Power equipment digital monitoring system based on power system

Abstract

The invention discloses a power equipment digital monitoring system based on a power system, and relates to the technical field of power grid information monitoring. Comprising a data acquisition module, an item set acquisition module, a frequent set marking module, an associated node marking module, a reference operation vector acquisition module, a deviation base angle acquisition module, a real-time deviation angle acquisition module and an interference node output module. The technical problem that the fault condition of a fault node in a power system may be misjudged due to data fluctuation of other nodes, and the data fluctuation of the other nodes is ignored due to the fact that the data fluctuation does not reach a set limit value, so that the fault node is wrongly marked is solved. According to the method, the interference node associated with the abnormal node can be checked preferentially, so that the possibility that the fault node is misjudged by data fluctuation of other nodes is reduced, and meanwhile, the situation that the data fluctuation of other nodes is neglected due to the fact that the data fluctuation does not reach a set limit value, and consequently the fault node is marked mistakenly is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid information monitoring, and in particular to a digital monitoring system for power equipment based on a power system. Background Art

[0002] The intelligent power equipment monitoring system is a system that realizes real-time monitoring and data processing of power grid operation status, equipment operation parameters, power supply quality, etc. through integrated sensors, communication equipment and information processing systems. Its purpose is to improve the reliability, safety and operation efficiency of the power system. It involves monitoring various aspects of the power system, such as voltage, current, frequency and other parameters, as well as detecting and predicting potential faults or abnormal conditions. It is an important tool for improving the level of intelligence, informationization and automation of the power grid, and can provide strong support for the stable operation and optimized management of the power grid.

[0003] However, when monitoring the operating data of each node in the power grid, there is a problem of mutual influence between the data of each node. Specifically, the fault condition of the faulty node may be misjudged due to data fluctuations of other nodes, and the data fluctuations of other nodes are ignored because they do not reach the set limit, resulting in incorrect marking of the faulty node, which is not conducive to the troubleshooting of the faulty node. Based on this, a digital monitoring system for power equipment based on the power system is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a digital monitoring system for power equipment based on an electric power system, which solves the technical problem that the fault condition of a faulty node in the electric power system may be misjudged due to data fluctuations of other nodes, and the data fluctuations of other nodes are ignored because they do not reach the set limit, thereby causing incorrect marking of the faulty node.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The digital monitoring system of power equipment based on the power system includes:

[0007] Data acquisition module: responsible for collecting the operating data of each node in the power system, namely voltage, current and frequency data;

[0008] Frequent set marking module: Analyze the running data of each node to obtain the item sets of different nodes, calculate the frequency of item sets, and mark the item sets whose frequency exceeds the threshold as frequent sets;

[0009] The associated node marking module extracts other nodes in the frequent set as the associated nodes of the node;

[0010] Benchmark operation vector acquisition module: vectorizes the historical operation data of each node, calculates the mean of the operation vector within a preset number of times, and obtains the benchmark operation vector of each node;

[0011] Deviation base angle acquisition module: performs standard deviation processing on the angle between the running vector of each node and the reference running vector to determine the deviation base angle of each node;

[0012] Real-time deviation angle acquisition module: calculates the real-time deviation angle between the benchmark operation vector of the abnormal node and the real-time operation vectors of all associated nodes;

[0013] Interference node output module: compares the real-time deviation angle of the associated node with the deviation base angle of the abnormal node to obtain the interference node of the abnormal node.

[0014] As a further solution of the present invention: the specific method of obtaining the benchmark operation vector corresponding to each node is:

[0015] The voltage, current and frequency in the operating data are marked as V, I and F respectively, and the historical operating data of each node in the preset number of times n are vectorized into the operating vector Een (Ven, Ien, Fen), and the average of the operating vectors Een (Ven, Ien, Fen) corresponding to each node in the preset number of times n is used as the benchmark operating vector Ge (Vep, Iep, Fep) corresponding to each node, where e is the number of nodes in the power system, n is the preset number, e≥1, n≥1.

[0016] As a further solution of the present invention: the specific method of obtaining the deviation base angle corresponding to each node is:

[0017] S01: Calculate the modulus length of the benchmark running vector of each node and the modulus length of the running vector corresponding to each node in the preset number of times n by using the vector modulus length calculation formula, select one node from each node as the target node, obtain the modulus length of the benchmark running vector of the target node marked as ||G1||, and the modulus length of the running vector corresponding to the target node in the preset number of times n marked as ||E1 n||, obtain the cosine value cos(Eθn) between each running vector and the benchmark running vector by using the cosine formula, and then use the inverse cosine function to obtain the deviation angle Eθn between each running vector and the benchmark running vector, and then perform standard deviation processing on the deviation angle Eθn to obtain the standard value H of the deviation angle Eθn, compare and analyze the standard value H with the preset value Y2, and obtain the deviation base angle P1 of the target node;

[0018] S02: Finally, repeat step S01 to obtain the deviation base angle Pe of each node.

[0019] As a further solution of the present invention: the specific method of obtaining the deviation base angle of the target node is:

[0020] When the standard value H is less than the preset value Y2, the mean Eθp of the deviation angle Eθn is used as the deviation base angle P1 of the target node. When the standard value H is greater than or equal to the preset value Y2, the mean of the maximum and minimum values ​​of the deviation angle Eθn is used as the deviation base angle P1 of the target node.

[0021] As a further solution of the present invention: the specific method of obtaining the real-time deviation angle between the reference operation vector of the abnormal node and the real-time operation vectors of all associated nodes is:

[0022] The real-time operation data of all associated nodes of the abnormal node are quantified into a real-time operation vector Kr (LAr, LBr, LCr). The modulus length ||Kr|| corresponding to the real-time operation vector of each associated node is calculated by the vector modulus length calculation formula. The real-time deviation angle Kθr between the real-time operation vector of each associated node and the abnormal node benchmark operation vector is calculated by the arc cosine function, where r is the number of associated nodes corresponding to the abnormal node, e≥r≥1.

[0023] As a further solution of the present invention: the specific method of obtaining the interference node of the abnormal node is:

[0024] The associated nodes satisfying |Kθr-PA|≥Y3 are marked as interference nodes of the abnormal nodes, and the interference nodes are bound to the abnormal nodes and output at the same time, where PA is the deviation base angle corresponding to the abnormal node and Y3 is the preset value.

[0025] As a further solution of the present invention: the specific method of obtaining the deviation angle between each running vector and the reference running vector is:

[0026] Through the cosine formula cos(Eθn)=||G1|| / ||E1 n||, we can obtain the cosine value cos(Eθn) between each running vector and the benchmark running vector, and then use the formula: Eθn=arccos(cos(Eθn)) to calculate the deviation angle Eθn between each running vector and the benchmark running vector.

[0027] As a further solution of the present invention: the specific method of obtaining the real-time deviation angle between the real-time operation vector of each associated node and the reference operation vector of the abnormal node is:

[0028] Through the arc cosine function Kθr=arccos(cos(Kθr)), the real-time deviation angle Kθr between the real-time operation vector of each associated node and the abnormal node benchmark operation vector is calculated, where cos(Kθr)=||GA|| / ||Kr||, where ||GA|| is the module length corresponding to the abnormal node benchmark operation vector.

[0029] As a further solution of the present invention: the specific method of obtaining the frequent set of each node is:

[0030] The operating data of each node in the power system can be regarded as an item. Each node is merged with other remaining nodes in pairs to form multiple item sets. The number of times the item set data of each node appears in the data set is calculated, and the frequency corresponding to each item set is obtained based on the calculation results. Then a frequency threshold is set, and the item set whose frequency exceeds this threshold is defined as the frequent set of the corresponding node.

[0031] As a further solution of the present invention: the specific method of obtaining the frequency corresponding to each item set is:

[0032] S1: Select a node to analyze. First, mark the historical running data corresponding to the N times the analysis node is marked as an abnormal node as a data set, obtain the number of abnormal data in the other data sets, and define the ratio between the number of abnormal times and N as the frequency corresponding to each item set. Mark the item set with a frequency greater than or equal to Y1 as the frequent set of the analysis node, otherwise do nothing;

[0033] S2: Repeat the above step S1 to obtain the frequent sets corresponding to each node.

[0034] Beneficial effects of the present invention:

[0035] (1) The present invention, by marking the associated nodes of each node, can identify the node combination that often has abnormalities at the same time, which helps to distinguish between real failures and data fluctuations caused by the influence of other nodes, thereby reducing false alarms;

[0036] (2) The present invention can prioritize the interference nodes associated with the abnormal nodes by analyzing and marking the interference nodes of the abnormal nodes, thereby reducing the possibility that the faulty nodes are misjudged by the data fluctuations of other nodes. At the same time, it can avoid the data fluctuations of other nodes being ignored because they do not reach the set limit, resulting in the incorrect marking of the faulty nodes. By marking the association between the interference nodes and the abnormal nodes, the comprehensive scan of the entire power grid can be reduced, and the attention can be focused on the interference nodes related to the abnormal nodes. This targeted investigation and processing method helps to improve the reliability and stability of the power system and reduce the risk of power outages and other power grid failures. By emphasizing the association between the abnormal nodes and the interference nodes, and marking and outputting them, the system can more accurately determine the source of interference, help operation and maintenance personnel quickly locate and solve the problem, and further improve the reliability and stability of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention will be further described below in conjunction with the accompanying drawings.

[0038] Figure 1 It is a schematic diagram of the system framework structure of the present invention;

[0039] Figure 2 It is a schematic diagram of the reference running vector, real-time running vector and real-time deviation angle of the present invention. DETAILED DESCRIPTION

[0040] 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.

[0041] Embodiment 1

[0042] See also Figure 1-Figure 2 As shown, the present invention is a digital monitoring system for electric power equipment based on an electric power system, comprising:

[0043] The data acquisition module is used to collect and store the real-time or historical operating data of each node in the power system. The operating data includes voltage, current and frequency, so as to obtain the operating status of each node in the power grid in time and provide a data basis for subsequent analysis and processing;

[0044] The item set acquisition module analyzes the operating data of each node in the power system and obtains the item set of each node in the following way:

[0045] The operating data of each node in the power system can be regarded as an item;

[0046] For illustration, assume that there are three nodes in the power system, node A, node B, and node C, and the voltage, current, and frequency operating data of each node are considered as one item;

[0047] Merge each node with other remaining nodes in pairs to form multiple item sets, and then obtain the item sets of each node. For example, the item set {A, B} and the item set {A, C} represent the item sets composed of node A and nodes B and C respectively.

[0048] The frequent set labeling module is used to calculate the number of times the item set data of each node appears in the data set, and obtain the frequency corresponding to each item set according to the calculation results, and then set a frequency threshold. The item set whose frequency exceeds this threshold is defined as the frequent set of the corresponding node. The specific method is:

[0049] S1: Select node A as the analysis node, analyze each item set of node A one by one, first mark the historical operation data corresponding to node A when it is marked as an abnormal node N times as the data set of node A, obtain the number of data anomalies corresponding to nodes B and C in the data set of node A, that is, the number of data anomalies of nodes B and C when node A is marked as an abnormal node N times, and mark them as b and c respectively, N≥1;

[0050] The frequency AB and AC corresponding to the item set {A, B} and the item set {A, C} are calculated by AB = b / N and AC = c / N respectively;

[0051] Mark the item sets with frequency greater than or equal to Y1 in AB and AC as frequent sets of node A, otherwise do nothing, where Y1 is the frequency threshold, and the specific value of Y1 is formulated by relevant staff according to actual needs;

[0052] S2: Repeat the above step S1 to obtain the frequent itemsets corresponding to each node. By marking the frequent itemsets of each node, the node combinations that often have abnormalities in the system can be identified, which is convenient for targeted troubleshooting when subsequent nodes are abnormal.

[0053] The associated node marking module extracts other nodes that are frequently concentrated on each node and marks them as associated nodes corresponding to each node;

[0054] For example, when the item set {A, B} and the item set {A, C} are both frequent sets of node A, B and C are extracted and marked as associated nodes corresponding to node A;

[0055] By marking the associated nodes of each node, the system can identify combinations of nodes that often exhibit abnormalities at the same time, which helps to distinguish between real failures and data fluctuations caused by the influence of other nodes, thereby reducing false alarms.

[0056] Embodiment 2

[0057] As the second embodiment of the present invention, when the present application is specifically implemented, compared with the first embodiment, the technical solution of the present embodiment is different from the first embodiment only in that the present embodiment further includes a reference operation vector acquisition module, a deviation base angle acquisition module, a real-time deviation angle acquisition module and an interference node output module;

[0058] The benchmark operation vector acquisition module is used to analyze the historical operation data of each node and obtain the benchmark operation vector corresponding to each node according to the analysis results. The specific method is as follows:

[0059] The voltage, current and frequency in the operating data are marked as V, I and F respectively;

[0060] Vectorize the historical operation data of each node in the preset number of times n to obtain the operation vector Een(Ven, Ien, Fen) corresponding to each node in the preset number of times n, where e is the number of nodes in the power system, n is the preset number, e≥1, n≥1;

[0061] The average of the operation vectors Een (Ven, Ien, Fen) corresponding to each node in the preset number of times n is used as the reference operation vector Ge (Vep, Iep, Fep) corresponding to each node;

[0062] The deviation base angle acquisition module analyzes the angle between the reference running vector corresponding to each node and the running vector corresponding to each node in the preset number n, and obtains the deviation base angle corresponding to each node according to the analysis result, specifically in the following way:

[0063] The benchmark running vectors corresponding to each node and the running vector features corresponding to each node in the preset number of times n are placed in the same spatial coordinate system, and the directions of the vectors are kept unchanged, and the starting points of the running vectors corresponding to each node in the preset number of times n are moved to the starting point of the benchmark running vector of each node, thereby determining the intersection between the benchmark running vector of each node and its running vector;

[0064] S01: Calculate the vector modulus by the formula: Calculate and obtain the modulus length ||Ge|| of the benchmark operation vector of each node;

[0065] The vector modulus length calculation formula is: Calculate and obtain the modulus length ||Een|| of the operation vector corresponding to each node in the preset number of times n;

[0066] S02: Select one node from each node as the target node, mark the module length of the benchmark operation vector of the target node as ||G1||, and mark the module lengths of the operation vectors corresponding to the target node in the preset number of times n as ||E1 n||;

[0067] The cosine value cos(Eθn) between each running vector and the reference running vector is obtained by the cosine formula cos(Eθn)=||G1|| / ||E1 n||, and then the inverse cosine function (arccos) is used to calculate the deviation angle Eθn between each running vector and the reference running vector by the formula: Eθn=arccos(cos(Eθn));

[0068] The smaller the deviation angle Eθn is, the smaller the deviation between the corresponding running vector and the reference running vector is, and vice versa.

[0069] Perform standard deviation processing on the deviation angle Eθn to obtain a standard value H of the deviation angle Eθn;

[0070] When the standard value H is less than the preset value Y2, the mean value Eθp of the deviation angle Eθn is used as the deviation base angle P1 of the target node. When the standard value H is greater than or equal to the preset value Y2, the mean value of the maximum and minimum values ​​of the deviation angle Eθn is used as the deviation base angle P1 of the target node. The specific value of Y2 is determined by relevant staff according to actual needs.

[0071] The specific method of processing the standard deviation of Eθn is as follows: Determine the standard value H, where n≥i≥1;

[0072] S03: Repeat step S02 to obtain the deviation base angle Pe of each node, where e refers to the number of nodes in the power system;

[0073] The base deviation degree between the actual running vector of each node and the base running vector is quantified by the base deviation angle;

[0074] The real-time deviation angle acquisition module obtains the real-time operation vector of the real-time operation data of all associated nodes of the abnormal node when an abnormality occurs, and calculates the real-time deviation angle between the real-time operation vector of each associated node and the benchmark operation vector of the abnormal node. The specific method is as follows:

[0075] Obtain the real-time operation data of all associated nodes of the abnormal node, and quantify it into a real-time operation vector Kr(LAr, LBr, LCr), where r is the number of associated nodes corresponding to the abnormal node, e≥r≥1;

[0076] The vector modulus length calculation formula is: Calculate and obtain the modulus length ||Kr|| corresponding to the real-time operation vector of each associated node;

[0077] The real-time deviation angle Kθr between the real-time operation vector of each associated node and the abnormal node benchmark operation vector is calculated by the arccosine function Kθr=arccos(cos(Kθr)), where cos(Kθr)=||GA|| / ||Kr||, where ||GA|| is the modulus length corresponding to the abnormal node benchmark operation vector;

[0078] The interference node output module compares and analyzes the time deviation angle of each associated node with the deviation base angle of the abnormal node, marks the associated node with a larger difference as the interference node of the abnormal node, and outputs it together with the abnormal node. The specific method is as follows:

[0079] The associated nodes that satisfy |Kθr-PA|≥Y3 are marked as interference nodes of abnormal nodes. At the same time, the interference nodes are bound to the abnormal nodes and output at the same time, where Y3 is a preset value, and the specific value is formulated by relevant staff according to actual needs. PA is the deviation base angle corresponding to the abnormal node;

[0080] By analyzing and marking the interference nodes of abnormal nodes, when an abnormality occurs in a node, it is helpful to further analyze the mutual influence of data between nodes, give priority to troubleshooting the interference nodes associated with the abnormal node, reduce the possibility of the faulty node being misjudged due to data fluctuations of other nodes, and avoid the situation where data fluctuations of other nodes are ignored due to not reaching the set limit, thereby causing the faulty node to be incorrectly marked, thereby improving the troubleshooting efficiency of the power system, reducing the comprehensive scan of the entire power grid, and focusing on the interference nodes related to the abnormal node. This targeted troubleshooting and processing method helps to improve the reliability and stability of the power system and reduce the risk of power outages and other power grid failures. By emphasizing the correlation between abnormal nodes and interference nodes, and marking and outputting them, the system can more accurately determine the source of interference, help operation and maintenance personnel quickly locate and solve problems, further improve the reliability and stability of the power system, and thus reduce the risk of power outages and other power grid failures.

[0081] Embodiment 3

[0082] As the third embodiment of the present invention, when the present application is specifically implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments mentioned above for implementation.

[0083] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0084] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A digital monitoring system for power equipment based on a power system, characterized in that: include: Data acquisition module: responsible for collecting the operating data of each node in the power system, namely voltage, current and frequency data; Frequent set marking module: Analyze the running data of each node to obtain the item sets of different nodes, calculate the frequency of item sets, and mark the item sets whose frequency exceeds the threshold as frequent sets; The associated node marking module extracts other nodes in the frequent set as the associated nodes of the node; Benchmark operation vector acquisition module: vectorizes the historical operation data of each node, calculates the mean of the operation vector within a preset number of times, and obtains the benchmark operation vector of each node; Deviation base angle acquisition module: performs standard deviation processing on the angle between the running vector of each node and the reference running vector to determine the deviation base angle of each node; Real-time deviation angle acquisition module: calculates the real-time deviation angle between the benchmark operation vector of the abnormal node and the real-time operation vectors of all associated nodes; Interference node output module: compares the real-time deviation angle of the associated node with the deviation base angle of the abnormal node to obtain the interference node of the abnormal node.

2. The digital monitoring system for electric power equipment based on the electric power system according to claim 1 is characterized in that: The specific method of obtaining the benchmark operation vector corresponding to each node is: The voltage, current and frequency in the operating data are marked as V, I and F respectively, and the historical operating data of each node in the preset number of times n are vectorized into the operating vector Een (Ven, Ien, Fen), and the average of the operating vectors Een (Ven, Ien, Fen) corresponding to each node in the preset number of times n is used as the benchmark operating vector Ge (Vep, Iep, Fep) corresponding to each node, where e is the number of nodes in the power system, n is the preset number, e≥1, n≥1.

3. The digital monitoring system for electric power equipment based on the electric power system according to claim 2 is characterized in that: The specific method of obtaining the deviation base angle corresponding to each node is: S01: Calculate the modulus length of the benchmark running vector of each node and the modulus length of the running vector corresponding to each node in the preset number of times n by using the vector modulus length calculation formula, select one node from each node as the target node, obtain the modulus length of the benchmark running vector of the target node marked as ||G1||, and the modulus length of the running vector corresponding to the target node in the preset number of times n marked as ||E1 n||, obtain the cosine value cos(Eθn) between each running vector and the benchmark running vector by using the cosine formula, and then use the inverse cosine function to obtain the deviation angle Eθn between each running vector and the benchmark running vector, and then perform standard deviation processing on the deviation angle Eθn to obtain the standard value H of the deviation angle Eθn, compare and analyze the standard value H with the preset value Y2, and obtain the deviation base angle P1 of the target node; S02: Finally, repeat step S01 to obtain the deviation base angle Pe of each node.

4. The digital monitoring system for electric power equipment based on the electric power system according to claim 3 is characterized in that: The specific method of obtaining the deviation base angle of the target node is: When the standard value H is less than the preset value Y2, the mean Eθp of the deviation angle Eθn is used as the deviation base angle P1 of the target node. When the standard value H is greater than or equal to the preset value Y2, the mean of the maximum and minimum values ​​of the deviation angle Eθn is used as the deviation base angle P1 of the target node.

5. The digital monitoring system for electric power equipment based on the electric power system according to claim 4 is characterized in that: The specific method of obtaining the real-time deviation angle between the reference running vector of the abnormal node and the real-time running vectors of all associated nodes is as follows: The real-time operation data of all associated nodes of the abnormal node are quantified into a real-time operation vector Kr (LAr, LBr, LCr). The modulus length ||Kr|| corresponding to the real-time operation vector of each associated node is calculated by the vector modulus length calculation formula. The real-time deviation angle Kθr between the real-time operation vector of each associated node and the abnormal node benchmark operation vector is calculated by the arc cosine function, where r is the number of associated nodes corresponding to the abnormal node, e≥r≥1.

6. The digital monitoring system for electric power equipment based on the electric power system according to claim 5 is characterized in that: The specific method of obtaining the interference node of the abnormal node is: The associated nodes satisfying |Kθr-PA|≥Y3 are marked as interference nodes of the abnormal nodes, and the interference nodes are bound to the abnormal nodes and output at the same time, where PA is the deviation base angle corresponding to the abnormal node and Y3 is the preset value.

7. The digital monitoring system for electric power equipment based on the electric power system according to claim 3 is characterized in that: The specific method of obtaining the deviation angle between each running vector and the reference running vector is as follows: Through the cosine formula cos(Eθn)=||G1|| / ||E1 n||, we can obtain the cosine value cos(Eθn) between each running vector and the benchmark running vector, and then use the formula: Eθn=arccos(cos(Eθn)) to calculate the deviation angle Eθn between each running vector and the benchmark running vector.

8. The digital monitoring system for electric power equipment based on the electric power system according to claim 5 is characterized in that: The specific method of obtaining the real-time deviation angle between the real-time operation vector of each associated node and the reference operation vector of the abnormal node is: Through the arc cosine function Kθr=arccos(cos(Kθr)), the real-time deviation angle Kθr between the real-time operation vector of each associated node and the abnormal node benchmark operation vector is calculated, where cos(Kθr)=||GA|| / ||Kr||, where ||GA|| is the module length corresponding to the abnormal node benchmark operation vector.

9. The digital monitoring system for electric power equipment based on the electric power system according to claim 5 is characterized in that: The specific method of obtaining the frequent set of each node is: The operating data of each node in the power system can be regarded as an item. Each node is merged with other remaining nodes in pairs to form multiple item sets. The number of times the item set data of each node appears in the data set is calculated, and the frequency corresponding to each item set is obtained based on the calculation results. Then a frequency threshold is set, and the item set whose frequency exceeds this threshold is defined as the frequent set of the corresponding node.

10. The digital monitoring system for electric power equipment based on electric power system according to claim 1, characterized in that: The specific method to obtain the frequency corresponding to each item set is: S1: Select a node to analyze. First, mark the historical running data corresponding to the N times the analysis node is marked as an abnormal node as a data set, obtain the number of abnormal data in the other data sets, and define the ratio between the number of abnormal times and N as the frequency corresponding to each item set. Mark the item set with a frequency greater than or equal to Y1 as the frequent set of the analysis node, otherwise do nothing; S2: Repeat the above step S1 to obtain the frequent sets corresponding to each node.

Images