A method and device for automatically generating substation mirror intervals based on semantic recognition

Abstract

A method for automatically generating mirrored intervals in a substation based on semantic recognition is characterized by the following steps: Step 1: generating a centralized control system model library based on the model information and topological relationships of equipment and measurement points in the substation, and generating a topological map of the original interval based on the system library; Step 2: extracting the mirrored interval corresponding to the original interval from the system library, and generating a topological map of the mirrored interval based on the topological map of the original interval. This method is simple to use, avoids the tedious workload and potential errors caused by extensive manual operations by operation and maintenance personnel, and significantly improves the efficiency and accuracy of interval topological map generation.

Description

Technical Field

[0001] The present invention relates to the field of power system maintenance, and more particularly to a method and device for automatically generating substation mirror intervals based on semantic recognition. Background Art

[0002] At present, the new generation of centralized control station equipment monitoring systems usually adopt a partitioned architecture and are built with equipment as the core. They can combine new technologies such as big data, the Internet of Things, artificial intelligence, and visualization to build a new substation operation and maintenance model of "centralized control station + unmanned substation + equipment master system" in data perception, information transmission, global sharing, and intelligent applications, thereby realizing remote integrated monitoring and control of main and auxiliary equipment, intelligent analysis of equipment status and fault warning, remote online intelligent inspections, and full life cycle management of equipment.

[0003] In existing technologies, the new generation of centralized control stations can be integrated with autonomously controllable substations, fully inheriting the experience and achievements of existing dispatching master stations, substations, and related system construction. A single centralized control station system can connect to more than 50 substations and provide centralized control for a large number of substations.

[0004] However, each substation contains a large number of bays. Substation bays, as crucial components of a substation, contain ultra-high (or high) voltage switchgear, power lines, power transformers, and other components connected to the substation busbar system, as well as protection, control, and measurement equipment for these lines and transformers. Therefore, to achieve effective centralized control of the substation, a bay topology diagram must be drawn for all equipment within each bay. Furthermore, since each bay contains a large amount of measurement point information, this information also requires manual association.

[0005] In traditional methods, drawing interval topology maps and associating measurement point information requires a lot of manual operations, which not only leads to a large amount of manual maintenance costs and places a huge burden on the work of monitoring personnel, but also introduces errors due to the uncontrollability of manual operations, and the errors are highly hidden.

[0006] It's important to note that due to the redundant design of the power grid, a large number of bay topologies are completely identical. Not only are the topology structures identical between mirrored bays, but information such as model associations, device names, and corresponding light-emitting diodes also exhibit significant similarity or overlap. Preliminary analysis revealed that this duplication accounts for approximately 40% of substation topologies.

[0007] In order to solve the above problems, there is an urgent need for a mirror interval generation method that significantly reduces the workload of manual operations and provides a high degree of automation and accuracy. Summary of the Invention

[0008] In order to address the deficiencies in the prior art, the purpose of the present invention is to provide a method and device for automatically generating substation mirror intervals based on semantic identification, which realizes identification and automatic generation of interval topology maps based on the similarity of semantic content such as device names and measurement point names between two mirror intervals.

[0009] The present invention adopts the following technical solutions.

[0010] A first aspect of the present invention relates to a method for automatically generating a mirror interval of a substation based on semantic recognition, wherein the method comprises the following steps: Step 1, generating a system library based on model information of equipment and measuring points in the substation, and generating a topology map of the original interval based on the system library; Step 2, extracting a mirror interval corresponding to the original interval from the system library, and generating a topology map of the mirror interval based on the topology map of the original interval.

[0011] Preferably, the topology map of the original interval includes the device ID and device name of all devices in the original interval, the measuring point ID and measuring point name of all measuring points, and the connection relationship between any two of all devices and all measuring points; the method is used to parse and obtain the device ID and device name of all devices, the measuring point ID and measuring point name of all measuring points, and the connection relationship between any two of all devices and all measuring points in XML format.

[0012] Preferably, the topology map of the original interval may be generated by inputting a storage path of the topology map of the original interval, and acquiring and parsing the topology map file from the storage path.

[0013] Preferably, the method for extracting the mirror interval corresponding to the original interval from the system library is: tracing the measurement point names and device names in the topology map of the original interval in the system library; based on the tracing, manually extracting all the measurement point names and device names corresponding to the original interval, and confirming the correspondence between the manually extracted measurement points and devices and the measurement points and devices in the original interval based on semantic recognition; based on the correspondence, obtaining the topological structure of the mirror interval; wherein, the mirror interval has exactly the same topological structure as the original interval.

[0014] Preferably, the method for generating the topology map of the mirror interval is as follows: step 2.1, obtaining all device names under the original interval and all device names under the mirror interval that are stored in the system library; step 2.2, deleting the interval prefix in all device names, and identifying all device names under the original interval and all device names under the mirror interval one by one, and determining the devices under the mirror interval with completely identical names as mirrors of the devices under the original interval; step 2.3, eliminating the devices that have been successfully determined, and performing similarity semantic recognition based on the device name on the remaining devices, and realizing mirror recognition of all devices based on similarity semantic recognition; step 2.4, using the method in steps 2.1 to 2.3 to perform mirror recognition on all measurement point names under the original interval and all measurement point names under the mirror interval; step 2.5, copying the topology map of the original interval to generate the topology map of the mirror interval, and filling all devices and measurement points obtained in the mirror recognition step, and all connection relationships into the topology map of the mirror interval.

[0015] Preferably, in step 2.3, the method for similarity identification is: obtaining the device name and measurement point name in the original interval with the interval prefix deleted and the device name and measurement point name in the mirror interval with the interval prefix deleted; taking the name with the longer string length as the basis, identifying each character one by one and obtaining the total length of repeated characters; the similarity is the ratio of the length of the repeated characters to the length of the longer string length.

[0016] Preferably, similarity levels are assigned to the device names and measurement point names in the mirror interval based on the similarity values.

[0017] Preferably, the colors in the device name and measurement point name lists are displayed based on the similarity level.

[0018] The second aspect of the present invention relates to a device for automatically generating substation mirror intervals based on semantic identification, wherein the device is implemented based on a method for automatically generating substation mirror intervals based on semantic identification described in the first aspect of the present invention.

[0019] The present invention offers the following advantages: compared to existing technologies, the method and device for automatically generating substation mirror bays based on semantic recognition can automatically generate bay topology maps based on the similarity of semantic content, such as device names and measurement point names, between two mirror bays. This method is simple to use, avoids the tedious workload and potential errors often associated with extensive manual operations by operations and maintenance personnel, and significantly improves the efficiency and accuracy of bay topology map generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic flow chart of the steps of a method for automatically generating substation mirror intervals based on semantic recognition according to the present invention;

[0021] Figure 2 A schematic diagram of a list of equipment names and measuring point names in a method for automatically generating substation mirror intervals based on semantic recognition according to the present invention;

[0022] Figure 3 A schematic diagram of similarity level allocation in a method for automatically generating substation mirror intervals based on semantic recognition according to the present invention;

[0023] Figure 4 It is a bay topology diagram of the original bay in the method for automatically generating substation mirror bays based on semantic recognition according to the present invention;

[0024] Figure 5 The present invention provides a topological diagram of a mirror interval in a method for automatically generating a mirror interval in a substation based on semantic recognition. DETAILED DESCRIPTION

[0025] The present application will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present application.

[0026] Figure 1 The figure is a flow chart of the steps of the method for automatically generating substation mirror interval based on semantic recognition according to the present invention. Figure 1 As shown, a method for automatically generating substation mirror intervals based on semantic identification is provided, wherein the method includes steps 1 and 2.

[0027] Step 1: Generate a system library based on model information of equipment and measuring points in the substation, and generate a topology map of the original bay based on the system library.

[0028] Specifically, the system library in the present invention can be a database for storing centralized control system model information, wherein the model information includes the ID, name, description, and any relationship between any two of the devices and measuring points in the substation.

[0029] In one embodiment of the present invention, a system library can be generated based on the model information of the substation. The system library needs to include the corresponding information for generating the original interval and mirror interval topology diagrams. For example, the system library may include the ID numbers of all substation interval devices and measuring points, device names, measuring point names, connection relationships between devices, connection relationships between measuring points, or connection relationships between a certain measuring point and a certain device. This information can be obtained by reading the information of the corresponding control system in the substation. It should be noted that the connection relationship here can be either the actual geographical location relationship between the two devices, or the uplink and downlink relationship of the network connection topology in the actual data transmission process. The measuring point can be recorded as the actual location or line it is located on, what equipment is at this actual location, what the equipment is at both ends of this line, etc.

[0030] Preferably, the topology map of the original interval includes the device ID and device name of all devices in the original interval, the measuring point ID and measuring point name of all measuring points, and the connection relationship between any two of all devices and all measuring points; the method is used to parse the device ID and device name of all devices, the measuring point ID and measuring point name of all measuring points, and the connection relationship between any two of all devices and all measuring points in XML format.

[0031] In one embodiment of the present invention, a topology map of the original interval can be generated based on the system library, and a topology map of the mirrored interval can be generated using this topology map as a reference. The process of generating the topology map of the original interval can be manually implemented. For example, an operation and maintenance personnel can manually generate the corresponding topology map structure based on the information in the system library and assign corresponding device names to the topology map.

[0032] The above process can be implemented remotely. Therefore, in the present invention, the original interval topology map that has been generated can also be imported through the software used in the method.

[0033] Preferably, the topology map of the original interval may be generated by inputting a storage path of the topology map of the original interval, and acquiring and parsing the topology map file from the storage path.

[0034] In one embodiment of the present invention, the 14412 version of Sifang Relay Protection's CSGC3000 / NCCS system and the 7.12 version of Jincang Database can be used as the database system of the centralized control station monitoring system to implement the functions of the present invention.

[0035] Therefore, the storage path of the generated interval topology map can be imported through the system tool, and then the equipment and measurement point information under the interval can be analyzed accordingly.

[0036] Step 2: extract the mirror interval corresponding to the original interval from the system library, and generate a topology map of the mirror interval based on the topology map of the original interval.

[0037] In the present invention, after parsing the interval topology of the original interval, the mirror interval related to the above information can be extracted from the system library. In the present invention, the mirror interval can be set and the equipment and measurement point information under the mirror interval can be automatically obtained through software tools.

[0038] Preferably, the method for extracting the mirror interval corresponding to the original interval from the system library is as follows: tracing the measurement point names and device names in the topological diagram of the original interval in the system library; based on the tracing, manually extracting all the measurement point names and device names corresponding to the original interval, and confirming the correspondence between the manually extracted measurement points and devices and the measurement points and devices in the original interval based on semantic recognition; based on the correspondence, obtaining the topological structure of the mirror interval; wherein the mirror interval and the original interval have exactly the same topological structure. In one embodiment of the present invention, the mirror interval information corresponding to the original interval can be found from the system library based on the relevant information of the devices and measurement points. At the same time, the mirror device related to the above information can be extracted from the system library.

[0039] Preferably, the method for generating the topology map of the mirror interval is as follows: step 2.1, obtaining all device names under the original interval and all device names under the mirror interval that have been stored in the system library; step 2.2, deleting the interval prefix in all device names, and identifying all device names under the original interval and all device names under the mirror interval one by one, and determining the devices under the mirror interval with completely identical names as mirrors of the devices under the original interval; step 2.3, eliminating the devices that have been successfully determined, and performing similarity semantic recognition based on the device name on the remaining devices, and realizing mirror recognition of all devices based on similarity semantic recognition; step 2.4, using the method in steps 2.1 to 2.3 to perform mirror recognition on all measurement point names under the original interval and all measurement point names under the mirror interval; step 2.5, copying the topology map of the original interval to generate the topology map of the mirror interval, and filling all devices and all measurement points obtained in the mirror recognition step, and all connection relationships into the topology map of the mirror interval.

[0040] The specific content to be filled in here can include device name, measurement point name, connection relationship, etc.

[0041] In the present invention, devices and measurement points can be compared and matched separately. For devices, duplicate prefixes may exist in the device name. For example, when the original bay and the mirror bay are both bays on transformer No. 1, the name prefixes of all or some of the devices may include a prefix similar to "Transformer No. 1" based on manual naming rules. In another example, when the bay is a busbar bay, prefixes similar to "110kV I Bus" may also be included. In the present invention, these prefixes can be deleted.

[0042] The method for finding these intervals can be to compare the names of all devices in the same interval and find the same content, which is the interval prefix described in the present invention. After deleting the interval prefix in all device names, the comparison is performed again.

[0043] During the comparison process, devices with identical names can first be compared one by one to achieve a mirror match. Subsequently, since device naming is manually performed during system construction, inconsistent naming may occur. To address this issue, after completing the identical comparison, similarity identification can be performed.

[0044] Preferably, in step 2.3, the method for similarity identification is: obtaining the device name and measurement point name in the original interval with the interval prefix deleted and the device name and measurement point name in the mirror interval with the interval prefix deleted; using the name with the longer string length as a reference, identifying each character one by one and obtaining the total length of repeated characters; the similarity is the ratio of the length of the repeated characters to the length of the longer string length.

[0045] In the method of the present invention, the method for similarity identification takes into account the problem of incorrect writing of a character when manually filling in the device name. Therefore, a longer device name can be used as a reference and another device name can be used to compare each character one by one to obtain similarity data.

[0046] Similarity is the ratio of the length of the same characters to the total length of characters. Through this comparison, similar device names can be screened out to achieve matching of mirrored devices. In order to prevent confusion between devices that are too similar but essentially different, the method of the present invention prioritizes removing the common character strings of the names in this interval before comparison. For example, the "472 circuit breaker", "4726 knife switch", "47261 ground switch" in the original interval and the "462 circuit breaker", "4626 knife switch", "46261 ground switch" in the mirrored interval, remove "472" and "462" respectively, and then compare the device names "circuit breaker", "6 knife switch", "61 ground switch".

[0047] Through this comparison method, no matter what kind of non-standard naming method appears in the aforementioned manual naming process, the mirror device can be correctly found.

[0048] After comparing device names using the above method, you can also use the same method to compare measurement point names. Finally, input the matching results into the new mirror interval topology map, and the topology map of the mirror interval is generated.

[0049] Preferably, similarity levels are assigned to the device names and measurement point names in the mirror interval based on the similarity values.

[0050] In the list of device names and measuring point names generated by the present invention, in order to very intuitively obtain the similarity information between each device name, measuring point name and the associated devices and measuring points, they can be classified in a similarity level manner.

[0051] Preferably, the similarity level is displayed based on the colors in the device name and measurement point name lists. Figure 2 A schematic diagram of a list of equipment names and measuring point names in a method for automatically generating substation mirror intervals based on semantic recognition according to the present invention; Figure 3 This is a schematic diagram of similarity level allocation in a method for automatically generating substation mirror intervals based on semantic recognition according to the present invention. Figure 2 、 Figure 3 As shown, based on the relevant information in the system library, a list of device names and measurement point names can be generated. After solving the similarity using the above method, different colors can be used to mark the device names and measurement point names with different similarity levels.

[0052] In the method of the present invention, after obtaining information that matches the topology information of the original interval, a completely identical topology map can be generated based on the original interval. Figure 4 It is a bay topology diagram of the original bay in the method for automatically generating substation mirror bays based on semantic recognition according to the present invention; Figure 5 This is a topological diagram of the mirror interval in the automatic generation method of the substation mirror interval based on semantic recognition of the present invention. Figure 4 As shown, Figure 5 The topology diagram in the Figure 4 After copying the topology map, you can also replace the corresponding name information in the topology map according to the name information of the devices and measurement points mirrored in the previous step. The result of the replacement can be the new topology map.

[0053] The present invention replaces the device name and connection relationship during the replacement process, but retains the optical sign information of the mirror interval under the same structure, which greatly reduces the workload of manual input and improves the accuracy of the topology map.

[0054] The second aspect of the present invention relates to a device for automatically generating a substation mirror interval based on semantic identification, which is implemented based on a method for automatically generating a substation mirror interval based on semantic identification in the first aspect of the present invention.

[0055] The beneficial effect of the present invention lies in that, compared with existing technologies, the method and device for automatically generating substation mirror bays based on semantic recognition can automatically generate bay topology maps based on the similarity of semantic content, such as device names and measurement point names, between two mirror bays. This method is simple, avoids the tedious workload and potential errors caused by extensive manual operations by operation and maintenance personnel, and significantly improves the efficiency and accuracy of bay topology map generation.

[0056] The applicant of the present invention has made a detailed explanation and description of the implementation examples of the present invention in conjunction with the drawings in the specification. However, those skilled in the art should understand that the above implementation examples are only preferred implementation plans of the present invention, and the detailed description is only to help readers better understand the spirit of the present invention, and is not a limitation on the scope of protection of the present invention. On the contrary, any improvements or modifications based on the inventive spirit of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for automatically generating substation mirror intervals based on semantic recognition, characterized in that: The method comprises the following steps: Step 1: generating a system library based on model information of equipment and measuring points in the substation, and generating a topology map of the original interval based on the system library; Step 2: extracting a mirror interval corresponding to the original interval from the system library, and generating a topology map of the mirror interval based on the topology map of the original interval; The topology diagram of the original interval includes the device IDs and device names of all devices in the original interval, the measurement point IDs and measurement point names of all measurement points, and the connection relationship between any two of the devices and the measurement points; The method for extracting the mirror interval corresponding to the original interval from the system library is: Tracing the measurement point names and device names in the topology diagram of the original interval in the system library; Based on the tracing, manually extract all the measuring point names and equipment names corresponding to the original interval, and confirm the correspondence between the manually extracted measuring points and equipment and the measuring points and equipment in the original interval based on the semantic recognition; Based on the corresponding relationship, a topological structure of the mirror interval is obtained; wherein the mirror interval has the same topological structure as the original interval; The method for generating the topological graph of the mirror interval is: Step 2.1, obtaining all device names in the original interval and all device names in the mirror interval that are stored in the system library; Step 2.2: Delete the interval prefix from all device names, identify all device names in the original interval and all device names in the mirrored interval one by one, and determine that the devices in the mirrored interval with completely identical names are mirror images of the devices in the original interval. Step 2.3: Eliminate the devices that have been successfully identified, perform semantic recognition based on the similarity of the device names on the remaining devices, and implement mirror recognition of all devices based on the similarity semantic recognition; Step 2.4, using the methods in steps 2.1 to 2.3 to perform mirror recognition on all the measurement point names in the original interval and all the measurement point names in the mirror interval; Step 2.5: copy the topology map of the original interval to generate the topology map of the mirror interval, and fill all the devices and all the measurement points obtained in the mirror identification step and all the connection relationships into the topology map of the mirror interval.

2. The method for automatically generating substation mirror intervals based on semantic identification according to claim 1, characterized in that: The method is used to parse and obtain the device IDs and device names of all devices, the measuring point IDs and measuring point names of all measuring points, and the connection relationship between any two of all devices and all measuring points in an XML manner.

3. The method for automatically generating substation mirror intervals based on semantic recognition according to claim 2, characterized in that: The topological map of the original interval may also be generated in the following manner: The storage path of the topology map of the original interval is input, and the topology map file is acquired and parsed from the storage path.

4. The method for automatically generating substation mirror intervals based on semantic recognition according to claim 3 is characterized in that: In step 2.3, the similarity identification method is: Acquire the device name and the measuring point name in the original interval with the interval prefix deleted and the device name and the measuring point name in the mirrored interval with the interval prefix deleted; Based on the name with the longest character string, identify each character one by one and obtain the total length of repeated characters; The similarity is the ratio of the length of the repeated characters to the length of the longer character string.

5. The method for automatically generating substation mirror intervals based on semantic recognition according to claim 4, characterized in that: Based on the value of the similarity, similarity levels are assigned to the device names and measurement point names in the mirroring interval.

6. The method for automatically generating substation mirror intervals based on semantic recognition according to claim 5, characterized in that: The colors in the device name and measurement point name lists are displayed based on the similarity levels.

7. A device for automatically generating substation mirror intervals based on semantic recognition, characterized by: The device is implemented based on a method for automatically generating substation mirror intervals based on semantic recognition as described in any one of claims 1-6.

Images