Similar map searching method and device, electronic equipment and storage medium

Abstract

The invention discloses a similar map searching method and device, electronic equipment and a storage medium, the method and device are applied to the electronic equipment, and specifically, a pre-trained standardization network is utilized to carry out standardization processing on a collected to-be-matched original PRPD map, and a binary semantic segmentation map is obtained; extracting a plurality of correlation parameters of the binary semantic segmentation map; on the basis of the multiple correlation parameters, correlation parameters of the binary semantic segmentation map and each historical PRPD map in a historical map database are calculated, multiple map similarities are obtained, and the historical map database comprises multiple historical PRPD maps subjected to standardization processing through a standardization network; and selecting the historical PRPD atlas corresponding to the maximum atlas similarity in the plurality of atlas similarities as a target historical PRPD atlas similar to the original PRPD atlas. By means of the scheme, comparison search can be conducted without manual means, a large amount of manpower and material resources do not need to be consumed, and therefore the search efficiency is improved.

Description

Technical Field

[0001] This application relates to the field of ultra-high frequency partial discharge monitoring technology, and more specifically, to a similar spectrum search method, apparatus, electronic device and storage medium. Background Technology

[0002] Ultra-high frequency (UHF) partial discharge (PD) monitoring technology is currently the primary method for identifying the types of internal partial discharge defects in GIS (Gas Insulated Switch) equipment. During UHF detection, accumulated PRPD (Phase-Resolved Partial Discharge) maps provide statistical information on the phase, amplitude, and number of discharges of the PD signal, and are widely used in GIS PD type identification. However, because the PRPD map characteristics vary depending on the degree of internal defect in GIS equipment, PRPD map analysis requires comparison with historical defects in the GIS equipment. This involves searching historical PRPD maps based on the current PRPD map. Currently, this comparison is done manually, requiring significant manpower and resources, resulting in extremely low search efficiency. Summary of the Invention

[0003] In view of this, this application provides a similar map search method, apparatus, electronic device and storage medium for automatically searching for target historical PRPD maps similar to the currently acquired PRPD map from multiple historical PRPD maps, so as to improve search efficiency.

[0004] To achieve the above objectives, the following solution is proposed:

[0005] A similarity graph search method, applied to electronic devices, includes the following steps:

[0006] The collected original PRPD maps to be matched are normalized using a pre-trained normalization network to obtain a binary semantic segmentation map;

[0007] Extract multiple relevance parameters from the binary semantic segmentation graph;

[0008] Based on the multiple correlation parameters, the correlation parameters between the binary semantic segmentation graph and each historical PRPD graph in the historical graph database are calculated to obtain multiple graph similarities. The historical graph database includes multiple historical PRPD graphs that have been standardized using the standardization network.

[0009] The historical PRPD map corresponding to the largest map similarity among the multiple map similarities is determined as the search result.

[0010] Optionally, the plurality of correlation parameters include phase correlation parameters and amplitude correlation parameters that are respectively related to the phase and amplitude of the PRPD spectrum.

[0011] Optionally, the step of extracting multiple relevance parameters from the binary semantic segmentation map includes the following steps:

[0012] The binarized semantic segmentation map is obtained by setting each background pixel to 0 and each partial discharge pixel to 1.

[0013] The phase correlation parameter P and amplitude correlation parameter A are obtained by calculating the binarized spectrum based on the following formulas:

[0014]

[0015] Where g(i,j) is the label value at coordinate (i,j) in the binary map, PN is the total number of phase coordinates in the binary semantic segmentation map, and AN is the total number of amplitude coordinates in the binary semantic segmentation map.

[0016] Optionally, the historical map database includes multiple sub-databases.

[0017] Optionally, the plurality of sub-databases include a burr tip map sub-database, a suspended electrode map sub-database, a metal particle map sub-database, and an insulation defect map sub-database.

[0018] Optionally, the step of calculating the correlation parameters between the standardized map and each historical PRPD map in the historical map database based on the multiple correlation parameters to obtain multiple map similarities includes the following steps:

[0019] The discharge types of the standardized spectrum are analyzed and calculated;

[0020] Select a target sub-database from the plurality of sub-databases according to the discharge type;

[0021] The similarity between the standardized map and each historical PRPD map in the target sub-database is calculated sequentially based on the multiple correlation parameters to obtain the similarity of the multiple maps.

[0022] Optionally, the discharge type is burr tip discharge, floating electrode discharge, metal particle discharge, or insulation defect discharge.

[0023] A similarity map search device, applied to an electronic device, the similarity map search device comprising:

[0024] The standardization processing module is configured to use a pre-trained standardization network to standardize the collected original PRPD map to be matched, and obtain a binary semantic segmentation map.

[0025] The parameter extraction module is configured to extract multiple relevance parameters from the binary semantic segmentation graph;

[0026] The similarity calculation module is configured to calculate the correlation parameters between the binary semantic segmentation graph and each historical PRPD graph in the historical graph database based on the multiple correlation parameters, thereby obtaining multiple graph similarities. The historical graph database includes multiple historical PRPD graphs that have been standardized using the standardization network.

[0027] The target selection module is configured to determine the historical PRPD map corresponding to the largest map similarity among the multiple map similarities as the search result.

[0028] An electronic device includes at least one processor and a memory connected to the processor, wherein:

[0029] The memory is used to store computer programs or instructions;

[0030] The processor is used to execute the computer program or instructions to enable the electronic device to implement the interference pattern recognition method as described above.

[0031] A computer-readable storage medium is applied to an electronic device, the storage medium carrying one or more computer programs that can be executed by the electronic device, thereby enabling the electronic device to implement the interference pattern identification method as described above.

[0032] As can be seen from the above technical solution, this application discloses a similarity map search method, apparatus, electronic device, and storage medium. The method and apparatus are applied to electronic devices, specifically by using a pre-trained standardization network to standardize the collected original PRPD map to be matched, obtaining a binary semantic segmentation map; extracting multiple relevance parameters from the binary semantic segmentation map; calculating the relevance parameters between the binary semantic segmentation map and each historical PRPD map in the historical map database based on these multiple relevance parameters, obtaining multiple map similarities. The historical map database includes multiple historical PRPD maps that have been standardized using the standardization network; and selecting the historical PRPD map with the highest map similarity among the multiple map similarities as the target historical PRPD map similar to the original PRPD map. This solution allows for comparison and search without manual intervention, saving significant manpower and resources, thus improving search efficiency. Attached Figure Description

[0033] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0034] Figure 1 This is a flowchart of a similarity map search method according to an embodiment of this application;

[0035] Figure 2 This is a schematic diagram of the original PRPD map in an embodiment of this application;

[0036] Figure 3 This is a schematic diagram of a portion of the historical PRPD map dataset in an embodiment of this application;

[0037] Figure 4 This is a schematic diagram of a portion of the standardized historical PRPD map dataset according to an embodiment of this application.

[0038] Figure 5 This is a schematic diagram of the search results obtained in an embodiment of this application;

[0039] Figure 6 This is a flowchart of a similarity map search method according to an embodiment of this application;

[0040] Figure 7 This is a block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0042] Similar PRPD map search is not merely a simple contour comparison. When the phase statistical distribution of partial discharge signals within a PRPD map is similar, but the phase interval where the signals converge shifts, the morphological features of the PRPD map change from an image perspective, rendering traditional two-dimensional image-based similarity search algorithms ineffective. Therefore, this invention designs a PRPD map search algorithm based on similarity matching. First, the existing historical PRPD map database and the PRPD map to be searched are standardized. Second, methods for extracting phase correlation parameters and signal amplitude correlation parameters are designed. Finally, a similarity measurement method for PRPD maps that can withstand phase shift changes is constructed to calculate the similarity between the map to be searched and all PRPD maps in the historical map database. This patent converts historically accumulated PRPD map databases from various manufacturers and the PRPD map to be searched to a unified standard, and achieves the search for the most similar historical map when the phase features of the PRPD map shift, significantly reducing the time required for manual experts to review historical partial discharge case maps and improving the efficiency of partial discharge analysis and judgment within GIS equipment based on PRPD maps. Based on the above principles, the following specific implementation method is proposed.

[0043] Figure 1 This is a flowchart of a similarity map search method according to an embodiment of this application.

[0044] like Figure 1 As shown, the similarity map search method provided in this embodiment is applied to an electronic device to find the historical PRPD map with the highest similarity to the original PRPD map to be matched from a historical map database. The electronic device can be understood as a computer, server, cloud platform, or embedded device with data computing and information processing capabilities. The similarity map search method specifically includes the following steps:

[0045] S1. Standardize the original PRPD map to be matched to obtain a binary semantic segmentation map.

[0046] Specifically, a pre-trained standardization network is used to standardize the collected original PRPD maps that need to be matched, resulting in a binary semantic segmentation map. The original PRPD map is shown below. Figure 2 As shown.

[0047] Because existing historical PRPD maps are obtained from various UHF partial discharge devices, the data formats of the corresponding historical PRPD maps are inconsistent. This inconsistency manifests in aspects such as the background color of the maps, the rendering color of the partial discharge pixels, and the background format (with amplitude-phase coordinate grids). Figure 3As shown, this requires training a PRPD graph standardization network by labeling the PRPD standardized dataset to unify the format of historical graphs, laying the foundation for subsequent graph similarity measurements.

[0048] In this application, each PRPD map in the standard dataset used to train the normalized network contains 2000 pairs of local scattering pixel maps. The normalized network uses the PspNet semantic segmentation network for binary classification semantic segmentation of local scattering pixels and background pixels within the PRPD map, with MobileNet as the backbone network for feature extraction. The pyramid pooling in the PSPBlock of PspNet uses average pooling operations with kernel sizes of 1*1, 2*2, 3*3, and 6*6 to train the loss function.

[0049] The parameters for training this standardized network are set as follows: the input PRPD map size is the standard [473, 473], the training parameters are set as batch_size = 4, epoch = 115, learning rate lr = 0.0001, and the trained network is saved as biaozhun.pth.

[0050] In addition, this application requires standardizing the historical PRPD maps accumulated over time, and storing the multiple standardized PRPD maps into a historical map database, Datasets_n = [Data1_n, Data2_n, Data3_n, ..., Datan_n] and the PRPD map to be identified, Data_n. Datasets_n contains n historical PRPD maps, such as... Figure 4 As shown. Furthermore, this historical map database includes multiple sub-databases, specifically including a burr tip map sub-database, a suspended electrode map sub-database, a metal particle map sub-database, and an insulation defect map sub-database.

[0051] S2. Extract multiple relevance parameters from the binary semantic segmentation graph.

[0052] The phase correlation parameter P and signal amplitude correlation parameter A are extracted from the binary semantic segmentation graph obtained after standardization. Input Datasets_n and Data_n, containing n+1 graphs. The total number of phase and amplitude coordinates of the standardized PRPD graphs are uniformly represented as PN and AN. The i-th PRPD graph is binarized, and the histogram results of the PRPD graph in the phase and amplitude coordinate directions are used as the extracted phase correlation parameter P and amplitude correlation parameter A.

[0053] After binarizing the PRPD map, the process only has two values, 0 and 1. The histogram of the binarized PRPD map in phase and amplitude coordinates actually represents the frequency of occurrence of the partial discharge pixel at a fixed phase and amplitude. In the traditional field of image processing, the histogram of the combined horizontal and vertical coordinates has a one-to-one correspondence with the image morphology. Therefore, this process can be regarded as the feature extraction result of the PRPD map.

[0054] First, input the standardized binary semantic segmentation map PRPD_bz from the original map.

[0055] Then, based on PRPD_bz, the positions corresponding to the background pixels in the map are marked as 0, and the positions corresponding to the partial discharge pixels are marked as 1, thus obtaining the corresponding binarized map PRPD_B. The value at the position with coordinates (i,j) in the binarized map is represented as g(i,j).

[0056] Finally, the corresponding phase correlation parameter P and signal amplitude correlation parameter are calculated using the following formulas:

[0057]

[0058] Where g(i,j) is the label value at coordinate (i,j) in the binary map, PN is the total number of phase coordinates in the binary semantic segmentation map, and AN is the total number of amplitude coordinates in the binary semantic segmentation map.

[0059] S3. Calculate the graph similarity between the binary semantic segmentation graph and each historical PRPD graph.

[0060] Based on multiple relevance parameters of the binary semantic segmentation graph, the relevance parameters between the binary semantic segmentation graph and each historical PRPD graph in the historical graph database are calculated to obtain multiple graph similarities. The historical graph database includes multiple historical PRPD graphs that have undergone standardization using the aforementioned standardization network. Each historical PRPD graph is actually a binary semantic segmentation graph and is configured with two relevance parameters obtained through alignment processing. The specific processing procedure is the same as that for the binary semantic segmentation graph above.

[0061] In the specific calculation, the phase correlation parameter P_Data_n and amplitude correlation parameter A_Data_n of the PRPD_n to be identified are input. The phase correlation parameter of the i-th PRPDi_n in the historical database is P_Datai_n, and the amplitude correlation parameter is A_datai_n. A method for calculating the similarity between the spectra of PRPD_n and PRPDi_n is defined, and the similarity result D(data_n, datai_n) is obtained. The specific process is as follows:

[0062] First, input the phase correlation parameter P_Data_n of the original PRPD map to be retrieved and the phase correlation parameter P_Datai_n of the i-th historical PRPD map PRPDi_n in the historical database. The amplitude correlation parameter is a one-dimensional vector, representing the parameters corresponding to PN phase coordinates.

[0063] Then, fixing P_Data_n, PRPDi_n is cyclically shifted according to a∈[1,2,3,...,PN-1]. The phase similarity PP(a) between P_Data_n and PRPDi_n after each cyclic shift of a is defined as follows:

[0064]

[0065] Next, the phase similarity coefficient is defined as PP(a), where the maximum value of a∈[1,2,3,...,PN-1] is max(PP).

[0066] Next, input the phase correlation parameter A_Data_n of the PRPD map to be retrieved and the phase correlation parameter A_datai_n of the i-th PRPD i_n in the historical database. The amplitude correlation parameter is a one-dimensional vector, representing the parameters corresponding to AN amplitude coordinates. The amplitude similarity coefficient is defined as AA, and the formula is as follows:

[0067]

[0068] Finally, the graph similarity D(data_n, datai_n) is obtained using the following formula:

[0069] D(data_n,datai_n)=AA×max(PP).

[0070] The historical PRPD (Precipitation Point Discharge) database of this application includes multiple sub-databases: a burr tip discharge sub-database, a suspended electrode discharge sub-database, a metal particle discharge sub-database, and an insulation defect discharge sub-database. Based on this, when calculating the similarity of the discharge patterns, this application first analyzes and calculates the discharge type of the standardized pattern to determine its discharge type, specifically burr tip discharge, suspended electrode discharge, metal particle discharge, or insulation defect discharge. Then, based on the discharge type, a target sub-database is selected from the multiple sub-databases. Finally, the similarity of the correlation parameters between the standardized pattern and each historical PRPD pattern in the corresponding target sub-database is calculated sequentially based on multiple correlation parameters to obtain the multiple pattern similarities. This process avoids searching all historical PRPD patterns, significantly reducing computational complexity.

[0071] S4. Select the historical PRPD map corresponding to the highest map similarity as the search result.

[0072] The historical PRPD map corresponding to the highest map similarity among the multiple map similarities obtained above is selected as the target historical PRPD map similar to the original PRPD map, thus obtaining the search results, such as... Figure 5 As shown.

[0073] As can be seen from the above technical solution, this embodiment provides a similarity map search method. This method is applied to electronic devices, specifically by using a pre-trained standardization network to standardize the collected original PRPD maps to be matched, obtaining a binary semantic segmentation map; extracting multiple relevance parameters from the binary semantic segmentation map; calculating the relevance parameters between the binary semantic segmentation map and each historical PRPD map in the historical map database based on these multiple relevance parameters, obtaining multiple map similarities. The historical map database includes multiple historical PRPD maps that have been standardized using the standardization network; and selecting the historical PRPD map with the highest map similarity among the multiple map similarities as the target historical PRPD map similar to the original PRPD map. This solution allows for comparison and search without manual intervention, saving significant manpower and resources, thus improving search efficiency.

[0074] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0075] Although the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous.

[0076] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0077] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer.

[0078] Figure 6 This is a block diagram of a similarity map search device according to an embodiment of this application.

[0079] like Figure 6 As shown, the similarity map search device provided in this embodiment is applied to an electronic device to find the historical PRPD map with the highest similarity to the original PRPD map to be matched from the historical map database. The electronic device can be understood as a computer, server, cloud platform or embedded device with data computing and information processing capabilities. The similarity map search device specifically includes a standardization processing module 10, a parameter extraction module 20, a similarity calculation module 30 and a target selection module 40.

[0080] The standardization module is used to standardize the original PRPD map to be matched, resulting in a binary semantic segmentation map.

[0081] Specifically, a pre-trained standardization network is used to standardize the collected original PRPD maps that need to be matched, resulting in a binary semantic segmentation map. The original PRPD map is shown below. Figure 2 As shown.

[0082] Because existing historical PRPD maps are obtained from various UHF partial discharge devices, the data formats of the corresponding historical PRPD maps are inconsistent. This inconsistency manifests in aspects such as the background color of the maps, the rendering color of the partial discharge pixels, and the background format (with amplitude-phase coordinate grids). Figure 3 As shown, this requires training a PRPD graph standardization network by labeling the PRPD standardized dataset to unify the format of historical graphs, laying the foundation for subsequent graph similarity measurements.

[0083] In this application, each PRPD map in the standard dataset used to train the normalized network contains 2000 pairs of local scattering pixel maps. The normalized network uses the PspNet semantic segmentation network for binary classification semantic segmentation of local scattering pixels and background pixels within the PRPD map, with MobileNet as the backbone network for feature extraction. The pyramid pooling in the PSPBlock of PspNet uses average pooling operations with kernel sizes of 1*1, 2*2, 3*3, and 6*6 to train the loss function.

[0084] The parameters for training this standardized network are set as follows: the input PRPD map size is the standard [473, 473], the training parameters are set as batch_size = 4, epoch = 115, learning rate lr = 0.0001, and the trained network is saved as biaozhun.pth.

[0085] In addition, this application requires standardizing the historical PRPD maps accumulated over time, and storing the multiple standardized PRPD maps into a historical map database, Datasets_n = [Data1_n, Data2_n, Data3_n, ..., Datan_n] and the PRPD map to be identified, Data_n. Datasets_n contains n historical PRPD maps, such as... Figure 4 As shown. Furthermore, this historical map database includes multiple sub-databases, specifically including a burr tip map sub-database, a suspended electrode map sub-database, a metal particle map sub-database, and an insulation defect map sub-database.

[0086] The parameter extraction module is used to extract multiple relevant parameters from the binary semantic segmentation graph.

[0087] The phase correlation parameter P and signal amplitude correlation parameter A are extracted from the binary semantic segmentation graph obtained after standardization. Input Datasets_n and Data_n, containing n+1 graphs. The total number of phase and amplitude coordinates of the standardized PRPD graphs are uniformly represented as PN and AN. The i-th PRPD graph is binarized, and the histogram results of the PRPD graph in the phase and amplitude coordinate directions are used as the extracted phase correlation parameter P and amplitude correlation parameter A.

[0088] After binarizing the PRPD map, the process only has two values, 0 and 1. The histogram of the binarized PRPD map in phase and amplitude coordinates actually represents the frequency of occurrence of the partial discharge pixel at a fixed phase and amplitude. In the traditional field of image processing, the histogram of the combined horizontal and vertical coordinates has a one-to-one correspondence with the image morphology. Therefore, this process can be regarded as the feature extraction result of the PRPD map.

[0089] First, input the standardized binary semantic segmentation map PRPD_bz from the original map.

[0090] Then, based on PRPD_bz, the positions corresponding to the background pixels in the map are marked as 0, and the positions corresponding to the partial discharge pixels are marked as 1, thus obtaining the corresponding binarized map PRPD_B. The value at the position with coordinates (i,j) in the binarized map is represented as g(i,j).

[0091] Finally, the corresponding phase correlation parameter P and signal amplitude correlation parameter are calculated using the following formulas:

[0092]

[0093] Where g(i,j) is the label value at coordinate (i,j) in the binary map, PN is the total number of phase coordinates in the binary semantic segmentation map, and AN is the total number of amplitude coordinates in the binary semantic segmentation map.

[0094] The similarity calculation module is used to calculate the graph similarity between the binary semantic segmentation graph and each historical PRPD graph.

[0095] Based on multiple relevance parameters of the binary semantic segmentation graph, the relevance parameters between the binary semantic segmentation graph and each historical PRPD graph in the historical graph database are calculated to obtain multiple graph similarities. The historical graph database includes multiple historical PRPD graphs that have undergone standardization using the aforementioned standardization network. Each historical PRPD graph is actually a binary semantic segmentation graph and is configured with two relevance parameters obtained through alignment processing. The specific processing procedure is the same as that for the binary semantic segmentation graphs described above.

[0096] In the specific calculation, the phase correlation parameter P_Data_n and amplitude correlation parameter A_Data_n of the PRPD_n to be identified are input. The phase correlation parameter of the i-th PRPDi_n in the historical database is P_Datai_n, and the amplitude correlation parameter is A_datai_n. A method for calculating the similarity between the spectra of PRPD_n and PRPDi_n is defined, and the similarity result D(data_n, datai_n) is obtained. The specific process is as follows:

[0097] First, input the phase correlation parameter P_Data_n of the original PRPD map to be retrieved and the phase correlation parameter P_Datai_n of the i-th historical PRPD map PRPDi_n in the historical database. The amplitude correlation parameter is a one-dimensional vector, representing the parameters corresponding to PN phase coordinates.

[0098] Then, fixing P_Data_n, PRPDi_n is cyclically shifted according to a∈[1,2,3,...,PN-1]. The phase similarity PP(a) between P_Data_n and PRPDi_n after each cyclic shift of a is defined as follows:

[0099]

[0100] Next, the phase similarity coefficient is defined as PP(a), where the maximum value of a∈[1,2,3,...,PN-1] is max(PP).

[0101] Next, input the phase correlation parameter A_Data_n of the PRPD map to be retrieved and the phase correlation parameter A_datai_n of the i-th PRPD i_n in the historical database. The amplitude correlation parameter is a one-dimensional vector, representing the parameters corresponding to AN amplitude coordinates. The amplitude similarity coefficient is defined as AA, and the formula is as follows:

[0102]

[0103] Finally, the graph similarity D(data_n, datai_n) is obtained using the following formula:

[0104] D(data_n,datai_n)=AA×max(PP).

[0105] The historical PRPD (Precipitation Point Discharge) database of this application includes multiple sub-databases: a burr tip discharge sub-database, a suspended electrode discharge sub-database, a metal particle discharge sub-database, and an insulation defect discharge sub-database. Based on this, when calculating the similarity of the discharge patterns, this application first analyzes and calculates the discharge type of the standardized pattern to determine its discharge type, specifically burr tip discharge, suspended electrode discharge, metal particle discharge, or insulation defect discharge. Then, based on the discharge type, a target sub-database is selected from the multiple sub-databases. Finally, the similarity of the correlation parameters between the standardized pattern and each historical PRPD pattern in the corresponding target sub-database is calculated sequentially based on multiple correlation parameters to obtain the multiple pattern similarities. This process avoids searching all historical PRPD patterns, significantly reducing computational complexity.

[0106] The target selection module is used to select the historical PRPD map corresponding to the maximum map similarity as the search result.

[0107] The historical PRPD map corresponding to the highest map similarity among the multiple map similarities obtained above is selected as the target historical PRPD map similar to the original PRPD map, thus obtaining the search results, such as... Figure 5 As shown.

[0108] As can be seen from the above technical solution, this embodiment provides a similarity map search device. This device is applied to electronic devices, specifically by using a pre-trained standardization network to standardize the collected original PRPD maps to be matched, obtaining a binary semantic segmentation map; extracting multiple relevance parameters from the binary semantic segmentation map; calculating the relevance parameters between the binary semantic segmentation map and each historical PRPD map in the historical map database based on the multiple relevance parameters, obtaining multiple map similarities, where the historical map database includes multiple historical PRPD maps that have been standardized using the standardization network; and selecting the historical PRPD map with the highest map similarity among the multiple map similarities as the target historical PRPD map similar to the original PRPD map. This solution allows for comparison and search without manual intervention, saving significant manpower and resources, thereby improving search efficiency.

[0109] The units described in the embodiments of this disclosure can be implemented in software or in hardware. The name of a unit does not necessarily limit the unit itself; for example, the first acquisition unit can also be described as "a unit that acquires at least two Internet Protocol addresses".

[0110] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0111] Figure 7 This is a block diagram of an electronic device according to an embodiment of this application.

[0112] The following is for reference. Figure 7 This document illustrates a structural diagram suitable for implementing the electronic device in the embodiments of this disclosure. The terminal device in the embodiments of this disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. This electronic device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this disclosure.

[0113] The electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from an input device 706 into a random access memory (RAM) 703. The RAM also stores various programs and data required for the operation of the electronic device. The processing unit, ROM, and RAM are interconnected via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0114] Typically, the following devices can be connected to the I / O interface: input devices including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 707 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 708 including, for example, magnetic tapes, hard disks, etc.; and communication devices 709. Communication device 709 allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although electronic devices with various devices are shown in the figures, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0115] This application also provides an embodiment of a computer-readable storage medium.

[0116] The aforementioned computer-readable storage medium is applied to an electronic device and carries one or more computer programs. When these programs are executed by the electronic device, the device uses a pre-trained standardization network to standardize the acquired original PRPD maps to be matched, obtaining a binary semantic segmentation map. Multiple relevance parameters are extracted from the binary semantic segmentation map. Based on these parameters, relevance parameters are calculated between the binary semantic segmentation map and each historical PRPD map in a historical map database, resulting in multiple map similarities. The historical map database includes multiple historical PRPD maps that have been standardized using the standardization network. The historical PRPD map with the highest similarity among these similarities is selected as the target historical PRPD map similar to the original PRPD map. This solution eliminates the need for manual comparison and search, reducing the need for significant manpower and resources, thus improving search efficiency.

[0117] It should be noted that the computer-readable medium described above in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0118] In this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0119] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0120] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0121] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0122] The technical solution provided by the present invention has been described in detail above. Specific examples have been used to illustrate the principle and implementation of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A similarity graph search method, applied to electronic devices, characterized in that, The similarity map search method includes the following steps: The collected original PRPD maps to be matched are normalized using a pre-trained normalization network to obtain a binary semantic segmentation map; Extract multiple relevance parameters from the binary semantic segmentation graph; Based on the multiple correlation parameters, the correlation parameters between the binary semantic segmentation graph and each historical PRPD graph in the historical graph database are calculated to obtain multiple graph similarities. The historical graph database includes multiple historical PRPD graphs that have been standardized using the standardization network. The historical PRPD map corresponding to the largest map similarity among the multiple map similarities is determined as the search result.

2. The similarity map search method as described in claim 1, characterized in that, The plurality of correlation parameters include phase correlation parameters and amplitude correlation parameters, which are respectively related to the phase and amplitude of the PRPD spectrum.

3. The similarity map search method as described in claim 2, characterized in that, The extraction of multiple relevance parameters from the binary semantic segmentation map includes the following steps: The binarized semantic segmentation map is obtained by setting each background pixel to 0 and each partial discharge pixel to 1. The phase correlation parameter P and amplitude correlation parameter A are obtained by calculating the binarized spectrum based on the following formulas: Where g(i,j) is the label value at coordinate (i,j) in the binary map, PN is the total number of phase coordinates in the binary semantic segmentation map, and AN is the total number of amplitude coordinates in the binary semantic segmentation map.

4. The similarity map search method as described in claim 1, characterized in that, The historical map database includes multiple sub-databases.

5. The similarity map search method as described in claim 4, characterized in that, The multiple sub-databases include a burr tip map database, a suspended electrode map database, a metal particle map database, and an insulation defect map database.

6. The similarity map search method as described in claim 5, characterized in that, The process of calculating the correlation parameters between the standardized map and each historical PRPD map in the historical map database based on the multiple correlation parameters to obtain multiple map similarities includes the following steps: The discharge types of the standardized spectrum are analyzed and calculated; Select a target sub-database from the plurality of sub-databases according to the discharge type; The similarity between the standardized map and each historical PRPD map in the target sub-database is calculated sequentially based on the multiple correlation parameters to obtain the similarity of the multiple maps.

7. The similarity map search method as described in claim 6, characterized in that, The discharge types are burr tip discharge, floating electrode discharge, metal particle discharge, or insulation defect discharge.

8. A similarity map search device, applied to electronic equipment, characterized in that, The similarity map search device includes: The standardization processing module is configured to use a pre-trained standardization network to standardize the collected original PRPD map to be matched, and obtain a binary semantic segmentation map. The parameter extraction module is configured to extract multiple relevance parameters from the binary semantic segmentation graph; The similarity calculation module is configured to calculate the correlation parameters between the binary semantic segmentation graph and each historical PRPD graph in the historical graph database based on the multiple correlation parameters, thereby obtaining multiple graph similarities. The historical graph database includes multiple historical PRPD graphs that have been standardized using the standardization network. The target selection module is configured to determine the historical PRPD map corresponding to the largest map similarity among the multiple map similarities as the search result.

9. An electronic device, characterized in that, The electronic device includes at least one processor and a memory connected to the processor, wherein: The memory is used to store computer programs or instructions; The processor is used to execute the computer program or instructions to enable the electronic device to implement the interference pattern recognition method as described in any one of claims 1 to 7.

10. A computer-readable storage medium for use in electronic devices, characterized in that, The storage medium carries one or more computer programs that can be executed by the electronic device, thereby enabling the electronic device to implement the interference pattern recognition method as described in any one of claims 1 to 7.

Images