Automatic checking system and method for unplanned operation of transformer substation based on artificial intelligence

Abstract

The invention relates to the field of artificial intelligence, and discloses an automatic checking system and method for unplanned operation of a transformer substation based on artificial intelligence, and the method comprises the steps: collecting an equipment operation state, an operation instruction, sensor data and alarm information in real time, and transmitting the collected data to a central platform in real time through a high-speed communication network; the collected real-time data, historical operation data, alarm records and equipment maintenance logs are stored in a central platform; comparing the real-time operation data with historical operation records, and identifying whether unplanned operation which is not approved or exceeds a specified operation range exists or not; performing risk assessment on the unplanned operation based on the property of the operation, the influence range and the equipment state, and judging the influence of the operation on the substation equipment and system; and based on real-time data and historical data, self-learning is carried out through a deep learning algorithm, and the precision and efficiency of the unplanned operation recognition model are continuously optimized. The method has the advantage that the operation safety and the management efficiency are improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an automatic verification system and method for unplanned operations of a substation based on artificial intelligence. Background Art

[0002] With the improvement of automation and intelligence level of power system, substation is an important node of power transmission, and its operation safety and reliability are crucial to the stability of the whole power system. Substation not only needs to undertake basic tasks such as daily power conversion, distribution and dispatching, but also needs to deal with emergencies such as equipment failure, load fluctuation, natural disasters, etc. Therefore, the management and maintenance of substations are extremely complex and demanding to ensure the efficient operation of power system and the stability of power supply. The traditional substation operation management mode often relies on manual inspection and verification, and has certain limitations. Manual inspection is not only a huge workload, but also often relies on the experience and judgment of on-site staff. On the one hand, it is easy to miss or misdetect, and on the other hand, it is difficult to monitor and feedback all operations in real time and accurately. In addition, manual verification usually has a certain time delay, and it is impossible to detect potential safety hazards in time, especially when the load is high or the equipment fails, it is easy to miss the key detection window. Therefore, it is necessary to design an automatic verification system and method for unplanned substation operations based on artificial intelligence to improve operation safety and management efficiency. Summary of the invention

[0003] 1. Technical issues to be solved

[0004] In view of the deficiencies in the prior art, the present invention provides an automatic verification system and method for unplanned operations of substations based on artificial intelligence, which has the advantages of improving operation safety and management efficiency, and solves the problems in the above-mentioned background technology.

[0005] (II) Technical solution

[0006] In order to achieve the above-mentioned purpose of improving operation safety and management efficiency, the present invention provides the following technical solutions: an automatic verification system for unplanned operations of substations based on artificial intelligence, comprising a data acquisition module, a data storage module, an unplanned operation identification module, an intelligent decision-making module and an optimization and feedback module;

[0007] The data acquisition module is used to collect the operating status, operation data, operation instructions, sensor data, and alarm information of each device in the substation in real time;

[0008] The data storage module is used to store historical operation data, alarm records, equipment maintenance logs, and use big data analysis technology to conduct in-depth mining to identify potential long-term risks or trends. Through data mining, the operation management process is optimized and support is provided for future decision-making;

[0009] The unplanned operation identification module performs intelligent analysis on real-time operation data based on artificial intelligence to identify whether there are unplanned operations that have not been approved or exceed the specified operation scope, and determines whether the operation is abnormal or unauthorized by comparing historical operation records, operation instructions and equipment status;

[0010] After the intelligent decision-making module identifies an unplanned operation, the system will automatically generate an alarm message and push it to the relevant person in charge, and provide decision support. According to the severity of the operation and the scope of equipment impact, the risk is intelligently assessed and a response strategy is given;

[0011] The optimization and feedback module is based on a deep learning algorithm, self-learns from historical data and real-time data, optimizes the accuracy and efficiency of unplanned job recognition, and gradually adjusts model parameters through feedback learning to improve system recognition capabilities.

[0012] The automatic verification method of unplanned operation of a substation based on artificial intelligence includes the following steps:

[0013] S1: Collect equipment operation status, operation instructions, sensor data, and alarm information in real time, and transmit the collected data to the central platform in real time through a high-speed communication network;

[0014] Preferably, S1 further includes establishing communication connections with various devices and sensors in the substation, obtaining real-time operation data, operation instructions issued by the real-time automation system, collecting alarm information detected by sensors and equipment in the substation, and transmitting the collected real-time data to the central monitoring platform through a high-speed communication network.

[0015] S2: Store the collected real-time data as well as historical operation data, alarm records, and equipment maintenance logs on a central platform, and use big data analysis technology to analyze historical operation data to identify long-term risks, operation patterns, and trends;

[0016] Preferably, S2 further includes integrating the collected real-time data, historical operation data, alarm records, and equipment maintenance logs into the data storage system of the central platform for classified archiving, and using big data analysis technology to deeply analyze historical operation data to identify potential risks such as equipment aging, excessive operation frequency, repetitive operations, and long-term unmaintained equipment, and predict future equipment failures or safety hazards.

[0017] S3: Compare real-time operation data with historical operation records to identify whether there are unplanned operations that have not been approved or exceed the prescribed operation scope. Perform intelligent analysis on real-time operation data and use the trained AI model to identify whether there are unplanned operations.

[0018] Preferably, S3 further includes comparing the real-time operation data with the historical operation records of the substation to check whether the current operation has been approved and is within the specified scope, and using the historical operation data and abnormal operation behavior data to train the artificial intelligence model. The model identifies unplanned operations by analyzing the characteristics of the operations. During the training process, the historical operation data is learned using a machine learning algorithm to establish a recognition model.

[0019] S4: Conduct risk assessment on unplanned operations based on the nature of the operations, scope of impact and equipment status, and determine the impact of the operations on substation equipment and systems;

[0020] Preferably, S4 further includes classifying real-time operations, identifying the nature of the operations, evaluating the impact scope of unplanned operations by analyzing the equipment and areas involved in the operations, and establishing a risk assessment model using multi-dimensional data in combination with the nature of the operations, the impact scope and the status of the equipment.

[0021] S5: Based on real-time data and historical data, the system conducts self-learning through deep learning algorithms to continuously optimize the accuracy and efficiency of the unplanned job recognition model. It also optimizes the recognition model and system configuration based on feedback information to improve recognition accuracy.

[0022] Preferably, S5 further includes using a deep learning algorithm to train historical data, establishing an unplanned job recognition model, and learning new data obtained from job feedback, system status monitoring, and alarm information in real time through online learning and adaptive adjustment mechanisms. Through reinforcement learning or incremental learning, the model is adapted to new environmental changes, real-time data is combined with historical data, and the model is continuously adjusted to adapt to future unplanned job patterns.

[0023] (III) Beneficial effects

[0024] Compared with the prior art, the present invention provides an automatic verification system and method for unplanned operations of substations based on artificial intelligence, which has the following beneficial effects:

[0025] The present invention collects the equipment operation status, operation instructions, sensor data, and alarm information in real time, and transmits the collected data to the central platform in real time through a high-speed communication network; the collected real-time data and historical operation data, alarm records, and equipment maintenance logs are stored in the central platform, and the historical operation data are analyzed by big data analysis technology to identify long-standing risks, operation modes, and trends; the real-time operation data is compared with the historical operation records to identify whether there are unplanned operations that have not been approved or exceed the specified operation scope, and the real-time operation data is intelligently analyzed, and the trained Al model is used to identify whether there are unplanned operations; based on the nature of the operation, the scope of influence, and the equipment status, the risk of unplanned operations is assessed to determine the impact of the operation on the substation equipment and system; based on real-time data and historical data, self-learning is performed through a deep learning algorithm, the accuracy and efficiency of the unplanned operation recognition model are continuously optimized, and the recognition model and system configuration are optimized according to feedback information to improve the recognition accuracy. It has the advantages of improving operation safety and management efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a schematic diagram of the structure of the present invention.

[0027] Figure 2 Schematic diagram of the method of the present invention. DETAILED DESCRIPTION

[0028] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Example 1

[0030] The present invention provides a technical solution: an automatic verification method for unplanned operations of a substation based on artificial intelligence, comprising the following steps:

[0031] S1: Collect equipment operation status, operation instructions, sensor data, and alarm information in real time, and transmit the collected data to the central platform in real time through a high-speed communication network;

[0032] Establish communication connections with various devices and sensors in the substation to obtain real-time operation data. It can accurately capture the working status of the equipment, such as current, voltage, temperature, pressure and other parameters, covering all key equipment in the substation, ensuring that the operating status of each device can be monitored in real time. The high-precision data of equipment and sensors provides detailed and accurate operation information, which is helpful for subsequent analysis.

[0033] Real-time collection of operating instructions issued by operators or automation systems, such as equipment start and stop, adjustment operations, etc., records information such as instruction type, executor, time, and operation target. Ensure that all operating instructions can be recorded and archived in real time, and the operation process is clear and traceable. Real-time collection of instructions reduces the tediousness of manual recording and improves operation management efficiency. The association of instruction data with equipment operating status enables abnormalities in the operation execution process to be identified in a timely manner.

[0034] Collect alarm information detected by sensors and equipment in the substation, including early warning information such as overcurrent, overvoltage, temperature limit, equipment failure, etc. Alarm information is transmitted in real time to ensure that any abnormal situation in the substation can be discovered and responded to immediately. Through alarm data, necessary safety measures can be taken quickly to avoid further expansion of the fault. Collect various types of alarm information to fully cover the potential safety hazards of substation equipment.

[0035] The collected real-time data is transmitted to the central monitoring platform through a high-speed communication network. Ensure that the data is transmitted to the platform with almost no delay, support real-time monitoring and decision-making, use a high-bandwidth, low-latency communication network to ensure the stability and efficiency of data transmission, and high-speed communication supports large-scale data transmission, ensuring that the platform can receive and process large amounts of data in a timely manner.

[0036] S2: Store the collected real-time data as well as historical operation data, alarm records, and equipment maintenance logs on a central platform, and use big data analysis technology to analyze historical operation data to identify long-term risks, operation patterns, and trends;

[0037] The collected real-time data, historical operation data, alarm records, equipment maintenance logs and other information are integrated into the data storage system of the central platform for classified archiving. The data from different sources are integrated to form a complete data warehouse to facilitate subsequent comprehensive analysis, ensure the long-term storage and traceability of historical data, facilitate long-term trend analysis, optimize storage space through data archiving technology, and ensure the efficiency and reliability of data storage.

[0038] Using big data analysis technology, we deeply analyze historical operation data, identify potential risks such as equipment aging, excessive operation frequency, repetitive operations, and long-term unmaintained equipment, and predict future equipment failures or safety hazards. We identify possible equipment failures or safety hazards in advance, make preparations in advance, predict future risks and equipment status based on the changing trends of historical data, avoid sudden failures, provide accurate risk assessment reports, provide data support for managers' decision-making, and optimize resource allocation and operation plans.

[0039] By analyzing historical operation data, common operation modes can be identified, such as equipment maintenance cycle, operator operation frequency, operation type, etc., to find out whether there are abnormal operation behaviors, such as frequent temporary operations or unapproved operations. It helps to analyze the operation behaviors of operators and identify whether there are irregular operation behaviors. Through the analysis of operation modes, the operation process and operation specifications can be optimized, unnecessary temporary operations or emergency operations can be reduced, and frequent abnormal operation behaviors can be identified to adjust the operation plan or conduct safety training in a timely manner.

[0040] S3: Compare real-time operation data with historical operation records to identify whether there are unplanned operations that have not been approved or exceed the prescribed operation scope. Perform intelligent analysis on real-time operation data and use the trained AI model to identify whether there are unplanned operations.

[0041] Compare the real-time operation data with the substation's historical operation records to check whether the current operation has been approved and complies with the specified scope, including comparing the operation instructions, approval process, and equipment status to determine whether the operation is an unplanned operation. Ensure that each operation is approved before execution to avoid unauthorized operations. By comparing historical records, avoid repetitive operations or erroneous operations, and identify whether there are behaviors that exceed the specified scope of operations. Real-time comparison can ensure that the operation complies with safety and operating procedures, thereby reducing risks.

[0042] The artificial intelligence model is trained using historical operation data and abnormal operation behavior data. The model identifies unplanned operations by analyzing the characteristics of the operations. During the training process, the machine learning algorithm is used to learn the historical operation data and establish a recognition model. Through the training of the AI ​​model, unplanned operations can be automatically identified, reducing the cost of manual intervention. The model can identify the difference between normal operations and unplanned operations, and can extract potential risks from the behavior patterns of operations. As the operation data increases, the model is continuously optimized, can adapt to new operation patterns, and improve the accuracy of recognition.

[0043] Intelligent analysis of real-time operation data, including automated evaluation of operation type, operator, operation duration, equipment status, and other information. The system uses a trained AI model to analyze whether real-time operations meet the predetermined operation specifications. It can identify unplanned operations in real time during the operation execution process to avoid discovering problems after the fact. Through intelligent analysis, it can automatically determine whether the operation meets the preset rules and respond in real time to reduce manual intervention. The AI ​​model can efficiently analyze complex real-time data, accurately detect potential unplanned operations, and issue alarms in a timely manner.

[0044] S4: Conduct risk assessment on unplanned operations based on the nature of the operations, scope of impact and equipment status, and determine the impact of the operations on substation equipment and systems;

[0045] Classify real-time operations and identify the nature of the operations, such as troubleshooting, equipment maintenance, system adjustments, etc. The system analyzes the possible risks based on the type of operation and determines whether it is an unplanned operation. The system can distinguish different types of operations and clarify the level of risk that each operation may bring. For example, troubleshooting may be a high-risk operation, while equipment adjustment may be a low-risk operation. The classification of the nature of the operation provides a preliminary basis for subsequent risk assessment and helps the system respond quickly.

[0046] The impact scope of unplanned operations is assessed by analyzing the equipment and areas involved in the operation. The system calculates the number of equipment involved, its importance and its role in the power system, and determines the potential impact of the operation on the overall operation of the substation. The system can assess the impact of unplanned operations on different equipment and areas in the substation, identify which operations may affect key equipment or cause large-scale failures, and assign different processing priorities to different operations based on the scope of impact, giving priority to unplanned operations that have the greatest impact on the system, helping managers fully understand the potential risks of unplanned operations and formulate response strategies in advance.

[0047] Real-time monitoring of the equipment status involved, including the equipment's health status, maintenance history, service life, load conditions, etc., evaluates whether the equipment can withstand the load of the operation and whether there are hidden dangers by integrating sensor data. Real-time acquisition of the equipment's working status ensures that the health status of the equipment can be accurately judged when there is no planned operation. The system can analyze the equipment's load and status, and evaluate whether the operation may aggravate equipment failure or cause equipment damage. Using equipment status data, the system can predict the equipment risks that may be brought about by the operation in advance, providing an accurate basis for decision-making.

[0048] The risk assessment model is established by combining the nature of the operation, the scope of impact and the status of the equipment with multi-dimensional data. The model uses a weighted scoring method to conduct risk assessment on the operation and calculate the overall risk level that the operation may bring. The assessment model is trained using machine learning methods to improve the accuracy of the assessment. By establishing a risk assessment model, the system can quantify the risk level of unplanned operations and derive an accurate risk score. Based on the risk assessment results, the system automatically generates recommended response measures, such as whether to immediately suspend the operation or whether to take additional safety checks. Through continuous learning and optimization, the model can improve the precision and accuracy of risk assessment and adapt to complex operating environments.

[0049] S5: Based on real-time data and historical data, the system conducts self-learning through deep learning algorithms to continuously optimize the accuracy and efficiency of the unplanned job recognition model. It also optimizes the recognition model and system configuration based on feedback information to improve recognition accuracy.

[0050] Deep learning algorithms such as deep neural networks (DNN), convolutional neural networks (CNN) or recurrent neural networks (RNN) are used to train historical data and establish an unplanned job recognition model. The system continuously learns the job patterns in historical data to identify and predict the occurrence of unplanned jobs. During the model training process, cross-validation, hyperparameter tuning and other technologies are used to ensure the efficiency and accuracy of training. Through the training of deep learning models, the system can self-learn and recognize the complex patterns of unplanned jobs, reducing manual intervention. Through training, the system can intelligently analyze real-time job data, accurately identify potential unplanned jobs, and improve the accuracy and efficiency of recognition. Parameters are continuously adjusted during model training to ensure that each iteration can improve the accuracy of recognition, especially the ability to accurately recognize in complex environments.

[0051] The system uses online learning and adaptive adjustment mechanisms to learn new data obtained from job feedback, system status monitoring, and alarm information in real time. Through reinforcement learning or incremental learning, the model can adapt to new environmental changes. Real-time data is combined with historical data, and the model continuously adjusts to adapt to future unplanned operation modes. The system can dynamically adjust according to real-time operation data, quickly adapt to changes in operation modes, and improve the flexibility of unplanned operation recognition. New data learned in real time helps the model identify new operation modes, thereby optimizing operation recognition accuracy. Through continuous adaptive learning, the system can continuously optimize recognition accuracy in a changing environment and reduce false positives and false negatives.

[0052] The system will obtain feedback from management personnel, equipment feedback and alarm information, and adjust the model based on the actual results after the operation is executed. If the identified unplanned operation is judged as a misjudgment or missed judgment, the model will adjust the weight and optimize the parameters according to the new feedback information. Through continuous feedback adjustment, the system can identify and correct misjudgments or missed judgments, optimize the accuracy of the unplanned operation identification model, and the feedback mechanism helps the system make up for possible deficiencies, making the model more intelligent and avoiding the influence of human factors on the accuracy of the model. Through the optimization of feedback information, the system can cope with more complex operation situations and improve the robustness and accuracy of the overall model.

[0053] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0054] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automatic verification system for unplanned substation operations based on artificial intelligence, characterized in that: It includes data acquisition module, data storage module, unplanned operation identification module, intelligent decision-making module and optimization and feedback module; The data acquisition module is used to collect the operating status, operation data, operation instructions, sensor data, and alarm information of each device in the substation in real time; The data storage module is used to store historical operation data, alarm records, equipment maintenance logs, and use big data analysis technology to conduct in-depth mining to identify potential long-term risks or trends. Through data mining, the operation management process is optimized and support is provided for future decision-making; The unplanned operation identification module performs intelligent analysis on real-time operation data based on artificial intelligence to identify whether there are unplanned operations that have not been approved or exceed the specified operation scope, and determines whether the operation is abnormal or unauthorized by comparing historical operation records, operation instructions and equipment status; After the intelligent decision-making module identifies an unplanned operation, the system will automatically generate an alarm message and push it to the relevant person in charge, and provide decision support. According to the severity of the operation and the scope of equipment impact, the risk is intelligently assessed and a response strategy is given; The optimization and feedback module is based on a deep learning algorithm, self-learns from historical data and real-time data, optimizes the accuracy and efficiency of unplanned job recognition, and gradually adjusts model parameters through feedback learning to improve system recognition capabilities.

2. The method for automatically checking unplanned substation operations based on artificial intelligence according to claim 1 is characterized in that: The following steps are involved: S1: Collect equipment operation status, operation instructions, sensor data, and alarm information in real time, and transmit the collected data to the central platform in real time through a high-speed communication network; S2: Store the collected real-time data as well as historical operation data, alarm records, and equipment maintenance logs on a central platform, and use big data analysis technology to analyze historical operation data to identify long-term risks, operation patterns, and trends; S3: Compare real-time operation data with historical operation records to identify whether there are unplanned operations that have not been approved or exceed the prescribed operation scope. Perform intelligent analysis on real-time operation data and use the trained AI model to identify whether there are unplanned operations. S4: Conduct risk assessment on unplanned operations based on the nature of the operations, scope of impact and equipment status, and determine the impact of the operations on substation equipment and systems; S5: Based on real-time data and historical data, the system conducts self-learning through deep learning algorithms to continuously optimize the accuracy and efficiency of the unplanned job recognition model. It also optimizes the recognition model and system configuration based on feedback information to improve recognition accuracy.

3. The method for automatically checking unplanned operation of a substation based on artificial intelligence according to claim 1 is characterized in that: The S1 further includes establishing communication connections with various devices and sensors in the substation, obtaining real-time operation data, operation instructions issued by the real-time automation system, collecting alarm information detected by sensors and equipment in the substation, and transmitting the collected real-time data to the central monitoring platform through a high-speed communication network.

4. The method for automatically checking unplanned substation operations based on artificial intelligence according to claim 1 is characterized in that: The S2 further includes integrating the collected real-time data, historical operation data, alarm records, and equipment maintenance logs into the data storage system of the central platform for classification and archiving, and using big data analysis technology to deeply analyze historical operation data, identify potential risks such as equipment aging, excessive operation frequency, repetitive operations, and long-term unmaintained equipment, and predict future equipment failures or safety hazards.

5. The automatic verification system and method for unplanned operation of substation based on artificial intelligence according to claim 1 is characterized in that: The S3 further includes comparing the real-time operation data with the historical operation records of the substation to check whether the current operation has been approved and is within the prescribed scope, and using the historical operation data and abnormal operation behavior data to train the artificial intelligence model. The model identifies unplanned operations by analyzing the characteristics of the operations. During the training process, the historical operation data is learned using a machine learning algorithm to establish a recognition model.

6. The automatic verification system and method for unplanned operation of substation based on artificial intelligence according to claim 1 is characterized in that: The S4 further includes classifying real-time operations, identifying the nature of the operations, evaluating the impact scope of unplanned operations by analyzing the equipment and areas involved in the operations, and establishing a risk assessment model using multi-dimensional data in combination with the nature of the operations, the impact scope and the equipment status.

7. The automatic verification system and method for unplanned operation of substation based on artificial intelligence according to claim 1 is characterized in that: The S5 further includes using a deep learning algorithm to train historical data, establish an unplanned job recognition model, and use online learning and adaptive adjustment mechanisms to learn new data obtained from job feedback, system status monitoring, and alarm information in real time. Through reinforcement learning or incremental learning, the model is adapted to new environmental changes. Real-time data is combined with historical data, and the model is continuously adjusted to adapt to future unplanned job patterns.

Images