A method for monitoring secondary circuit faults in a substation

Abstract

The present invention relates to the field of secondary circuit monitoring, and more specifically, the present invention relates to a method for monitoring faults in a secondary circuit of a substation. The method includes: obtaining the operation data of secondary circuit equipment after preprocessing, constructing an initial isolation tree based on the operation data in a preset time period, taking the operation data at future times outside the preset time period as change samples and inputting them into the initial isolation tree, and calculating the degree of change of any node after the change samples are input into the initial isolation tree; taking the nodes with a degree of change greater than a preset change threshold as research nodes, calculating the node change uniformity of the initial isolation tree according to the degree of change of the research nodes, and in response to the node change uniformity being less than a reconstruction threshold, reconstructing the initial isolation tree to obtain an updated isolation tree; calculating the anomaly score of the operation data in the updated isolation tree based on the isolation forest, and completing the fault monitoring. Through the technical solution of the present invention, the accuracy of the fault monitoring result can be improved.

Description

Technical Field

[0001] The present invention relates to the field of secondary circuit monitoring. More specifically, the present invention relates to a method for monitoring faults in the secondary circuit of a substation. Background Art

[0002] The secondary circuit generally includes a relay protection circuit, an automatic control circuit, a measurement circuit, a communication circuit, etc., which ensure the real-time monitoring of the substation, fault detection, and the stability of the power system. Although the secondary circuit seems auxiliary, its healthy operation directly affects the safety of the substation and the reliability of the power grid. Especially when a power equipment fails, the secondary circuit can quickly respond through rapid detection and automatic protection, thereby preventing the expansion of the fault.

[0003] Isolation forest is a machine learning algorithm for anomaly detection. Its core idea is to "isolate" data points by randomly partitioning the data space. The isolation forest constructs multiple trees, and each tree gradually divides the data into different regions by randomly selecting features and cut points until all data points are "isolated". Anomaly points usually require fewer partitioning steps to be isolated, so their path lengths in the tree are shorter. Compared with traditional distance- or density-based anomaly detection methods, the isolation forest has higher efficiency when dealing with large-scale data and does not depend on the distribution form of the data, being applicable to high-dimensional and sparse data sets.

[0004] The existing Chinese patent application document with the publication number CN118554450A discloses a method and device for monitoring the safety of the power grid operation state based on big data. Among them, the above application document constructs an isolation forest model for all data slice groups; and calculates the anomaly scores of each data slice group in the isolation forest model. When the anomaly score is greater than the set value, the corresponding data slice group is abnormal data, and at this time, the power grid operation state corresponding to the time sequence of the data slice group is abnormal. However, the aging of secondary circuit equipment will increase the electrical contact resistance, resulting in a collective shift of the collected data, and further leading to inaccurate anomaly detection results when using the isolation forest for anomaly detection. Summary of the Invention

[0005] To solve the problem of inaccurate anomaly monitoring results, the present invention proposes a method for monitoring faults in the secondary circuit of a substation.

[0006] The present invention discloses a method for monitoring faults in the secondary circuit of a substation, including: obtaining the operation data of secondary circuit equipment after preprocessing, constructing an initial isolation tree based on the operation data in a preset time period, taking the operation data at future moments outside the preset time period as change samples and inputting them into the initial isolation tree, and calculating the degree of change of any node after the change samples are input into the initial isolation tree; taking the nodes with a degree of change greater than a preset change threshold as research nodes, calculating the node change uniformity of the initial isolation tree according to the degree of change of the research nodes, and in response to the node change uniformity being less than a reconstruction threshold, reconstructing the initial isolation tree to obtain an updated isolation tree; calculating the anomaly score of the operation data in the updated isolation tree based on the isolation forest to complete fault monitoring.

[0007] An initial isolation tree is constructed using the operation data within a preset time period, and by inputting change samples at future moments, the degree of change of nodes is evaluated, thereby monitoring the changes in the system in real time. When the node change uniformity is lower than the set reconstruction threshold, the reconstruction of the isolation tree is automatically triggered to ensure that the tree structure can reflect the latest system dynamics and avoid a decrease in monitoring accuracy due to changes in data distribution.

[0008] Preferably, the preprocessing is to fill in the missing operation data using the mean filling method.

[0009] Preferably, the degree of change satisfies the relational expression:

[0010] , represents the node in the th layer of the initial isolation tree with the degree of change, and respectively represent the number of operation data of the node in the th layer of the initial isolation tree before and after the change samples are input into the initial isolation tree, represents the total number of nodes in the th layer of the initial isolation tree in the th layer, represents the normalization function.

[0011] It can not only measure the change range of the operation data of a single node before and after the change, but also compare the changes of different nodes to ensure that the monitoring of the equipment status has high sensitivity and accuracy. Through the quantitative processing of the degree of change, it helps to discover potential operation anomalies, and further provides guarantee for subsequent fault monitoring and improves the accuracy of the monitoring results.

[0012] Preferably, the degree of change also satisfies the relational expression:

[0013] , Represents the initial isolated tree Middle Layer Node The degree of change, and They represent the initial isolated tree before and after the sample is changed and input into the initial isolated tree. Middle Layer Node The number of running data, Represents the initial isolated tree Middle The total number of nodes in the layer, and They represent the initial isolated tree before and after the sample is changed and input into the initial isolated tree. Middle Layer Node The variance of the running data, Represents the normalization function.

[0014] By further introducing the consideration of data variance, the degree of change of each node in the initial isolated tree is comprehensively evaluated, which not only measures the relative change in the running data of the node before and after the change, but also combines the change in data volatility.

[0015] Preferably, the node change uniformity includes: for any initial isolated tree, calculating the mean change degree of all research nodes in the initial isolated tree, respectively calculating the absolute difference between the change degree of each research node in the initial isolated tree and the mean change degree, and using the cumulative value of all absolute differences through negative correlation mapping as the node change uniformity. This method can effectively reflect the uniformity of changes in each node in the tree structure.

[0016] By calculating the mean of the degree of change of all research nodes, and then evaluating the degree of deviation of each node change, if the degree of change of a node is far from the mean, the corresponding absolute difference is large. Finally, these differences are converted into the impact on the uniformity of node change through negative correlation mapping.

[0017] Preferably, the node change uniformity also satisfies the relationship:

[0018] , Represents the initial isolated tree The uniformity of node changes, Represents the initial isolated tree Research Node The depth of the layer, Represents the initial isolated tree The average layer depth of all research nodes in , Represents the total number of research nodes of the initial isolated tree , Represents the initial isolated tree in the research nodes degree of change of Represents the initial isolated tree average degree of change of all research nodes in Represents the normalization function Represents the exponential function

[0019] By calculating the deviation from the mean of the node layer depth through normalization, the distribution characteristics of the nodes in the tree structure are reflected. Then, the exponential function is used to weight the deviation from the mean of the degree of change of the nodes, enhancing the influence of the nodes with uneven degrees of change on the uniformity

[0020] Preferably, the reconstruction of the initial isolated tree to obtain the updated isolated tree includes: using the change sample that is the rightmost input into the initial isolated tree in chronological order as the operation data in the updated preset period to construct the updated isolated tree

[0021] Preferably, the completion of the fault monitoring includes: setting an anomaly threshold, and generating and sending an alarm signal in response to the average anomaly score of the operation data in a continuous preset period being greater than the anomaly threshold

[0022] Advantages of the present invention

[0023] By constructing an initial isolated tree and introducing change samples in the future period, the present invention can flexibly adapt to data changes and detect possible anomalies or faults in the loop. By calculating the degree of change and the change uniformity of each node, and reconstructing in response to nodes with too large changes, the system can effectively update the isolated tree model, thereby improving the accuracy and sensitivity of fault detection. In addition, the method also ensures that when a fault occurs, an alarm signal can be automatically triggered by setting an anomaly score threshold, helping the operation and maintenance personnel to respond quickly, reducing the outage risk and fault expansion, and improving the reliability and safety of the substation Brief Description of the Drawings

[0024] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features, and advantages of the exemplary embodiments of the present invention will become readily understood. In the drawings, several embodiments of the present invention are shown in an exemplary rather than restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0025] Figure 1 is a flowchart of a method for monitoring secondary circuit faults in a substation according to an embodiment of the present invention Detailed Embodiments

[0026] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] It should be understood that when terms such as "first" and "second" are used in the claims, the description and the drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "including" and "comprising" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0028] The present invention provides a method for monitoring faults in the secondary circuit of a substation. As Figure 1 shown, a method for monitoring faults in the secondary circuit of a substation includes steps S1 - S3, which are specifically described below.

[0029] S1, obtain the operation data of the secondary circuit equipment after pre - processing, construct an initial isolation tree according to the operation data in a preset time period, use the operation data at future moments outside the preset time period as change samples and input them into the initial isolation tree, and calculate the change degree of any node after the change samples are input into the initial isolation tree.

[0030] It should be noted that in a substation, the secondary circuit refers to an auxiliary circuit related to functions such as control, protection, monitoring, and dispatching of power equipment, and usually cannot directly carry electrical energy transmission. Secondary circuit data refers to parameters such as voltage, current, and temperature in the circuit. By monitoring the secondary circuit data in a substation, potential fault risks can be detected in a timely manner, and then measures can be taken in advance to prevent major accidents.

[0031] In one embodiment, a remote terminal unit (RTU) system is used to collect the initial operation data of the secondary circuit, and the preset collection frequency is 1 Hz. Preferably, the initial operation data includes voltage, current, and temperature.

[0032] Perform pre - processing on the initial operation data to obtain the operation data after pre - processing. The pre - processing is to fill in the missing initial operation data using the mean filling method.

[0033] It should be noted that secondary circuit equipment in a substation, such as relays, protection devices, control equipment, and communication equipment, usually needs to operate all - day long. Long - term high - load work will cause heat accumulation in electrical components and circuits, thereby accelerating the aging process of the equipment.

[0034] Since equipment aging can cause the collected data to shift. For example, the aging of equipment such as cables, terminal blocks, and relays can increase the electrical contact resistance, which in turn causes the temperature to rise. This can lead to a collective shift in the collected operation data, resulting in inaccurate anomaly detection results when using the isolation forest algorithm for anomaly detection.

[0035] Exemplarily, the preset time period is set to 10 minutes. An initial isolation tree is constructed based on the operation data within the preset time period.

[0036] It should be noted that since the initial isolation tree has been determined, the splitting threshold for each layer in the initial isolation tree has been determined. Therefore, the input of changed samples will not generate new nodes in the isolation tree.

[0037] The operation data at future times outside the preset time period is used as the changed sample and input into the initial isolation tree. Exemplarily, if the preset time period is 10 minutes, then the operation data after the 10th minute is used as the changed sample. Calculate the degree of change of any node in the initial isolation tree after the changed sample is input. The degree of change satisfies the relational expression:

[0038] , represents the node at the th layer in the initial isolation tree The degree of change of . and respectively represent the number of operation data of the node at the th layer in the initial isolation tree before and after the changed sample is input into the initial isolation tree . represents the total number of nodes at the th layer in the initial isolation tree . represents the normalization function.

[0039] The above calculation method can not only measure the change range of operation data of a single node before and after the change, but also compare the changes of different nodes, ensuring high sensitivity and accuracy in monitoring the equipment status. Through the quantitative processing of the degree of change, it helps to discover potential operation anomalies, and thus provides guarantee for subsequent fault monitoring and improves the accuracy of monitoring results.

[0040] In one embodiment, the degree of change also satisfies the relational expression:

[0041] , represents the node at the th layer in the initial isolation tree Nodes of the layer The degree of change and respectively represent the number of running data of the nodes in the th layer of the initial isolation tree before the changed sample is input and after the changed sample is input into the initial isolation tree The number of running data of the nodes in the represents the initial isolation tree th layer of the total number of nodes in the initial isolation tree and respectively represent the variance of the running data of the nodes in the th layer of the initial isolation tree before the changed sample is input and after the changed sample is input into the initial isolation tree The variance of the running data of the nodes in the represents the normalization function.

[0042] By further introducing the consideration of data variance, comprehensively evaluate the degree of change of each node in the initial isolation tree, not only measure the relative change amplitude of the running data of the node before and after the change, but also combine the change of data volatility.

[0043] One measures the change in the number of node data, and the other measures the change in the variance of the node running data, which can more comprehensively reflect the dynamic change of the node. Doing so can not only capture the change in the number of data, but also consider the change in the data fluctuation range, enhance the sensitivity and accuracy to the node change, and help to more accurately identify the abnormal fluctuation and potential risk of the device running state.

[0044] S2. Take the nodes with the degree of change greater than the preset change threshold as the research nodes, calculate the node change uniformity of the initial isolation tree according to the degree of change of the research nodes, and in response to the node change uniformity being less than the reconstruction threshold, reconstruct the initial isolation tree to obtain an updated isolation tree.

[0045] In one embodiment, obtain the research nodes. Exemplarily, the preset change threshold is set to 0.7.

[0046] Calculate the node change uniformity of the initial isolation tree according to the degree of change of the research nodes. The node change uniformity includes: for any initial isolation tree, calculate the mean value of the degree of change of all research nodes in the initial isolation tree, calculate the absolute difference between the degree of change of each research node in the initial isolation tree and the mean value of the degree of change respectively, and take the result of the cumulative value of all absolute differences through negative correlation mapping as the node change uniformity.

[0047] It should be noted that as the change samples are continuously input, the node features in the initial isolation tree will change. If the degree of change of each node in the initial isolation tree is large, it indicates that there is a fault in the secondary circuit equipment, resulting in large fluctuations in the data points. If the nodes with large changes in the isolation tree are concentrated in a certain part of the isolation tree, it indicates that due to equipment aging, the collected data drifts. At this time, it indicates that the initial isolation tree no longer meets the data distribution characteristics and needs to be reconstructed.

[0048] Exemplarily, in response to the node change uniformity being less than the reconstruction threshold, and the reconstruction threshold being set to 0.4, reconstructing the initial isolation tree to obtain an updated isolation tree includes: using the change sample that is input into the initial isolation tree on the rightmost side in chronological order as the operation data in the updated preset period to construct the updated isolation tree.

[0049] In one embodiment, the node change uniformity also satisfies the relational expression:

[0050] , represents the node change uniformity of the initial isolation tree , represents the depth of the research node in the initial isolation tree , represents the average depth of all research nodes in the initial isolation tree , represents the total number of research nodes of the initial isolation tree , represents the degree of change of the research node in the initial isolation tree , represents the average value of the degree of change of all research nodes in the initial isolation tree , represents the normalization function represents the exponential function.

[0051] By normalizing and calculating the deviation of the node depth from the mean, the distribution characteristics of the nodes in the tree structure are reflected. Then, the exponential function is used to weight the deviation of the node change degree, enhancing the influence of the nodes with uneven change degrees on the uniformity. This can not only measure the uniformity of the node depth but also comprehensively consider the amplitude and volatility of the node changes, helping to identify those nodes with large or uneven changes after the change, thus effectively reflecting the overall change uniformity of the initial isolation tree.

[0052] S3. Based on the isolation forest, calculate the anomaly score of the operation data in the updated isolation tree to complete the fault monitoring.

[0053] In one embodiment, the anomaly score of the running data in the isolation tree is updated based on the isolation forest. Calculating the anomaly score is a well-known technique to those skilled in the art and will not be elaborated here.

[0054] An anomaly threshold is set. In response to the average anomaly score of the running data in a continuous preset period being greater than the anomaly threshold, an alarm signal is generated and sent.

[0055] Exemplarily, the anomaly threshold is set to 0.8.

[0056] Although this specification has shown and described multiple embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will think of many changes, alterations, and alternative ways without departing from the spirit and concept of the present invention. It should be understood that various alternative solutions of the embodiments of the present invention described herein can be adopted in the practice of the present invention.

[0057] The above are all preferred embodiments of the present invention, and the protection scope of the present invention is not limited thereby. Therefore, all equivalent changes made according to the structure, shape, and principle of the present invention shall be covered within the protection scope of the present invention.

Claims

1. A method for monitoring secondary circuit faults in a substation, characterized in that: include: Obtain the pre-processed operating data of the secondary circuit equipment, build an initial isolated tree based on the operating data in a preset period, input the operating data at future times outside the preset period as a change sample into the initial isolated tree, and calculate the degree of change of any node after the change sample is input into the initial isolated tree; Nodes whose degree of change is greater than a preset change threshold are taken as research nodes, and the node change uniformity of the initial isolated tree is calculated according to the degree of change of the research nodes. In response to the node change uniformity being less than a reconstruction threshold, the initial isolated tree is reconstructed to obtain an updated isolated tree; Based on the isolation forest calculation, the abnormal scores of the running data in the isolation tree are updated to complete the fault monitoring; The node change uniformity includes: for any initial isolated tree, calculating the mean change degree of all research nodes in the initial isolated tree, respectively calculating the absolute difference between the change degree of each research node in the initial isolated tree and the mean change degree, and using the cumulative value of all absolute differences through negative correlation mapping as the node change uniformity; The reconstructing the initial isolated tree to obtain the updated isolated tree includes: using the rightmost change sample input into the initial isolated tree in time sequence as the running data in the updated preset time period to construct the updated isolated tree.

2. A substation secondary circuit fault monitoring method according to claim 1, characterized in that: The preprocessing is to fill in the missing running data using the mean filling method.

3. A substation secondary circuit fault monitoring method according to claim 1, characterized in that: The degree of change satisfies the relationship: , Represents the initial isolated tree Middle Layer Node The degree of change, and They represent the initial isolated tree before and after the sample is changed and input into the initial isolated tree. Middle Layer Node The number of running data, Represents the initial isolated tree Middle The total number of nodes in the layer, Represents the normalization function.

4. A substation secondary circuit fault monitoring method according to claim 1, characterized in that: The degree of change also satisfies the relationship: , Represents the initial isolated tree Middle Layer Node The degree of change, and They represent the initial isolated tree before and after the sample is changed and input into the initial isolated tree. Middle Layer Node The number of running data, Represents the initial isolated tree Middle The total number of nodes in the layer, and They represent the initial isolated tree before and after the sample is changed and input into the initial isolated tree. Middle Layer Node The variance of the running data, Represents the normalization function.

5. A substation secondary circuit fault monitoring method according to claim 1, characterized in that: The completion of fault monitoring comprises: An abnormality threshold is set, and in response to the average abnormality score of the running data in a continuous preset period being greater than the abnormality threshold, an alarm signal is generated and sent.

Images