A real-time monitoring method and system for power system network security

Abstract

The present invention provides a method and system for real-time monitoring of power system network security, comprising: a first fault diagnosis unit for performing initial fault diagnosis based on operating data and outputting the initial fault diagnosis result; a segmentation processing unit for segmenting the operating data based on the initial fault diagnosis result and outputting segmented data; a second fault diagnosis unit for performing detailed fault diagnosis based on the segmented data and outputting the detailed fault diagnosis result; and a fault location unit for outputting the minimum fault operating data segment. The present invention can accurately locate the specific time when a fault occurs, thereby shortening troubleshooting time for operation and maintenance personnel. It can also more quickly predict faults, prevent sudden equipment failures, reduce downtime, and ensure the safe and stable operation of the power system.

Description

Technical Field

[0001] The present invention relates to an electric power system, and in particular to a method and system for real-time monitoring of electric power system network security. Background Art

[0002] Power system equipment monitoring and fault diagnosis are crucial and important in ensuring the safe and stable operation of power systems. Power system equipment monitoring collects various operational data (such as current, voltage, temperature, and vibration) in real time, enabling accurate understanding of the equipment's operating status. This helps proactively identify potential problems, prevent failures, ensure the continued efficient operation of the power system, and reduce sudden power outages. Power system equipment fault diagnosis analyzes equipment operating data and fault characteristics after a fault occurs to quickly locate the source of the fault and determine its type and severity. For example, patent document CN111488947A obtains power fingerprint data, voiceprint data, temperature data, and vibration data from power system equipment. It first determines the device's operating mode based on the power fingerprint data. Then, using a pre-trained fault detection model for that mode, it performs fault detection on the device based on the voiceprint, temperature, and vibration data. This differentiated processing is tailored to the characteristics of different operating modes, improving the accuracy of equipment fault detection. Patent document CN114993383A utilizes intelligent algorithms to confirm distribution room status variables, significantly improving the accuracy of distribution room status assessment and the reliability of fault warnings. By using intelligent algorithms to calculate the transformer's state quantities, a more comprehensive assessment of its status is conducted, moving beyond the limitations of single state quantity analysis and enhancing the accuracy of transformer fault warnings. The intelligent algorithm for low-voltage switchgear can confirm its status from a more macroscopic perspective, significantly improving the timeliness and accuracy of switchgear status assessments. However, existing real-time monitoring of power system network security has the following shortcomings: Existing monitoring methods only issue fault alarms after power system equipment fails, but cannot determine the specific moment of the failure, requiring operations and maintenance personnel to spend a significant amount of time troubleshooting the cause of the failure. Fault prediction models use fault data for prediction, resulting in untimely fault predictions, which can lead to sudden equipment failure, even threatening personnel safety and causing significant economic losses. Summary of the Invention

[0003] In view of the problems existing in the prior art, the present invention provides a method and system for real-time monitoring of power system network security.

[0004] In one aspect, the present invention provides a power system network security real-time monitoring system, comprising:

[0005] Data acquisition unit, used to collect operating data and monitoring data of power system equipment;

[0006] a first fault diagnosis unit, configured to perform initial fault diagnosis based on the operating data and output an initial fault diagnosis result;

[0007] A segmentation processing unit, configured to segment the operating data according to the initial fault diagnosis result and output segmented data;

[0008] a second fault diagnosis unit, configured to perform detailed fault diagnosis based on the segmented data and output a detailed fault diagnosis result;

[0009] The fault location unit is used to determine whether to adjust the number of segments based on the detailed fault diagnosis result; if the number of segments does not need to be adjusted, the minimum fault operation data segment where the fault occurred is output; otherwise, the number of segments is adjusted through the segment processing unit, and the second fault diagnosis unit performs detailed fault diagnosis based on the adjusted number of segments and outputs the detailed fault diagnosis result after the adjustment of the number of segments until the number of segments does not need to be adjusted.

[0010] Furthermore, the system further comprises a monitoring processing unit, which is configured to segment the monitoring data according to the minimum fault operation data segment and output the minimum fault monitoring data segment.

[0011] Furthermore, the specific process of segmenting the operating data according to the fault diagnosis result is: dividing the operating data into equal parts;

[0012] The specific process of adjusting the number of segments by the segment processing unit is: increasing the number of segments into which the operating data is equally divided by one segment.

[0013] Furthermore, the system further includes a fault prediction unit, which is used to predict potential faults based on the operating data.

[0014] Furthermore, the fault prediction unit is trained and updated based on an operation data segment preceding the minimum fault operation data segment.

[0015] In another aspect, the present invention provides a method for real-time monitoring of power system network security, which is applied to the above-mentioned system, and comprises the following steps:

[0016] Step S1: The data acquisition unit collects the operating data D and monitoring data M of the power system equipment, and the length of the operating data D and the monitoring data M are both n;

[0017] Step S2: the first fault diagnosis unit performs initial fault diagnosis on the operating data D and outputs an initial fault diagnosis result R;

[0018] Step S3: The segmentation processing unit divides the operation data D into k segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2, ..., D k ;

[0019] Step S4: The second fault diagnosis unit analyzes the segmented data D1, D2, ..., D k Perform detailed fault diagnosis and output detailed fault diagnosis results R1, R2, ..., R k ;

[0020] Step S5: The fault location unit locates the fault according to the detailed fault diagnosis results R1, R2, ..., R k Determine whether to adjust the number of segments k; if the number of segments k does not need to be adjusted, then output the minimum fault operation data segment D where the fault occurs F Otherwise, the number of segments is adjusted by the segment processing unit, and the second fault diagnosis unit performs detailed fault diagnosis according to the adjusted number of segments and outputs the detailed fault diagnosis result after the adjusted number of segments until the number of segments does not need to be adjusted.

[0021] Furthermore, the method further comprises step S6: the monitoring processing unit is used to segment the monitoring data according to the minimum fault operation data segment, and output the minimum fault monitoring data segment M. F .

[0022] Furthermore, in step S3, the specific process of the segmentation processing unit dividing the operating data D into k segments according to the fault diagnosis result R is as follows: the operating data D is divided into k segments: D1, D2, ..., D k , where each segment D i The size of

[0023] In step S5, the specific process of adjusting the number of segments by the segment processing unit is: increasing the value of the number k of segments into which the operating data is equally divided by 1.

[0024] Furthermore, the method further includes step S7: a fault prediction unit is used to predict potential faults based on the operating data.

[0025] Furthermore, in step S7, the fault prediction unit is based on the minimum fault operation data segment D F The previous running data segment D F-1 Trained and updated.

[0026] The present invention provides a real-time monitoring method and system for power system network security, comprising: a data acquisition unit, a first fault diagnosis unit, a segmentation processing unit, a second fault diagnosis unit, and a fault location unit. The minimum fault operation data segment output by the fault location unit can accurately locate the specific time when the fault occurs, thereby shortening the troubleshooting time of operation and maintenance personnel and improving operation and maintenance efficiency. In addition, the fault prediction unit in the present invention is trained and updated based on the operation data segment preceding the minimum fault operation data segment, enabling more rapid fault prediction, preventing sudden equipment failure, reducing downtime, ensuring the safe and stable operation of the power system, and effectively reducing equipment maintenance costs and socioeconomic risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0028] Figure 1 The structure of a real-time monitoring system for power system network security provided by the embodiment of the present invention Figure 1 ;

[0029] Figure 2 The structure of a real-time monitoring system for power system network security provided by the embodiment of the present invention Figure 2 ;

[0030] Figure 3 The structure of a real-time monitoring system for power system network security provided by the embodiment of the present invention Figure 3 ;

[0031] Figure 4 The structure of a real-time monitoring system for power system network security provided by the embodiment of the present invention Figure 4 ;

[0032] Figure 5 The present invention provides a flowchart of a method for real-time monitoring of power system network security according to an embodiment of the present invention. DETAILED DESCRIPTION

[0033] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiment is only a unit embodiment of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0034] As an embodiment of the present invention, Figure 1 As shown, this embodiment provides a power system network security real-time monitoring system, including:

[0035] Data acquisition unit, used to collect operating data and monitoring data of power system equipment;

[0036] Specifically, the power system equipment includes generators, frequency converters, inverters, transformers, mutual inductors, switchgear, etc. The operational data includes current, voltage, power, temperature, humidity, temperature rise, vibration, etc. Key power system equipment is also equipped with video surveillance to obtain monitoring data.

[0037] a first fault diagnosis unit, configured to perform initial fault diagnosis based on the operating data and output an initial fault diagnosis result;

[0038] Specifically, the first fault diagnosis unit adopts a machine learning model, which includes a support vector machine, a convolutional neural network, Q-Learning, and deep learning. It uses historical diagnostic data for training, performs initial fault diagnosis based on real-time operating data, and outputs initial fault diagnosis results.

[0039] A segmentation processing unit, configured to segment the operating data according to the initial fault diagnosis result and output segmented data;

[0040] a second fault diagnosis unit, configured to perform detailed fault diagnosis based on the segmented data and output a detailed fault diagnosis result;

[0041] Specifically, the second fault diagnosis unit adopts a machine learning model, which includes support vector machine, convolutional neural network, Q-Learning, and deep learning.

[0042] The fault location unit is used to determine whether to adjust the number of segments based on the detailed fault diagnosis result; if the number of segments does not need to be adjusted, the minimum fault operation data segment where the fault occurred is output; otherwise, the number of segments is adjusted through the segment processing unit, and the second fault diagnosis unit performs detailed fault diagnosis based on the adjusted number of segments and outputs the detailed fault diagnosis result after the adjustment of the number of segments until the number of segments does not need to be adjusted.

[0043] Specifically, the initial fault diagnosis result is that a fault exists, the segmentation processing unit divides the operating data into two segments, the second fault diagnosis unit performs detailed fault diagnosis based on the two segments of segmented data, and outputs a detailed fault diagnosis result; if the detailed fault diagnosis result is that a fault exists, the segmentation processing unit divides the operating data into three segments, the second fault diagnosis unit performs detailed fault diagnosis based on the three segments of segmented data, and outputs a detailed fault diagnosis result; if the detailed fault diagnosis result is that no fault exists, the data segment where the fault exists when the operating data is divided into two segments is output as the minimum fault operating data segment where the fault occurs.

[0044] Furthermore, if Figure 2 As shown, the system further includes a monitoring processing unit, which is used to segment the monitoring data according to the minimum fault operation data segment and output the minimum fault monitoring data segment.

[0045] Specifically, for some longer operating data, the operation and maintenance personnel need to retrieve a large amount of monitoring videos for investigation after discovering a fault. The present invention can accurately locate the minimum fault operating data segment through the segmented processing unit combined with the first fault diagnosis unit and the second fault diagnosis unit, and can determine the corresponding minimum fault monitoring data segment, thereby shortening the fault investigation time of the operation and maintenance personnel and improving the operation and maintenance efficiency.

[0046] Furthermore, the specific process of segmenting the operating data according to the fault diagnosis result is: dividing the operating data into equal parts;

[0047] The specific process of adjusting the number of segments by the segment processing unit is: increasing the number of segments into which the operating data is equally divided by one segment.

[0048] Furthermore, if Figure 3 As shown, the system further includes a fault prediction unit, which is used to predict potential faults based on the operating data.

[0049] Furthermore, the fault prediction unit is trained and updated based on an operation data segment preceding the minimum fault operation data segment.

[0050] Specifically, existing power system fault prediction systems use fault data to establish a fault prediction unit. While this method can provide high prediction accuracy, it suffers from a certain degree of hysteresis, making it impossible to perform effective preventive maintenance on the power system. Faults can only be addressed passively, impacting the grid's operational efficiency. However, before a fault occurs, the operating data of power system equipment may contain some fault signs. The fault prediction unit of the present invention is trained and updated based on the operating data segment preceding the minimum fault operating data segment. This allows for timely prediction of potential faults in power system equipment, preventing sudden equipment failures, reducing downtime, and ensuring the safe and stable operation of the power system.

[0051] As another embodiment of the present invention, Figure 4 As shown, this embodiment provides a power system network security real-time monitoring system, including:

[0052] Data acquisition unit, used to collect operating data and monitoring data of power system equipment;

[0053] a first fault diagnosis unit, configured to perform initial fault diagnosis based on the operating data and output an initial fault diagnosis result;

[0054] A segmentation processing unit, configured to divide the operating data into equal parts according to the initial fault diagnosis result and output segmented data;

[0055] a second fault diagnosis unit, configured to perform detailed fault diagnosis based on the segmented data and output a detailed fault diagnosis result;

[0056] The fault location unit is used to determine whether to adjust the number of segments based on the detailed fault diagnosis result; if the number of segments does not need to be adjusted, the minimum fault operation data segment where the fault occurred is output; otherwise, the number of segments into which the operation data is equally divided is increased by one segment, and the second fault diagnosis unit performs detailed fault diagnosis based on the adjusted number of segments and outputs the detailed fault diagnosis result after the number of segments is adjusted until the number of segments does not need to be adjusted.

[0057] A monitoring processing unit is used to segment the monitoring data according to the minimum fault operation data segment and output the minimum fault monitoring data segment.

[0058] A fault prediction unit is used to predict potential faults according to the operation data, and the fault prediction unit is trained and updated according to an operation data segment preceding the minimum fault operation data segment.

[0059] On the other hand, Figure 5 As shown, the present invention provides a method for real-time monitoring of power system network security, which is applied to the above-mentioned system, and the method includes the following steps:

[0060] Step S1: The data acquisition unit collects the operating data D and monitoring data M of the power system equipment, and the length of the operating data D and the monitoring data M are both n;

[0061] Step S2: the first fault diagnosis unit performs initial fault diagnosis on the operating data D and outputs an initial fault diagnosis result R;

[0062] Specifically, the first fault diagnosis unit adopts a machine learning model, which includes support vector machine, convolutional neural network, Q-Learning, and deep learning.

[0063] Step S3: The segmentation processing unit divides the operation data D into k segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2, ..., D k ;

[0064] Step S4: The second fault diagnosis unit analyzes the segmented data D1, D2, ..., D k Perform detailed fault diagnosis and output detailed fault diagnosis results R1, R2, ..., R k ;

[0065] Specifically, the second fault diagnosis unit adopts a machine learning model, which includes support vector machine, convolutional neural network, Q-Learning, and deep learning.

[0066] More specifically, the second fault diagnosis unit selects different machine learning models according to the change of k. In fault diagnosis, the choice of machine learning models has a significant impact on accuracy, especially when processing data of different lengths. Data sets of different lengths usually contain different amounts of information and feature complexity. Therefore, it is crucial to select a suitable machine learning model for each data length. For data with a small sample size, a simple model can effectively avoid overfitting and provide more reliable diagnostic results. As the amount of data increases, more complex models can better capture the potential patterns in the data, thereby improving accuracy. Flexible adjustment of machine learning models to adapt to data of different lengths can not only optimize model performance, but also effectively improve the accuracy of fault diagnosis. The application of this strategy in fields such as power systems is particularly important, providing strong support for ensuring equipment safety and system stability.

[0067] Step S5: The fault location unit locates the fault according to the detailed fault diagnosis results R1, R2, ..., R k Determine whether to adjust the number of segments k; if the number of segments k does not need to be adjusted, then output the minimum fault operation data segment D where the fault occurs FOtherwise, the number of segments is adjusted by the segment processing unit, and the second fault diagnosis unit performs detailed fault diagnosis according to the adjusted number of segments and outputs the detailed fault diagnosis result after the adjusted number of segments until the number of segments does not need to be adjusted.

[0068] Furthermore, the fault location unit locates the fault according to the detailed fault diagnosis results R1, R2, ..., R k The specific process of determining whether to adjust the number of segments k is as follows: If there is segment data D i , i∈{1,2,…,k}, such that R i =1, then the number of segments k needs to be adjusted; if for all segment data D i , i∈{1,2,…,k} such that all R i =0, there is no need to adjust the number of segments.

[0069] Specifically, for example, the initial fault diagnosis result R=1, 1 indicates a fault, that is, the operating data D has a fault; the segmentation processing unit divides the operating data D into 2 segments according to the initial fault diagnosis result R, and outputs segmented data D1 and D2; the second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1 and D2, and outputs detailed fault diagnosis results R1 and R2; if R1=0 and R2=1, it indicates that the segmented data D2 has a fault; the segmentation processing unit divides the operating data D into 3 segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2 and D3; the second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1, D2 and D3, and outputs detailed fault diagnosis results R1, R2 and R3; if R1=0, R2=0 and R3=0, it indicates that the segmentation is terminated, and the minimum fault operating data segment D where the fault occurs is output. F It is the segmented data D2 when the operating data D is divided into two segments.

[0070] Specifically, for example, the initial fault diagnosis result R=1, 1 indicates a fault, that is, the operating data D has a fault; the segmentation processing unit divides the operating data D into 2 segments according to the initial fault diagnosis result R, and outputs segmented data D1 and D2; the second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1 and D2, and outputs detailed fault diagnosis results R1 and R2; if R1=0, R2=1, it indicates that the segmented data D2 has a fault; the segmentation processing unit divides the operating data D into 3 segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2, and D3; the second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1, D2, and D3. D2, D3 perform detailed fault diagnosis and output detailed fault diagnosis results R1, R2, R3; if R1 = 0, R2 = 0, R3 = 1, it means that the segmented data D3 has a fault; the segmentation processing unit divides the operating data D into 4 segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2, D3, D4. The second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1, D2, D3, D4 and outputs detailed fault diagnosis results R1, R2, R3, R4; if R1 = 0, R2 = 0, R3 = 0, R4 = 0, it means that the segmentation is terminated and the minimum fault operating data segment D where the fault occurs is output. F It is the segmented data D3 when the operating data D is divided into 3 segments.

[0071] Specifically, for example, the initial fault diagnosis result R=1, 1 indicates a fault, that is, the operating data D has a fault; the segmentation processing unit divides the operating data D into 2 segments according to the initial fault diagnosis result R, and outputs segmented data D1 and D2; the second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1 and D2, and outputs detailed fault diagnosis results R1 and R2; if R1=0, R2=1, it indicates that the segmented data D2 has a fault; the segmentation processing unit divides the operating data D into 3 segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2, and D3; the second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1, D2, and D3. D2, D3 perform detailed fault diagnosis and output detailed fault diagnosis results R1, R2, R3; if R1 = 0, R2 = 1, R3 = 0, it means that the segmented data D2 has a fault; the segmentation processing unit divides the operating data D into 4 segments according to the initial fault diagnosis result R, and outputs segmented data D1, D2, D3, D4. The second fault diagnosis unit performs detailed fault diagnosis on the segmented data D1, D2, D3, D4 and outputs detailed fault diagnosis results R1, R2, R3, R4; if R1 = 0, R2 = 0, R3 = 0, R4 = 0, it means that the segmentation is terminated and the minimum fault operating data segment D where the fault occurs is output. F It is the segmented data D2 when the operating data D is divided into 3 segments.

[0072] Furthermore, the method further comprises step S6: the monitoring processing unit is used to segment the monitoring data according to the minimum fault operation data segment, and output the minimum fault monitoring data segment M. F .

[0073] Specifically, when processing longer operating data, operation and maintenance personnel often need to retrieve a large amount of monitoring videos to troubleshoot. To this end, the present invention proposes a new method that combines a segmented processing unit with a first fault diagnosis unit and a second fault diagnosis unit to achieve accurate positioning of fault data. This method can accurately identify the smallest fault operating data segment and find the corresponding minimum fault monitoring data segment. This innovation not only greatly shortens the troubleshooting time of operation and maintenance personnel, but also significantly improves operation and maintenance efficiency. By quickly locating the source of the fault, the operation and maintenance team can perform maintenance and repairs more efficiently, thereby ensuring the stable operation of the equipment.

[0074] Furthermore, in step S3, the specific process of the segmentation processing unit dividing the operating data D into k segments according to the fault diagnosis result R is as follows: the operating data D is divided into k segments: D1, D2, ..., D k , where each segment D i The size of

[0075] In step S5, the specific process of adjusting the number of segments by the segment processing unit is: increasing the value of the number k of segments into which the operating data is equally divided by 1.

[0076] Furthermore, the method further includes step S7: a fault prediction unit is used to predict potential faults based on the operating data.

[0077] Furthermore, in step S7, the fault prediction unit is based on the minimum fault operation data segment D F The previous running data segment D F-1 Trained and updated.

[0078] Fault prediction in power systems primarily relies on fault data to build prediction models. While this approach offers good accuracy, it often suffers from a time delay, hindering effective preventative maintenance and ultimately forcing reactive response to faults, impacting the overall operational efficiency of the power grid. However, before a fault occurs, the operating data of power system equipment often exhibits some fault symptoms, providing an opportunity for early warning. The fault prediction unit of the present invention focuses on extracting these potential fault symptoms, specifically by analyzing the preceding operating data segment of the minimum fault data segment for training and updating. This approach enables the system to more promptly identify potential equipment faults, resulting in more accurate predictions. Compared to traditional methods, this symptom-based prediction mechanism enables faster and more accurate fault identification, significantly improving the maintenance efficiency and reliability of the power system. The ability to make timely predictions not only effectively prevents sudden equipment failures but also significantly reduces downtime caused by faults. This innovation enables the power system to take appropriate measures before a fault occurs, thereby ensuring safe and stable system operation. Overall, by extracting fault symptoms, the present invention improves the timeliness and accuracy of predictions, providing strong support for efficient power system management.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A real-time monitoring system for power system network security, characterized in that: include: Data acquisition unit, used to collect operating data and monitoring data of power system equipment; a first fault diagnosis unit, configured to perform initial fault diagnosis based on the operating data and output an initial fault diagnosis result; A segmentation processing unit, configured to segment the operating data according to the initial fault diagnosis result and output segmented data; The specific process of segmenting the operating data according to the fault diagnosis result is as follows: dividing the operating data into equal parts; a second fault diagnosis unit, configured to perform detailed fault diagnosis based on the segmented data and output a detailed fault diagnosis result; a fault location unit, configured to determine whether to adjust the number of segments based on the detailed fault diagnosis result; and output the minimum fault operation data segment where the fault occurred if the number of segments does not need to be adjusted; Otherwise, the number of segments is adjusted through the segmentation processing unit, and the second fault diagnosis unit performs detailed fault diagnosis based on the adjusted number of segments, and outputs the detailed fault diagnosis results after the number of segments is adjusted until the number of segments does not need to be adjusted; the specific process of adjusting the number of segments through the segmentation processing unit is: increasing the number of segments into which the operating data is equally divided by one segment.

2. The power system network security real-time monitoring system according to claim 1 is characterized by: The system further includes a monitoring processing unit configured to segment the monitoring data according to the minimum fault operation data segment and output the minimum fault monitoring data segment.

3. The power system network security real-time monitoring system according to claim 1 is characterized by: The system further includes a fault prediction unit, which is configured to perform potential fault prediction based on the operation data.

4. The power system network security real-time monitoring system according to claim 3 is characterized by: The fault prediction unit is trained and updated according to an operation data segment preceding the minimum fault operation data segment.

5. A method for real-time monitoring of power system network security, applied to the system according to any one of claims 1 to 4, characterized in that: The method comprises the following steps: Step S1: The data acquisition unit collects the operating data of the power system equipment and monitoring data , running data and monitoring data The length of ; Step S2: The first fault diagnosis unit analyzes the operating data Perform initial fault diagnosis and output initial fault diagnosis results ; Step S3: The segment processing unit performs the following steps based on the initial fault diagnosis result: The operating data Divided into Segment, output segmented data ; Segment processing unit according to the fault diagnosis results The operating data Divided into The specific process of the segment is: run data Divide into part: , where each segment The size is ; Step S4: The second fault diagnosis unit analyzes the segmented data Perform detailed fault diagnosis and output detailed fault diagnosis results ; Step S5: The fault location unit determines the fault location based on the detailed fault diagnosis result. Determine whether to adjust the number of segments ; If you do not need to adjust the number of segments , then the minimum fault operation data segment where the fault occurs is output Otherwise, the segment number is adjusted by the segment processing unit, and the second fault diagnosis unit performs detailed fault diagnosis according to the adjusted segment number, and outputs the detailed fault diagnosis result after the segment number is adjusted, until the segment number does not need to be adjusted; the specific process of adjusting the segment number by the segment processing unit is: the number of segments into which the operating data is equally divided The value of is increased by 1.

6. The method for real-time monitoring of power system network security according to claim 5, characterized in that: The method further comprises step S6: the monitoring processing unit is used to segment the monitoring data according to the minimum fault operation data segment and output the minimum fault monitoring data segment. .

7. The method for real-time monitoring of power system network security according to claim 5, characterized in that: The method further comprises step S7: a fault prediction unit is configured to perform potential fault prediction based on the operating data.

8. The method for real-time monitoring of power system network security according to claim 7, characterized in that: In step S7, the fault prediction unit is based on the minimum fault operation data segment The previous running data segment Trained and updated.

Images