Method for processing wrong virtual loop after verification of secondary virtual loop of intelligent substation

Abstract

The invention provides a method for processing an error virtual loop after verification of a secondary virtual loop of an intelligent substation, and belongs to the technical field of intelligent substation model verification. Comprising the following steps: acquiring all possible error reasons of the secondary virtual loop; acquiring knowledge of the secondary loop which is called and still not matched correctly; defining a fuzzy matrix between the secondary loop knowledge and the error reason; according to the fuzzy vector of the to-be-detected error virtual loop, utilizing the fuzzy matrix to obtain a fuzzy vector of an error reason; and diagnosing the error cause of the to-be-detected error virtual loop according to the sorting result of the error cause fuzzy vectors. According to the method, the error result of the virtual loop verification is analyzed and processed, the error reason of the wrong virtual loop can be positioned, a user is helped to quickly position the problem, and the integrity of the virtual loop verification is improved.

Description

Technical Field

[0001] The invention belongs to the technical field of intelligent substation model verification, and in particular relates to a method for processing an erroneous virtual circuit after a secondary virtual circuit verification of an intelligent substation. Background Art

[0002] The configuration file (SCD file) of the smart substation contains the configuration of the entire station's intelligent electronic equipment, communication parameters and substation information. The virtual terminal connection in this file reflects the corresponding connection relationship of the secondary equipment in the station, which is the key to the transmission of the entire station's electrical quantity information and the core of the secondary system. Therefore, the verification of the configuration file is very important.

[0003] At present, there are clear standards for the syntax and semantics of configuration files, but there is a lack of verification basis for the functional verification of virtual terminal connections. Therefore, patent document CN115563303A proposes a functional verification method for secondary virtual circuits in smart substations. This method imports existing SCD files to obtain virtual terminal connection information, extracts and processes knowledge, forms a set of generalized virtual circuit relationship rule bases based on knowledge graphs, and verifies the correctness of secondary virtual circuit connections through consistency reasoning of physical virtual circuits. This method can verify the correct virtual circuits, but for virtual circuits with incorrect verification results, there is a lack of further analysis and processing methods. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method for processing an erroneous virtual circuit after verification of the secondary virtual circuit of an intelligent substation, which can locate the virtual circuit with an erroneous verification result after verifying the configuration file of the intelligent substation using a knowledge graph algorithm.

[0005] In order to solve one, part or all of the above technical problems, the technical solution adopted by the present invention is: A method for processing an erroneous virtual circuit after checking a secondary virtual circuit of an intelligent substation, comprising: Get all possible error causes of the secondary virtual circuit; Get called and still not matched with the correct secondary circuit knowledge; Define the fuzzy matrix between secondary loop knowledge and error causes; According to the fuzzy vector of the faulty virtual loop to be detected, the fuzzy vector of the fault cause is obtained by using the fuzzy matrix; The error cause of the to-be-detected erroneous virtual circuit is diagnosed according to the sorting result of the error cause fuzzy vector.

[0006] Furthermore, the method also includes: optimizing the fuzzy matrix using the current diagnosis result.

[0007] Furthermore, the error causes include IED serial number error, bay serial number error, virtual terminal model error, voltage level error, set error, and whether in the same bay error.

[0008] Furthermore, the fuzzy matrix is ​​expressed as ,in , , , m is the amount of secondary loop knowledge obtained, n is the number of error causes; R represents a fuzzy relationship between secondary loop knowledge and error causes; A quantitative fuzzy value that reflects the degree of correlation between the i-th secondary circuit knowledge and the j-th error cause.

[0009] Further, the method of obtaining the fuzzy vector of the error cause by using the fuzzy matrix includes: The fuzzy vector of the false virtual loop to be detected is expressed as , is the membership degree of the i-th secondary loop knowledge, ; The fuzzy vector of the error cause is defined as , is the membership degree of the jth error cause, ; but .

[0010] Further, the method for diagnosing the error cause of the to-be-detected erroneous virtual circuit according to the sorting result of the error cause fuzzy vector includes: Arrange the elements in the error cause fuzzy vector from large to small, then the error cause corresponding to the first element after sorting is the most likely error cause of the faulty virtual loop to be detected, and the next most likely error cause is the error cause corresponding to the subsequent elements.

[0011] Furthermore, the method of optimizing the fuzzy matrix using the current diagnosis result includes: optimizing the fuzzy matrix using a fuzzy neural network.

[0012] Further, the method of optimizing the fuzzy matrix using the fuzzy neural network includes: Construct a fuzzy neural network, use the fuzzy vector as the input of the neural network, and the neural network obtains the actual output through synthetic operation. , , whose operation is expressed as , are the weights between the input pattern and the output pattern in the neural network. is the membership degree of the i-th secondary loop knowledge in the fuzzy vector, ; make ,in is the expected output, is the actual output, Represents the output error; but ,in represents the weight at time t, It represents the new weight obtained after the weight at time t is corrected once, a is the proportional factor, , after the above method converges, the new weight As the corresponding element value in the fuzzy matrix.

[0013] Compared with the prior art, the present invention has the following beneficial effects: The present invention analyzes and processes the erroneous results of the virtual circuit verification, can locate the error cause of the erroneous virtual circuit, help users to quickly locate the problem, and improve the integrity of the virtual circuit verification. DETAILED DESCRIPTION

[0014] In order to better understand the present invention, the content of the present invention is further clearly set forth in conjunction with the embodiments below, but the protection content of the present invention is not limited to the following embodiments. In the following description, a large number of specific details are provided in order to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details.

[0015] After verifying the smart substation configuration file using a knowledge graph algorithm (such as patent document CN115563303A), the verification results of the virtual circuit include correct, measured and controlled, pending confirmation, wrong, missing and repeated.

[0016] Correct: The actual imaginary circuit exactly matches the secondary circuit knowledge.

[0017] Measurement and control: The function of the measurement and control virtual loop is mainly to send the operating status to the upper-level monitoring system or the master station system, which is content that does not need to be verified.

[0018] To be confirmed: The actual virtual circuit cannot be found and is not related to any knowledge in the knowledge base, so its purpose cannot be determined.

[0019] Missing: Virtual circuits that should be connected but are not connected. That is, after all verifications are completed, there is still no matching secondary circuit knowledge.

[0020] Repeat: A virtual circuit with the same function already exists in the interval, that is, both ends of the virtual circuit can match the used knowledge.

[0021] Error: One end of the actual virtual circuit connection matches unused knowledge. That is, there is a secondary circuit knowledge that is not used, and the virtual terminal model extracted from one end of this virtual circuit matches the secondary circuit knowledge. Another case is the busbar protection interval error, that is, the reception and transmission of each interval of the busbar protection should correspond to the same interval.

[0022] Errors, omissions and duplications all require further analysis by the system. Omissions and duplications have clear knowledge to match them, and the analysis and matching process is relatively simple. However, for errors, it is possible that a single virtual circuit may have similar knowledge to multiple secondary circuits, which requires special processing and analysis.

[0023] Based on this, the present application proposes a method for analyzing and processing the error results of virtual circuit verification, which can locate the cause of the error.

[0024] Embodiment 1: The purpose of this embodiment is to provide a method for processing an erroneous virtual circuit after the secondary virtual circuit of a smart substation is verified.

[0025] The virtual circuits and secondary circuit knowledge classified as errors are not simply a one-to-one relationship. For example, if a device can receive tripping commands from two protections, A and B, and there are problems with the verification of both virtual circuits, it is necessary to determine the relationship between the virtual circuit and the knowledge of the two protection tripping circuits, A and B.

[0026] The processing method comprises: S1. Obtain all possible error causes of the secondary virtual circuit.

[0027] The causes of the error include IED serial number error, bay serial number error, virtual terminal model error, voltage level error, set error, and whether it is in the same bay error.

[0028] S2, obtains the secondary loop knowledge that is called and still not matched correctly.

[0029] From the current interval of the secondary virtual circuit knowledge graph, obtain the secondary circuit knowledge entry that has been called but still has not been matched correctly.

[0030] S3. Define the fuzzy matrix between secondary circuit knowledge and error causes.

[0031] The fuzzy matrix is ​​expressed as ,in , , , m is the amount of secondary loop knowledge obtained, n is the number of error causes; R represents a fuzzy relationship between secondary loop knowledge and error causes; The quantitative fuzzy value that reflects the correlation between the i-th secondary circuit knowledge and the j-th error cause, The larger it is, the greater the role of the j-th error cause in the i-th secondary circuit knowledge, that is, the greater the possibility that this error cause is the main error cause.

[0032] The initial values ​​in the fuzzy matrix R are comprehensively evaluated based on historical records or expert systems.

[0033] S4. Obtain a fuzzy vector of the cause of the error according to the fuzzy vector of the erroneous virtual loop to be detected.

[0034] The fuzzy vector of the false virtual loop to be detected is expressed as , the fuzzy vector of the false virtual loop to be detected is constructed based on the secondary loop knowledge, is the membership degree of the i-th secondary loop knowledge, .

[0035] The fuzzy vector of the error cause is defined as , is the membership degree of the jth error cause, .but .

[0036] S5. Diagnose the error cause of the erroneous virtual circuit to be detected according to the sorting result of the error cause fuzzy vector.

[0037] Arrange the elements in the error cause fuzzy vector Y from large to small, then the error cause corresponding to the first element after sorting is the most likely error cause of the faulty virtual loop to be detected, and the next most likely error cause is the error cause corresponding to the subsequent elements.

[0038] S6. Optimize the fuzzy matrix using the current diagnosis results.

[0039] Construct a fuzzy neural network, take the fuzzy vector X as the input of the neural network, and the neural network obtains the actual output through synthetic operation. , , whose operation is expressed as , It is the weights between the input mode and the output mode in the neural network, that is, the membership value from the fault phenomenon to the fault cause. The weight adjustment process is the adjustment process of the membership value. The basic idea of ​​weight adjustment is to use the deviation between the expected output and the actual output of the neural network as a reference for connection weight adjustment, and ultimately reduce this deviation. The specific adjustment process is: make ,in is the expected output, is the actual output, Represents the output error.

[0040] but ,in represents the weight at time t, It represents the new weight obtained after the weight at time t is corrected once, a is the proportional factor, satisfying , the above method can always converge to complete the weights , that is, the optimization of each membership value in the fuzzy matrix, that is, with the new weights As the corresponding element value in the fuzzy matrix, so as to achieve the effect of self-learning.

[0041] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Other modifications or equivalent substitutions made to the technical solution of the present invention by ordinary technicians in the field should be included in the scope of the claims of the present invention as long as they do not depart from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for processing an erroneous virtual circuit after verification of a secondary virtual circuit of an intelligent substation, characterized in that: include Get all possible error causes of the secondary virtual circuit; Get called and still not matched with the correct secondary circuit knowledge; Define the fuzzy matrix between secondary loop knowledge and error causes; According to the fuzzy vector of the faulty virtual loop to be detected, the fuzzy vector of the fault cause is obtained by using the fuzzy matrix; The error cause of the to-be-detected erroneous virtual circuit is diagnosed according to the sorting result of the error cause fuzzy vector.

2. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of a smart substation according to claim 1, characterized in that: Also includes: Optimize the fuzzy matrix using the current diagnosis results.

3. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of the smart substation according to claim 1, characterized in that: The error causes include IED serial number error, bay serial number error, virtual terminal model error, voltage level error, set error, and whether in the same bay error.

4. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of a smart substation according to claim 1, characterized in that: The fuzzy matrix is ​​expressed as ,in , , , m is the amount of secondary loop knowledge obtained, n is the number of error causes; R represents a fuzzy relationship between secondary loop knowledge and error causes; A quantitative fuzzy value that reflects the degree of correlation between the i-th secondary circuit knowledge and the j-th error cause.

5. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of a smart substation according to claim 1, characterized in that: Methods for obtaining the fuzzy vector of the error cause using the fuzzy matrix include: The fuzzy vector of the false virtual loop to be detected is expressed as , is the membership degree of the i-th secondary loop knowledge, ; The fuzzy vector of the error cause is defined as , is the membership degree of the jth error cause, ; but .

6. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of a smart substation according to claim 1, characterized in that: The method for diagnosing the error cause of the to-be-detected erroneous virtual circuit according to the sorting result of the error cause fuzzy vector includes: Arrange the elements in the error cause fuzzy vector from large to small, then the error cause corresponding to the first element after sorting is the most likely error cause of the faulty virtual loop to be detected, and the next most likely error cause is the error cause corresponding to the subsequent elements.

7. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of a smart substation according to claim 2, characterized in that: The method of optimizing the fuzzy matrix using the current diagnosis result includes: optimizing the fuzzy matrix using a fuzzy neural network.

8. The method for processing an erroneous virtual circuit after checking the secondary virtual circuit of a smart substation according to claim 7, characterized in that: The methods of optimizing fuzzy matrix using fuzzy neural network include: Construct a fuzzy neural network, use the fuzzy vector as the input of the neural network, and the neural network obtains the actual output through synthetic operation. , , whose operation is expressed as , are the weights between the input pattern and the output pattern in the neural network. is the membership degree of the i-th secondary loop knowledge in the fuzzy vector, ; make ,in is the expected output, is the actual output, Represents the output error; but ,in represents the weight at time t, It represents the new weight obtained after the weight at time t is corrected once, a is the proportional factor, , after the above method converges, the new weight As the corresponding element value in the fuzzy matrix.

Images