbers by unit name, and send a text message reminder to inform that the operation ticket has entered the preview link; S4.3. Wait for all relevant personnel to complete the signature confirmation, and receive the signature information through the D5000 system; S4.4. When the operation is confirmed, automatically update the status of the operation ticket to "execution".

[0015] Compared with the prior art, the beneficial effects of the present invention: 1. In this full-process intelligent verification and control method for power grid dispatching operation tickets, by setting the target operation state, it ensures that the operation has a clear goal, reduces misoperations caused by unclear goals during the operation process. By using the anomaly detection algorithm to identify potential risk points and providing early warning information in advance, it reduces operation risks. By simulating the operation preview, it verifies the correctness of the operation ticket before the actual operation and avoids power grid accidents caused by improper operations.

[0016] 2. In the full-process intelligent verification and control method for power grid dispatching operation tickets, manual review can quickly and accurately check the correctness of operation tickets, improving the review efficiency. Ensure that the operation tickets go through pre-notification and signature confirmation, enhancing the standardization and transparency of operations. Enable the operation process inspection function to monitor the operation progress in real time and ensure that the operation proceeds as planned. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is the overall method flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0019] Please refer to Figure 1 As shown, this embodiment provides a full-process intelligent verification and control method for power grid dispatching operation tickets, including the following steps: S1. The dispatcher selects the equipment to be operated in the dispatching management system and sets its target operating state; In this example, the equipment to be operated mainly includes circuit breakers, lines, buses, transformers, and power grid equipment; the target operating states include running, hot standby, cold standby, and maintenance states.

[0020] S2. Obtain the current operating state of the equipment to be operated from the D5000 system, combine it with historical dispatching operation tickets, intelligently compile a dispatching operation ticket, and identify risk points through an integrated anomaly detection algorithm to generate a real-time dispatching operation ticket; Specifically, the D5000 system is an advanced intelligent power grid dispatching technology support platform adopted by the State Grid. It integrates multiple key functions such as real-time monitoring, early warning, dispatching plan, security verification, and dispatching management. Through efficient data processing and intelligent operation support, it ensures the safe, stable, and efficient operation of the power system, while meeting the requirements of modern power grids for remote access, alarm direct transmission, horizontal connection, and vertical management. Through automated and intelligent dispatching means, it reduces manual intervention, improves dispatching efficiency, shortens power outage time, and enhances power supply reliability.

[0021] In this example, the specific steps for obtaining the current operating state of the equipment to be operated from the D5000 system, combining it with historical dispatching operation tickets, intelligently compiling a dispatching operation ticket, and identifying risk points through an integrated anomaly detection algorithm to generate a real-time dispatching operation ticket are as follows: S2.1. The dispatching management system obtains the current operating status of the selected equipment from the D5000 system; S2.2. The dispatching management system intelligently generates a draft operation ticket based on historical dispatching operation tickets; Among them, the historical dispatching operation tickets are automatically retrieved from the dispatching management system; S2.3. The dispatching management system identifies potential risk points in the draft operation ticket through an integrated anomaly detection algorithm and provides early warning information; In this example, the specific steps for the dispatching management system to identify potential risk points in the draft operation ticket through an integrated anomaly detection algorithm are as follows: S2.31. Collect historical operation ticket data and preprocess the collected historical operation ticket data; In this example, the specific steps for preprocessing the collected historical operation ticket data are as follows: Delete and correct missing values, outliers, and duplicate records: ; Among them, represents the cleaned dataset; represents the original dataset; represents the th record in the original dataset; represents the th record is valid; Standardize the historical operation ticket data to make it have the same scale: ; Among them, represents the standardized value; represents a certain value in the original data; represents the mean of the data; represents the standard deviation of the data.

[0022] Specifically, first, delete and correct missing values, outliers, and duplicate records to ensure the integrity and consistency of the dataset and improve data quality; second, standardize the historical operation ticket data to make it have the same scale, avoid the impact of feature value range differences on model training, and improve the convergence speed and performance of the model. These preprocessing steps in the full-process intelligent verification and control method for power grid dispatching operation tickets ensure the quality and consistency of the data, lay a solid foundation for subsequent model training and verification, and thus improve the correctness and safety of the operation tickets.

[0023] S2.32. Select features that can reflect operation safety according to the characteristics of dispatching operations; Among them, the features reflecting operation safety include operation type, operation time, equipment status, and environmental conditions; S2.33. Use an anomaly detection integration algorithm for judgment; In this example, the specific steps for using the anomaly detection integration algorithm for judgment are as follows: After preprocessing through S2.31, a feature matrix and a label vector ; Based on historical operation ticket data, construct a K-NN model, a DBSCAN model, and an autoencoder model, and combine these three models into an anomaly detection integration model; Specifically, the training steps of the K-NN model are as follows: Select a suitable value; For each data point of the historical operation ticket data , calculate its distance from all other data points; Find the nearest neighbors; Calculate the average distance of these neighbors as : ; Among them, represents the number of nearest neighbors selected; represents the th data point; represents the th nearest neighbor data point of the data point ; Sum them up; Specifically, the K-NN model mainly identifies outliers by calculating the average distance between each data point of the historical operation ticket data and its closest neighbors. This method can effectively capture changes in local density. For data points that are far from their nearest neighbors, the K-NN model can mark them as potential outliers.

[0024] The training steps of the DBSCAN model are as follows: Select suitable parameters (neighborhood radius) and (minimum number of neighborhood points); For each data point of the historical operation ticket data , calculate the number of points within its neighborhood ; Calculate the density : ; Among them, represents the maximum number of neighborhood points of all data points; Specifically, the DBSCAN model determines whether a data point is an outlier based on the density around it. By setting an appropriate neighborhood radius and the minimum number of neighborhood points , data points located in low-density regions can be effectively identified.

[0025] The training steps of the autoencoder model are as follows: Construct an autoencoder model, including an encoder and a decoder part; Use historical operation ticket data to train the autoencoder so that the model can learn the compressed representation of the data; For each data point of the historical operation ticket data , calculate its reconstruction error : ; ; ; Among them, represents the low-dimensional representation of the th data point ; represents the reconstructed value of the th data point ; represents the encoding part of the autoencoder, which maps the input data to a low-dimensional encoding space; represents the decoding part of the autoencoder, which reconstructs the original data from the encoding space; Specifically, the autoencoder model identifies outliers by learning the compressed representation of the input data and attempting to reconstruct the input data. If the reconstruction error of a certain data point is significantly higher than that of other data points, then this data point may be considered an outlier.

[0026] The three models identify outliers based on different principles. The K-NN model focuses on local outlier measurement, the DBSCAN model focuses on density differences, and the autoencoder model relies on the reconstruction error of the data. Combining these three models can make up for the limitations that a single model may have and improve the accuracy and robustness of outlier detection.

[0027] Among them, the specific expression of the outlier detection integrated model is: ; Among them, represents the probability that the data point belongs to the outlier class for the given feature vector ; represents the new feature vector of the th data point, ; Indicates the th data point; Indicates the distance to the nearest neighbor calculated by the K-NN model; Indicates the density calculated by the DBSCAN model; Indicates the reconstruction error calculated by the autoencoder model; Indicates the intercept term of the model; Indicates the feature 's influence on the model output; Indicates the feature 's influence on the model output; Indicates the feature 's influence on the model output; Considering the influence of the device health level on the probability of the abnormal class , the health level is introduced to further optimize the anomaly detection ensemble model: ; ; Among them, Indicates the anomaly detection ensemble model after introducing the health level ; Indicates the device health level; Indicates the feature 's influence on the model output; Among them, 's expression is: ; In the formula, Indicates the health level score of the device at time , which reflects the current overall health status of the device and is usually normalized to the interval [0,1], where 1 represents the best state and 0 represents the worst state; Indicates the weight of the temperature feature, which reflects the influence degree of temperature on the device health status; Indicates the weight of the vibration feature, which represents the influence degree of vibration on the device health status; Indicates the weight of the insulation resistance feature, which represents the influence degree of insulation resistance on the device health status; Indicates the maximum allowable operating temperature; Indicates the minimum allowable operating temperature; Indicates the actual temperature value measured at time ; Indicates the maximum allowable vibration level; Indicates the minimum allowable vibration level; Indicates the time The actual vibration value measured at time Represents the maximum allowable insulation resistance; Represents the minimum allowable insulation resistance; Represents at time The actual insulation resistance value measured; Quantified by a series of key performance indicators (KPIs) reflecting the device state, to judge the health score of the device at time ; To ensure that data with different dimensions and ranges can be compared with each other and that the final score is not unduly affected by the large value of a certain feature, each feature has been standardized; The temperature Is mapped to the interval [0,1], where 1 represents the optimal state of the temperature (i.e., the lowest), and 0 represents the worst state (i.e., the highest); Similarly, vibration and insulation resistance are also converted into scores on the same scale; The standardized feature scores are assigned corresponding weights to reflect their different degrees of influence on the device health state; Specifically, the health of the device may decline due to various factors, may increase due to maintenance, and there will also be an aging problem. In the initial stage, the aging may be slow, but as time increases, the aging rate will accelerate; The aging of the device also indicates that maintenance is required. Maintenance is not just the effect of a single event, but the cumulative effect of multiple maintenances. Introduce a cumulative maintenance function To reflect the cumulative effect of multiple maintenances: ; ; Among them, Represents the th maintenance effect; Represents the th maintenance time; Represents the Dirac delta function, indicating a pulse at moment; Then The expression of is: ; In the formula, Represents the health score of the optimized device; Represents; Represents the aging function, used to quantify the aging rate of the device over time; Specifically, , in the formula, And Represents two positive parameters, among which, Controls the maximum aging degree, Control the aging rate. When increases, it will gradually approach , simulating the phenomenon that the aging speed of the device gradually increases over time; Substitute the optimized health level into the anomaly detection integration model for further optimization to obtain a new anomaly detection integration model : ; Combine the K-NN model, DBSCAN model and autoencoder model, and introduce the device health level and cumulative maintenance function to optimize the anomaly detection integration model, so that the model not only depends on the operation ticket data itself, but also can be adjusted according to the actual operating conditions of the device, which helps to more accurately identify potential problems when the device performance deteriorates; by introducing the aging function and cumulative maintenance function, the model can better simulate the change trend of the device over time. Considering that the aging of the device will gradually increase the risk of failure, while maintenance can reverse or slow down this process to a certain extent, this method provides strong support for predictive maintenance.

[0028] Initialize the model parameters ; Use the logarithmic loss function: ; Among them, represents the logarithmic loss function; represents the number of samples in the dataset; represents the true label of the th sample; Considering that the historical operation ticket data in different time periods has different anomaly tendencies, introduce a timestamp to optimize the logarithmic loss function: ; Among them, represents the logarithmic loss function optimized by introducing a timestamp; represents the hyperparameter that controls the influence intensity of the timestamp; represents the th data timestamp; represents the weight function of the timestamp ; Specifically, by introducing a timestamp to optimize the logarithmic loss function , the model can better adapt to the anomaly tendencies in different time periods, improving the accuracy and stability of anomaly detection. This method not only enhances the time sensitivity and robustness of the model, but also improves the precision and recall rate of anomaly detection, supports real-time monitoring and early warning, thus providing strong guarantee for the safe operation of the power grid.

[0029] Use a new integrated anomaly detection model Evaluate the new draft operation ticket and calculate its anomaly probability : ; Among them, represents the new integrated anomaly detection model; According to the anomaly probability and the preset threshold Judge whether there are potential risks in the draft operation ticket: When , it is considered that there are potential risks in this draft; Otherwise, it is considered that this draft is normal.

[0030] S2.34. Output the judgment result.

[0031] S2.4. Combine the anomaly detection results to generate a real-time dispatching operation ticket.

[0032] S3. Use the operation verification model to perform a simulated operation preview on the dispatching operation ticket to check the correctness of the operation ticket; In this example, the specific steps to use the operation verification model to perform a simulated operation preview on the dispatching operation ticket to check the correctness of the operation ticket are as follows: S3.1. Use the historical operation ticket data to train the operation verification model; S3.2. Optimize the model parameters by minimizing the loss function; In this example, the specific expression for optimizing the model parameters by minimizing the loss function is: ; Among them, represents the model parameters; represents the minimum loss function; represents the content of the input operation ticket; represents the output operation result; represents the neural network function.

[0033] Among them, The function expression of ; Among them, represents the activation function; represents the weight matrix of the first layer; represents the weight matrix of the second layer; represents the bias vector of the first layer; represents the bias vector of the second layer; Specifically, by optimizing the model parameters, it is ensured that the model can accurately predict the operation results, thereby improving the performance and reliability of the model. By minimizing the loss function, the model can continuously adjust the parameters during the training process to best fit the historical data, providing an accurate prediction basis for the subsequent simulation operation preview.

[0034] S3.3. Input the content of the operation ticket to be verified into the trained model and output the simulation operation result; S3.4. Check the correctness of the operation ticket by comparing the simulation operation result with the expected result; In this example, the specific expression for checking the correctness of the operation ticket by comparing the simulation operation result with the expected result is: ; Among them, represents the overall evaluation function; represents the expected state of the th operation step; represents the actual state after the simulation operation of the th operation step; represents the total number of steps; represents the comparison function; When , it means that the simulation operation results of all steps are consistent with the expected results, and the operation ticket is correct; When , it means that the simulation operation result of at least one step is inconsistent with the expected result, and there is an error in the operation ticket.

[0035] Specifically, by calculating the overall evaluation function , the consistency between the simulation result and the expected result of each operation step is evaluated. When , it means that the simulation operation results of all steps are consistent with the expected results, and the operation ticket is correct; when , it means that the simulation operation result of at least one step is inconsistent with the expected result, and there is an error in the operation ticket. In this way, the correctness and safety of the operation ticket can be ensured, and power grid accidents caused by improper operations can be avoided.

[0036] S3.5. When the check is correct, send the operation ticket to manual review; S3.6. When inconsistencies are found, return the problematic operation ticket and the operation ticket to the operation ticket compilation link, recompile the operation ticket, and automatically perform simulation preview verification.

[0037] Specifically, the dispatching operation ticket is simulated and rehearsed through the operation verification model to check the correctness of the operation ticket. This process not only provides an intuitive and highly interactive simulation operation environment through virtual reality technology to help dispatchers discover and correct potential errors in advance, but also ensures the accuracy and feasibility of the operation ticket by comparing the simulation operation results with the expected results. This can significantly improve the safety and reliability of operations, reduce risks in actual operations, and at the same time enhance the operation skills and emergency handling capabilities of dispatchers, ensuring the efficiency and stability of power grid operations.

[0038] Through manual review, confirm the correctness of the operation ticket; Specifically, the specific steps for confirming the correctness of the operation ticket through manual review are as follows: Input the operation ticket information and conduct manual review; check the logical relationship between operation steps to confirm whether the operation sequence is reasonable and whether there are conflicts; check whether the operation ticket complies with the power system operation regulations; when it is correct, the operation ticket is transferred to the pre-warning link; when it is incorrect, the dispatcher directly marks the incorrect instructions on the operation ticket and returns it to the compilation link for re-compilation, simulation operation rehearsal, and dispatcher review until the operation ticket is correct.

[0039] Specifically, in the full-process intelligent verification and control method of power grid dispatching operation tickets, manual review ensures the correctness and safety of the operation ticket through steps such as input of operation ticket information, checking of the logical relationship of operation steps, and verification of compliance with operation regulations. When the operation ticket passes the review, it is transferred to the pre-warning link; if an error is found, the dispatcher marks the error and returns it to the compilation link until the operation ticket is completely correct, thereby ensuring the accuracy and reliability of power grid operations and preventing power grid accidents caused by improper operations.

[0040] S4. After the operation ticket is transferred to the pre-warning link, the reminder module is used to give an operation confirmation reminder. After the operation is confirmed, the operation automatically transfers to the execution link; In this example, the specific steps for the operation ticket to be transferred to the pre-warning link and then the reminder module to give an operation confirmation reminder and the operation automatically transfer to the execution link after confirmation are as follows: S4.1. Update the status of the operation ticket to "pre-warning" and generate a pre-warning reminder through the reminder module; S4.2. Group all monitoring station personnel, maintenance personnel, and user station duty personnel and their mobile phone numbers by unit name and send a text message reminder to inform that the operation ticket has entered the pre-warning link; S4.3. Wait for all relevant personnel to complete the signature confirmation and receive the signature information through the D5000 system; S4.4. When all signature confirmations are completed, automatically update the status of the operation ticket to "execution".

[0041] Specifically, by automating the preview and signature confirmation processes of operation tickets, it is ensured that all relevant personnel fully recognize and prepare for the operation tickets before they are officially executed. This process includes updating the status of the operation ticket to "preview" and automatically generating a notice, reminding the personnel of relevant units via text message, waiting for all signature confirmations, and automatically updating the status of the operation ticket to "execution" after the confirmations are completed. This mechanism not only improves the transparency and standardization of the operation ticket circulation but also ensures that all relevant personnel can timely understand and participate in the review process of the operation ticket, thus enhancing the safety and coordination of operations and guaranteeing the smooth progress of power grid operations.

[0042] After the operation ticket starts to be executed, after the dispatcher issues an operation order, the operation process inspection function is enabled.

[0043] Specifically, after the operation ticket starts to be executed and the dispatcher issues an operation order, the specific steps for enabling the operation process inspection function are as follows: The dispatcher issues an operation order through the intelligent dispatching control system; enables the operation process inspection function in the intelligent dispatching control system. When the inspection is qualified, the dispatching instruction is marked as green, otherwise it is marked as red to help the dispatcher check the completion of the operation; records all status changes during the operation process, generates an operation report, and archives it for future reference.

[0044] Specifically, the issuance of operation orders and the real-time monitoring of the operation process are realized through the intelligent dispatching control system. The dispatcher issues an operation order through the system and enables the operation process inspection function. The system automatically checks the completion of the operation. Qualified operation instructions are marked as green, and unqualified ones are marked as red to help the dispatcher timely discover and correct problems. In addition, the system also records all status changes during the operation process, generates an operation report and archives it for future reference to ensure the transparency and traceability of the operation. This mechanism not only improves the accuracy and safety of the operation but also provides detailed data support for subsequent audits and analyses, guaranteeing the efficiency and reliability of power grid operations.

[0045] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art of this industry should understand that the present invention is not limited by the above embodiments. The above embodiments and the descriptions in the specification are only the preferred examples of the present invention and are not used to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed.

Claims

1. A method for intelligent verification and control of the entire process of power grid dispatching operation tickets, characterized by: The following steps are involved: S1. Select the operated equipment in the dispatching management system and set its target operating state; S2. Obtain the current operating status of the operated equipment from the D5000 system, combine it with the historical dispatch operation ticket, intelligently compile the dispatch operation ticket, identify risk points through the anomaly detection integrated algorithm, and generate a real-time dispatch operation ticket; S3. Perform simulated operation rehearsal on the dispatching operation ticket through the operation verification model to check the correctness of the operation ticket; S4. After the operation ticket is transferred to the notice stage, the reminder module will be used to remind the user to confirm the operation. After the operation is confirmed, the operation will automatically be transferred to the execution stage.

2. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 1 is characterized by: In S1, the operated equipment mainly includes circuit breakers, lines, busbars, transformers, and power grid equipment; the target operating states include operation, hot standby, cold standby, and maintenance state.

3. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 1 is characterized in that: In S2, the current operating status of the operated equipment is obtained from the D5000 system, and the dispatch operation ticket is intelligently compiled in combination with the historical dispatch operation ticket. The risk points are identified through the anomaly detection integrated algorithm, and the specific steps of generating the real-time dispatch operation ticket are as follows: S2.1, the dispatch management system obtains the current operating status of the selected equipment from the D5000 system; S2.2, the dispatch management system intelligently generates a draft operation ticket based on the historical dispatch operation ticket; Among them, historical dispatch operation tickets are automatically retrieved from the dispatch management system; S2.3, the dispatch management system identifies potential risk points in the draft operation ticket through anomaly detection integrated algorithm and provides early warning information; S2.

4. Generate a real-time dispatch operation ticket based on the anomaly detection results.

4. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 3 is characterized by: In S2.3, the specific steps for the dispatch management system to identify potential risk points in the draft operation ticket through the anomaly detection integrated algorithm are: S2.

31. Collect historical operation ticket data and pre-process the collected historical operation ticket data; S2.

32. According to the characteristics of the dispatching operation, select features that can reflect the safety of the operation; Among them, the characteristics reflecting the safety of operation include operation type, operation time, equipment status, and environmental conditions; S2.33, use anomaly detection integrated algorithm to make judgments; S2.

34. Output the judgment result.

5. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 4 is characterized in that: In S2.31, the specific steps of preprocessing the data collected from historical operation tickets are as follows: Delete and correct missing values, outliers, and duplicate records: ； in, represents the cleaned data set; represents the original dataset; Indicates the original data set records; Indicates The record is valid; Standardize the historical operation ticket data to make them have the same scale: ； in, Indicates the normalized value; Represents a value in the original data; represents the mean of the data; represents the standard deviation of the data.

6. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 4 is characterized by: In S2.33, the specific steps of using the anomaly detection integrated algorithm to make judgments are: After preprocessing by S2.31, the feature matrix is ​​obtained and label vector ; According to the historical operation ticket data, a K-NN model, a DBSCAN model and an autoencoder model are constructed respectively, and these three models are combined into an anomaly detection integrated model; Among them, the anomaly detection integrated model The specific expression is: ； in, Represents a given feature vector , data points The probability of belonging to the anomaly class; Indicates The new feature vector of the data points is ; Indicates data points; Represents the nearest neighbor distance calculated by the K-NN model; Represents the density calculated by the DBSCAN model; Represents the reconstruction error calculated by the autoencoder model; represents the intercept term of the model; Representation characteristics The degree of impact on the model output; Representation characteristics The degree of impact on the model output; Representation characteristics The degree of impact on the model output; Considering the health of the equipment Probability of abnormal class The impact of introducing health Ensemble models for anomaly detection Further optimization: ； in, Indicates the level of health introduced The anomaly detection ensemble model behind; Indicates the health of the device; Representation characteristics The degree of impact on the model output; in, The expression is: ； In the formula, Indicates at time The health score of the equipment at that time; represents the weight of the temperature feature; represents the weight of the vibration feature; represents the weight of the insulation resistance characteristic; Indicates the maximum permissible operating temperature; Indicates the minimum permissible operating temperature; Indicates at time The actual temperature value measured at Indicates the maximum permissible vibration level; Indicates the minimum permissible vibration level; Indicates at time The actual vibration value measured at Indicates the maximum allowable insulation resistance; Indicates the minimum allowable insulation resistance; Indicates at time The actual insulation resistance value measured when And consider the cumulative maintenance function Equipment health Optimize and get the health of the equipment , the optimized health level Substitute into the anomaly detection ensemble model In this paper, we get a new anomaly detection integrated model ; Initialize model parameters ; Using the logarithmic loss function: ； in, represents the logarithmic loss function; Indicates the number of samples in the data set; Indicates The true labels of samples; Considering that historical operation ticket data in different time periods have different abnormal tendencies, the timestamp optimization logarithmic loss function is introduced: ； in, Represents the logarithmic loss function after the introduction of timestamp optimization; represents the hyperparameter that controls the strength of the timestamp effect; Indicates The timestamp of the data; Indicates timestamp The weight function of Using the new anomaly detection ensemble model Evaluate new draft operation tickets and calculate their abnormal probability : ； in, represents a new anomaly detection ensemble model; According to the abnormal probability and preset thresholds Determine whether the draft operation ticket has potential risks: when , then the draft is considered to have potential risks; Otherwise, the draft is considered normal.

7. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 1 is characterized in that: In S3, the specific steps of performing a simulated operation preview on the dispatch operation ticket through the operation verification model to check the correctness of the operation ticket are as follows: S3.

1. Use historical operation ticket data to train the operation verification model; S3.2, optimize model parameters by minimizing the loss function; S3.

3. Input the content of the operation ticket to be verified into the trained model and output the simulation operation result; S3.

4. Check the correctness of the operation ticket by comparing the simulated operation results with the expected results; S3.

5. When the check is correct, the operation ticket will be sent to manual review; S3.

6. When any inconsistency is found, the problematic operation ticket and the operation ticket will be returned to the operation ticket preparation stage, and the operation ticket will be re-prepared and automatically simulated and checked.

8. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 7 is characterized by: In S3.2, the specific expression for optimizing the model parameters by minimizing the loss function is: ； in, represents the model parameters; represents the minimum loss function; Indicates the input operation ticket content; Indicates the output operation result; Represents a neural network function.

9. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 7 is characterized by: In S3.4, by comparing the simulated operation result with the expected result, the specific expression for checking the correctness of the operation ticket is: ； in, represents the overall evaluation function; Indicates The expected status of each operation step; Indicates The actual status after the simulated operation of each operation step; Indicates the total number of steps; Represents a comparison function.

10. The method for intelligent verification and control of the entire process of power grid dispatching operation tickets according to claim 1 is characterized in that: In S4, after the operation ticket is transferred to the notice stage, the reminder module is used to remind the operation confirmation. After the operation is confirmed, the operation automatically flows to the execution stage. The specific steps are: S4.

1. Update the operation ticket status to "advance notice" and generate a advance notice reminder through the reminder module; S4.

2. Group all monitoring station personnel, operation and maintenance personnel, and user station duty personnel and their mobile phone numbers by unit name, and send SMS reminders to remind that the operation ticket has entered the notice phase; S4.

3. Wait for all relevant personnel to complete signature confirmation and receive signature information through the D5000 system; S4.

4. When the operation is confirmed, the operation ticket status will be automatically updated to "Execute".

Images