Converter station equipment operation and maintenance method based on space calculation, computer equipment and program product

Abstract

The invention relates to a converter station equipment operation and maintenance method and device based on space calculation, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring equipment information, target position information and environment information of to-be-analyzed equipment in a to-be-analyzed converter station; pre-processing the equipment information, the target position information and the environment information to obtain pre-processed equipment information, pre-processed position information and pre-processed environment information; performing feature extraction processing and fusion processing on the preprocessed equipment information, the preprocessed position information and the preprocessed environment information to obtain a fusion feature vector corresponding to the to-be-analyzed equipment; inputting the fusion feature vector into a pre-trained equipment operation and maintenance instruction prediction model to obtain a target equipment operation and maintenance instruction corresponding to the to-be-analyzed equipment; and performing corresponding operation and maintenance processing on the to-be-analyzed equipment according to the target equipment operation and maintenance instruction. By adopting the method, the operation and maintenance efficiency of converter station equipment can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of power grids, and in particular, to a method, device, computer device, computer-readable storage medium, and computer program product for operation and maintenance of converter station equipment based on spatial computing. Background Art

[0002] Currently, in order to ensure the safe and stable operation of a converter station, it is crucial to perform operation and maintenance on the converter station equipment.

[0003] In traditional technologies, when performing operation and maintenance on converter station equipment, manual inspection is generally adopted; however, this method is rather cumbersome and requires a large amount of time and manpower, resulting in low operation and maintenance efficiency of converter station equipment. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a method, device, computer device, computer-readable storage medium, and computer program product for operation and maintenance of converter station equipment based on spatial computing, which can improve the operation and maintenance efficiency of converter station equipment.

[0005] In a first aspect, the present application provides a method for operation and maintenance of converter station equipment based on spatial computing, including:

[0006] Obtaining device information, target location information, and environmental information of a device to be analyzed in a converter station to be analyzed;

[0007] Preprocessing the device information, the target location information, and the environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information;

[0008] Performing feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information;

[0009] Performing fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed;

[0010] Inputting the fusion feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain a target device operation and maintenance instruction corresponding to the device to be analyzed;

[0011] Performing corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0012] In one embodiment, obtaining the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed includes:

[0013] Obtaining the initial location information of the device to be analyzed in the converter station to be analyzed;

[0014] Performing conversion processing on the initial location information according to a preset correspondence relationship to obtain the target location information; the preset correspondence relationship is used to represent the correspondence relationship between the initial location information and the target location information.

[0015] In one embodiment, respectively performing feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environmental information to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information includes:

[0016] Taking the preprocessed device information as the main data, and taking the preprocessed location information and the preprocessed environmental information as auxiliary data, and inputting them into a feature extraction model for feature extraction processing to obtain the first feature vector;

[0017] Taking the preprocessed location information as the main data, and taking the preprocessed device information and the preprocessed environmental information as auxiliary data, and inputting them into the feature extraction model for feature extraction processing to obtain the second feature vector;

[0018] Taking the preprocessed environmental information as the main data, and taking the preprocessed device information and the preprocessed location information as auxiliary data, and inputting them into the feature extraction model for feature extraction processing to obtain the third feature vector.

[0019] In one embodiment, performing fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed includes:

[0020] Obtaining a first initial weight corresponding to the preprocessed device information, a second initial weight corresponding to the preprocessed location information, and a third initial weight corresponding to the preprocessed environmental information;

[0021] Performing normalization processing on the first initial weight, the second initial weight, and the third initial weight to obtain a first target weight corresponding to the preprocessed device information, a second target weight corresponding to the preprocessed location information, and a third target weight corresponding to the preprocessed environmental information;

[0022] Fuse the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the fused feature vector.

[0023] In one embodiment, before obtaining the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed, it further includes:

[0024] Obtain the voltage level information and device type information of the candidate converter stations;

[0025] Screen out the candidate converter stations that meet the preset conditions for both the voltage level information and the device type information from each of the candidate converter stations as the converter station to be analyzed.

[0026] In one embodiment, the pre-trained device operation and maintenance instruction prediction model is trained in the following manner:

[0027] Obtain the sample device information, sample location information, and sample environmental information of the sample devices in the sample converter station;

[0028] Preprocess the sample device information, the sample location information, and the sample environmental information respectively to obtain the preprocessed sample device information, the preprocessed sample location information, and the preprocessed sample environmental information;

[0029] Extract feature vectors from the preprocessed sample device information, the preprocessed sample location information, and the preprocessed sample environmental information respectively to obtain a first sample feature vector corresponding to the preprocessed sample device information, a second sample feature vector corresponding to the preprocessed sample location information, and a third sample feature vector corresponding to the preprocessed sample environmental information;

[0030] Fuse the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain the sample fused feature vector corresponding to the sample device;

[0031] Input the sample fused feature vector into the device operation and maintenance instruction prediction model to be trained to obtain the predicted device operation and maintenance instruction corresponding to the sample device;

[0032] Obtain the actual device operation and maintenance instruction corresponding to the sample device, and perform iterative training on the device operation and maintenance instruction prediction model to be trained according to the difference between the predicted device operation and maintenance instruction and the actual device operation and maintenance instruction to obtain the pre-trained device operation and maintenance instruction prediction model.

[0033] In a second aspect, the present application also provides a converter station device operation and maintenance device based on spatial computing, including:

[0034] An information acquisition module, configured to acquire device information, target location information, and environmental information of a device to be analyzed in a converter station to be analyzed;

[0035] An information processing module, configured to preprocess the device information, the target location information, and the environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information;

[0036] A feature extraction module, configured to perform feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information;

[0037] A feature fusion module, configured to perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed;

[0038] An instruction prediction module, configured to input the fusion feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain a target device operation and maintenance instruction corresponding to the device to be analyzed;

[0039] A device operation and maintenance module, configured to perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0040] In a third aspect, the present application further provides a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0041] Acquire device information, target location information, and environmental information of a device to be analyzed in a converter station to be analyzed;

[0042] Preprocess the device information, the target location information, and the environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information;

[0043] Perform feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information;

[0044] Perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed;

[0045] Input the fused feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain the target device operation and maintenance instruction corresponding to the device to be analyzed;

[0046] Perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0047] Fourthly, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0048] Obtain the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed;

[0049] Preprocess the device information, the target location information, and the environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information;

[0050] Extract feature vectors from the preprocessed device information, the preprocessed location information, and the preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information;

[0051] Fuse the first feature vector, the second feature vector, and the third feature vector to obtain a fused feature vector corresponding to the device to be analyzed;

[0052] Input the fused feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain the target device operation and maintenance instruction corresponding to the device to be analyzed;

[0053] Perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0054] Fifthly, the present application also provides a computer program product, including a computer program. When the computer program is executed by a processor, the following steps are implemented:

[0055] Obtain the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed;

[0056] Preprocess the device information, the target location information, and the environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information;

[0057] Perform feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information;

[0058] Perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed;

[0059] Input the fusion feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain a target device operation and maintenance instruction corresponding to the device to be analyzed;

[0060] Perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0061] The above-mentioned HVDC converter station device operation and maintenance method, device, computer device, storage medium, and computer program product based on spatial computing first obtain the device information, target location information, and environmental information of the device to be analyzed in the HVDC converter station to be analyzed, and preprocess the device information, target location information, and environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information, and perform feature extraction processing on the preprocessed device information, preprocessed location information, and preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information. Then, perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed. Then, input the fusion feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain a target device operation and maintenance instruction corresponding to the device to be analyzed. Finally, perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction. In this way, when performing operation and maintenance processing on the HVDC converter station device, after obtaining the device information, target location information, and environmental information of the device to be analyzed in the HVDC converter station to be analyzed, through a series of processes such as preprocessing, feature extraction processing, and fusion processing, based on a pre-trained device operation and maintenance instruction prediction model, the target device operation and maintenance instruction corresponding to the device to be analyzed can be quickly and directly obtained, which is beneficial to improving the determination efficiency of the device operation and maintenance instruction of the HVDC converter station device, and further improves the operation and maintenance efficiency of the HVDC converter station device; moreover, the entire process does not require manual intervention, avoiding the defects that the method of manual inspection is relatively cumbersome, requires a lot of time and manpower, and results in low operation and maintenance efficiency of the HVDC converter station device, and further improves the operation and maintenance efficiency of the HVDC converter station device. Description of the Drawings

[0062] To more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the following will briefly introduce the drawings required for the description in the embodiments of the present application or the related art. Obviously, the drawings described below are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0063] Figure 1 It is a schematic flowchart of a method for operation and maintenance of converter station equipment based on spatial computing in one embodiment;

[0064] Figure 2 It is a schematic flowchart of a method for operation and maintenance of converter station equipment based on spatial computing in another embodiment;

[0065] Figure 3 It is a schematic diagram of an intelligent interaction system in one embodiment;

[0066] Figure 4 It is a schematic diagram of the structural principle of a binocular vision measurement system in one embodiment;

[0067] Figure 5 It is a schematic diagram of the coordinate system conversion of LiDAR point cloud data in one embodiment;

[0068] Figure 6 It is a schematic diagram of the overall technical route of an intelligent interaction module in one embodiment;

[0069] Figure 7 It is a schematic diagram of the overall solution of intelligent operation assistance in one embodiment;

[0070] Figure 8 It is a schematic diagram of the technical framework for job result recognition in one embodiment;

[0071] Figure 9 It is a structural block diagram of a device for operation and maintenance of converter station equipment based on spatial computing in one embodiment;

[0072] Figure 10 It is an internal structure diagram of a computer device in one embodiment. Detailed implementation manners

[0073] In order to make the purpose, technical solutions and advantages of the present application clearer, the following further details the present application in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0074] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0075] In an exemplary embodiment, as Figure 1 shown, a method for operation and maintenance of converter station equipment based on spatial computing is provided. In this embodiment, an example is given where this method is applied to a server; it can be understood that this method can also be applied to a terminal, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. Among them, the terminal can be, but is not limited to, various personal computers, laptop computers, smart phones, and tablet computers; the server can be implemented by an independent server or a server cluster composed of multiple servers. In this embodiment, the method includes the following steps:

[0076] Step S101, obtain the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed.

[0077] Among them, the converter station to be analyzed refers to the converter station that needs equipment operation and maintenance.

[0078] Among them, the device to be analyzed refers to the converter station equipment that needs operation and maintenance.

[0079] Among them, the device information includes the device name, device model, etc. of the device to be analyzed. It should be noted that the device information is obtained by identifying the nameplate and pressure plate of the device to be analyzed.

[0080] Among them, the target location information is used to represent the coordinate information of the device to be analyzed in the LiDAR (Light Detection and Ranging) coordinate system.

[0081] Among them, the environmental information is used to represent the information around the device to be analyzed, including the location information of the devices around the device to be analyzed.

[0082] Exemplarily, the server determines the target aging degree of each candidate device in the converter station to be analyzed according to the aging degree of the key components in each candidate device; then, the server screens out the candidate devices with a target aging degree greater than the preset aging degree from each candidate device as the devices to be analyzed in the converter station to be analyzed; then, the server identifies the nameplate and pressure plate of the device to be analyzed through sensors to obtain the device information of the device to be analyzed; then, the server obtains the coordinate information of the device to be analyzed in the LiDAR coordinate system through the LiDAR measurement system as the target position information of the device to be analyzed; then, the server uses the position information of the devices within the preset range of the device to be analyzed as the environmental information of the device to be analyzed.

[0083] Step S102: Preprocess the device information, target position information, and environmental information respectively to obtain the preprocessed device information, preprocessed position information, and preprocessed environmental information.

[0084] Among them, the preprocessed device information refers to the device information after preprocessing.

[0085] Among them, the preprocessed position information refers to the target position information after preprocessing.

[0086] Among them, the preprocessed environmental information refers to the environmental information after preprocessing.

[0087] Exemplarily, the server performs noise removal, multi-view alignment, data reduction, surface reconstruction, etc. on the device information, target position information, and environmental information respectively to obtain the preprocessed device information, preprocessed position information, and preprocessed environmental information.

[0088] Step S103: Perform feature extraction processing on the preprocessed device information, preprocessed position information, and preprocessed environmental information respectively to obtain the first feature vector corresponding to the preprocessed device information, the second feature vector corresponding to the preprocessed position information, and the third feature vector corresponding to the preprocessed environmental information.

[0089] Among them, the first feature vector refers to the feature vector corresponding to the preprocessed device information.

[0090] Among them, the second feature vector refers to the feature vector corresponding to the preprocessed position information.

[0091] Among them, the third feature vector refers to the feature vector corresponding to the preprocessed environmental information.

[0092] Exemplarily, the server inputs the preprocessed device information, preprocessed location information, and preprocessed environmental information into the feature extraction model respectively. Through the feature extraction model, feature extraction processing is performed on the preprocessed device information, preprocessed location information, and preprocessed environmental information respectively, to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information.

[0093] Step S104: Perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed.

[0094] The fusion feature vector refers to the feature vector obtained by performing fusion processing on the first feature vector, the second feature vector, and the third feature vector.

[0095] Exemplarily, the server performs weighted summation processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed.

[0096] Step S105: Input the fusion feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain a target device operation and maintenance instruction corresponding to the device to be analyzed.

[0097] The device operation and maintenance instruction prediction model refers to a network model that can use the fusion feature vector corresponding to the device to be analyzed to obtain the target device operation and maintenance instruction corresponding to the device to be analyzed, such as a recurrent neural network model, a convolutional neural network model, etc.

[0098] The target device operation and maintenance instruction refers to the instruction information for performing operation and maintenance on the device to be analyzed, such as a device inspection instruction, a device maintenance instruction, etc.

[0099] Exemplarily, the server inputs the corresponding one of the devices to be analyzed into the pre-trained device operation and maintenance instruction prediction model to obtain multiple preset device operation and maintenance instructions corresponding to the device to be analyzed, and the prediction probability corresponding to each preset device operation and maintenance instruction; then, the server screens out the preset device operation and maintenance instruction with the largest corresponding prediction probability from each preset device operation and maintenance instruction as the target device operation and maintenance instruction corresponding to the device to be analyzed.

[0100] Step S106: Perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0101] Exemplarily, the server performs an integrity check on the target device operation and maintenance instruction (for example, the server checks whether key information such as device identification, instruction type, and execution parameters is included in the target device operation and maintenance instruction), and obtains the integrity check result corresponding to the target device operation and maintenance instruction; then, when the integrity check result passes, the server performs corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0102] In the above method for operation and maintenance of converter station equipment based on spatial computing, first, the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed are obtained, and the device information, target location information, and environmental information are respectively preprocessed to obtain the preprocessed device information, preprocessed location information, and preprocessed environmental information, and feature extraction processing is respectively performed on the preprocessed device information, preprocessed location information, and preprocessed environmental information to obtain the first feature vector corresponding to the preprocessed device information, the second feature vector corresponding to the preprocessed location information, and the third feature vector corresponding to the preprocessed environmental information. Then, the first feature vector, the second feature vector, and the third feature vector are fused to obtain the fused feature vector corresponding to the device to be analyzed. Then, the fused feature vector is input into the pre-trained device operation and maintenance instruction prediction model to obtain the target device operation and maintenance instruction corresponding to the device to be analyzed. Finally, corresponding operation and maintenance processing is performed on the device to be analyzed according to the target device operation and maintenance instruction. In this way, when performing operation and maintenance processing on converter station equipment, after obtaining the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed, through a series of processes such as preprocessing, feature extraction processing, and fusion processing, based on the pre-trained device operation and maintenance instruction prediction model, the target device operation and maintenance instruction corresponding to the device to be analyzed can be quickly and directly obtained, which is beneficial to improving the determination efficiency of the device operation and maintenance instruction of converter station equipment, and further improving the operation and maintenance efficiency of converter station equipment; moreover, the entire process does not require manual intervention, avoiding the defects that the method of manual inspection is relatively cumbersome, requires a large amount of time and manpower, and results in low operation and maintenance efficiency of converter station equipment, and further improves the operation and maintenance efficiency of converter station equipment.

[0103] In an exemplary embodiment, the above step S101 of obtaining the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed specifically includes the following content: obtaining the initial location information of the device to be analyzed in the converter station to be analyzed; performing conversion processing on the initial location information according to a preset correspondence relationship to obtain the target location information; the preset correspondence relationship is used to represent the correspondence relationship between the initial location information and the target location information.

[0104] Among them, the initial location information is used to represent the coordinate information of the device to be analyzed in the camera coordinate system.

[0105] Among them, the preset correspondence is used to represent the correspondence between the initial position information and the target position information. In an actual scenario, the preset correspondence refers to the coordinate conversion relationship between the LiDAR and the camera.

[0106] Exemplarily, the server obtains the coordinate information of the device to be analyzed in the camera coordinate system as the initial position information of the device to be analyzed in the converter station to be analyzed; then, the server takes the correspondence between the initial position information and the target position information as the preset correspondence; then, the server performs conversion processing on the initial position information according to the preset correspondence to obtain the target position information.

[0107] In this embodiment, by performing conversion processing on the initial position information according to the preset correspondence to obtain the target position information, the initial position information in different formats or coordinate systems can be uniformly converted into the target position information that meets the specific format requirements, thereby enabling data standardization in the entire analysis and avoiding the defect that data processing is prone to errors due to inconsistent position information formats.

[0108] In an exemplary embodiment, in step S103 above, feature extraction processing is respectively performed on the preprocessed device information, preprocessed position information, and preprocessed environment information to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed position information, and a third feature vector corresponding to the preprocessed environment information. The specific content is as follows: Taking the preprocessed device information as the main data, and the preprocessed position information and preprocessed environment information as auxiliary data, inputting them into a feature extraction model for feature extraction processing to obtain the first feature vector; taking the preprocessed position information as the main data, and the preprocessed device information and preprocessed environment information as auxiliary data, inputting them into a feature extraction model for feature extraction processing to obtain the second feature vector; taking the preprocessed environment information as the main data, and the preprocessed device information and preprocessed position information as auxiliary data, inputting them into a feature extraction model for feature extraction processing to obtain the third feature vector.

[0109] Among them, the main data can refer to the data with a relatively large corresponding weight.

[0110] Among them, the auxiliary data can refer to the data with a relatively small corresponding weight.

[0111] Exemplarily, the server takes the preprocessed device information as the main data, and the preprocessed location information and preprocessed environmental information as auxiliary data, and inputs them into the feature extraction model for feature extraction processing to obtain the feature vector corresponding to the preprocessed device information as the first feature vector; takes the preprocessed location information as the main data, and the preprocessed device information and preprocessed environmental information as auxiliary data, and inputs them into the feature extraction model for feature extraction processing to obtain the feature vector corresponding to the preprocessed location information as the second feature vector; takes the preprocessed environmental information as the main data, and the preprocessed device information and preprocessed location information as auxiliary data, and inputs them into the feature extraction model for feature extraction processing to obtain the feature vector corresponding to the preprocessed environmental information as the third feature vector.

[0112] In this embodiment, in the process of feature extraction processing of the preprocessed device information, preprocessed location information, and preprocessed environmental information, it is equivalent to considering multiple different data at the same time, so that the extracted feature vectors are more comprehensive, which is conducive to improving the determination accuracy of the feature vectors corresponding to the first feature vector, the second feature vector, and the third feature vector.

[0113] In an exemplary embodiment, in step S104 above, the first feature vector, the second feature vector, and the third feature vector are fused to obtain the fused feature vector corresponding to the device to be analyzed, which specifically includes the following content: obtaining the first initial weight corresponding to the preprocessed device information, the second initial weight corresponding to the preprocessed location information, and the third initial weight corresponding to the preprocessed environmental information; normalizing the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the preprocessed device information, the second target weight corresponding to the preprocessed location information, and the third target weight corresponding to the preprocessed environmental information; and fusing the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain the fused feature vector.

[0114] Among them, the first initial weight refers to the weight initially assigned to the preprocessed device information.

[0115] Among them, the second initial weight refers to the weight initially assigned to the preprocessed location information.

[0116] Among them, the third initial weight refers to the weight initially assigned to the preprocessed environmental information.

[0117] Among them, the first target weight refers to the weight finally assigned to the preprocessed device information.

[0118] Among them, the second target weight refers to the weight finally assigned to the preprocessed location information.

[0119] Among them, the third target weight refers to the weight finally assigned to the preprocessed environmental information.

[0120] Exemplarily, the server determines the first initial weight corresponding to the preprocessed device information, the second initial weight corresponding to the preprocessed location information, and the third initial weight corresponding to the preprocessed environmental information according to the respective importance levels of the preprocessed device information, the preprocessed location information, and the preprocessed environmental information (for example, determined based on prior knowledge and experience); then, the server normalizes the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the preprocessed device information, the second target weight corresponding to the preprocessed location information, and the third target weight corresponding to the preprocessed environmental information; for example, the first initial weight is 0.8, the second initial weight is 0.6, and the third initial weight is 0.6. After normalization, the first target weight corresponding to the preprocessed device information is 0.4, the second target weight corresponding to the preprocessed location information is 0.3, and the third target weight corresponding to the preprocessed environmental information is 0.3; then, the server fuses the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain a fused feature vector; for example, the server performs a weighted summation process on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain a fused feature vector.

[0121] In this embodiment, by using the target weight for fusion processing, a clear proportional relationship can be provided for different feature vectors during the fusion process, so that the obtained fused feature vector can accurately reflect the contribution degree of each piece of information, which is beneficial to improving the determination accuracy of the fused feature vector.

[0122] In an exemplary embodiment, before the above step S101 of obtaining the device information, the target location information, and the environmental information of the device to be analyzed in the converter station to be analyzed, the following specific contents are included: obtaining the voltage level information and the device type information of the candidate converter stations; screening out the candidate converter stations whose voltage level information and device type information both meet the preset conditions from each candidate converter station as the converter station to be analyzed.

[0123] Among them, the candidate converter station refers to the converter station to be selected.

[0124] Among them, the voltage level information refers to the voltage level of the candidate converter station, including high voltage, medium voltage, low voltage, etc.

[0125] Among them, the device type information refers to the number of types of devices in the candidate converter station, such as 10.

[0126] Among them, the preset condition refers to a pre-set judgment condition used to judge the voltage level information and equipment type information. For example, the voltage level information is greater than or equal to medium voltage, and the equipment type information is greater than 5. It should be noted that the preset condition can be determined according to the situation.

[0127] Exemplarily, the server extracts the voltage level information and equipment type information of the candidate converter stations from the converter station information of the candidate converter stations; then, the server screens out the candidate converter stations whose voltage level information and equipment type information both meet the preset conditions from each candidate converter station, and takes these candidate converter stations as the converter stations to be analyzed.

[0128] In this embodiment, by obtaining the voltage level information and equipment type information of the candidate converter stations and screening according to the preset conditions, it is possible to accurately find the converter stations that meet the conditions from multiple candidate converter stations, avoiding wasting time and resources analyzing on irrelevant or unqualified converter stations.

[0129] In an exemplary embodiment, the converter station equipment operation and maintenance method provided by the present application based on spatial computing further includes the training steps of a pre-trained equipment operation and maintenance instruction prediction model, which specifically includes the following content: obtaining the sample equipment information, sample location information, and sample environment information of the sample equipment in the sample converter station; respectively preprocessing the sample equipment information, sample location information, and sample environment information to obtain the preprocessed sample equipment information, preprocessed sample location information, and preprocessed sample environment information; respectively performing feature extraction processing on the preprocessed sample equipment information, preprocessed sample location information, and preprocessed sample environment information to obtain a first sample feature vector corresponding to the preprocessed sample equipment information, a second sample feature vector corresponding to the preprocessed sample location information, and a third sample feature vector corresponding to the preprocessed sample environment information; performing fusion processing on the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain a sample fusion feature vector corresponding to the sample equipment; inputting the sample fusion feature vector into the equipment operation and maintenance instruction prediction model to be trained to obtain a predicted equipment operation and maintenance instruction corresponding to the sample equipment; obtaining the actual equipment operation and maintenance instruction corresponding to the sample equipment, and iteratively training the equipment operation and maintenance instruction prediction model to be trained according to the difference between the predicted equipment operation and maintenance instruction and the actual equipment operation and maintenance instruction to obtain the pre-trained equipment operation and maintenance instruction prediction model.

[0130] Among them, the sample converter station refers to the converter station used to train the equipment operation and maintenance instruction prediction model.

[0131] Among them, the sample equipment refers to the converter station equipment used to train the equipment operation and maintenance instruction prediction model.

[0132] Among them, the sample equipment information refers to the equipment information of the sample equipment.

[0133] Among them, the sample location information refers to the location information of the sample device.

[0134] Among them, the sample environment information refers to the environment information of the sample device.

[0135] Among them, the preprocessed sample device information refers to the sample device information after preprocessing.

[0136] Among them, the preprocessed sample location information refers to the sample location information after preprocessing.

[0137] Among them, the preprocessed sample environment information refers to the sample environment information after preprocessing.

[0138] Among them, the first sample feature vector refers to the feature vector corresponding to the preprocessed sample device information.

[0139] Among them, the second sample feature vector refers to the feature vector corresponding to the preprocessed sample location information.

[0140] Among them, the third sample feature vector refers to the feature vector corresponding to the preprocessed sample environment information.

[0141] Among them, the sample fusion feature vector refers to the feature vector obtained by fusing the first sample feature vector, the second sample feature vector, and the third sample feature vector.

[0142] Among them, the predicted device operation and maintenance instruction refers to the predicted value corresponding to the device operation and maintenance instruction of the sample device.

[0143] Among them, the actual device operation and maintenance instruction refers to the actual value corresponding to the device operation and maintenance instruction of the sample device.

[0144] Exemplarily, in response to a model training instruction for a device operation and maintenance instruction prediction model to be trained, the server obtains sample device information, sample location information, and sample environment information of sample devices in a sample converter station from a database; then, the server preprocesses the sample device information, sample location information, and sample environment information respectively to obtain preprocessed sample device information, preprocessed sample location information, and preprocessed sample environment information; then, the server performs feature extraction processing on the preprocessed sample device information, preprocessed sample location information, and preprocessed sample environment information respectively to obtain a first sample feature vector corresponding to the preprocessed sample device information, a second sample feature vector corresponding to the preprocessed sample location information, and a third sample feature vector corresponding to the preprocessed sample environment information; then, the server performs fusion processing on the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain a sample fusion feature vector corresponding to the sample device; then, the server inputs the sample fusion feature vector into the device operation and maintenance instruction prediction model to be trained to obtain a predicted device operation and maintenance instruction corresponding to the sample device; then, the server obtains the actual device operation and maintenance instruction corresponding to the sample device, and obtains a loss value according to the difference between the predicted device operation and maintenance instruction and the actual device operation and maintenance instruction; then, the server adjusts the model parameters of the device operation and maintenance instruction prediction model to be trained according to the loss value; then, the server retrains the device operation and maintenance instruction prediction model with the adjusted model parameters until the loss value obtained by the trained device operation and maintenance instruction prediction model is less than the loss value threshold, then stops training, and uses the trained device operation and maintenance instruction prediction model as the pre-trained device operation and maintenance instruction prediction model.

[0145] In this embodiment, by pre-training the device operation and maintenance instruction prediction model, it is convenient to predict the target device operation and maintenance instruction corresponding to the device to be analyzed after obtaining the fusion feature vector corresponding to the device to be analyzed in actual application; moreover, the device operation and maintenance instruction prediction model receives new data in each iteration, performs internal improvement and optimization of the model, which is convenient for more effective prediction and is beneficial to improving the prediction accuracy of the device operation and maintenance instruction prediction model.

[0146] In an exemplary embodiment, as Figure 2 shown, another method for device operation and maintenance of a converter station based on spatial calculation is provided. Taking the application of this method to a server as an example, the method includes the following steps:

[0147] Step S201, obtain the voltage level information and device type information of the candidate converter stations; from each candidate converter station, screen out the candidate converter stations whose voltage level information and device type information both meet the preset conditions as the converter stations to be analyzed.

[0148] Step S202: Obtain the device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed.

[0149] Step S203: Preprocess the device information, target location information, and environmental information respectively to obtain the preprocessed device information, preprocessed location information, and preprocessed environmental information.

[0150] Step S204: Use the preprocessed device information as the main data, and the preprocessed location information and preprocessed environmental information as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the first feature vector.

[0151] Step S205: Use the preprocessed location information as the main data, and the preprocessed device information and preprocessed environmental information as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the second feature vector.

[0152] Step S206: Use the preprocessed environmental information as the main data, and the preprocessed device information and preprocessed location information as auxiliary data, and input them into the feature extraction model for feature extraction processing to obtain the third feature vector.

[0153] Step S207: Obtain the first initial weight corresponding to the preprocessed device information, the second initial weight corresponding to the preprocessed location information, and the third initial weight corresponding to the preprocessed environmental information.

[0154] Step S208: Perform normalization processing on the first initial weight, the second initial weight, and the third initial weight to obtain the first target weight corresponding to the preprocessed device information, the second target weight corresponding to the preprocessed location information, and the third target weight corresponding to the preprocessed environmental information.

[0155] Step S209: According to the first target weight, the second target weight, and the third target weight, perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain the fused feature vector.

[0156] Step S210: Input the fused feature vector into the pre-trained device operation and maintenance instruction prediction model to obtain the target device operation and maintenance instruction corresponding to the device to be analyzed.

[0157] Step S211: Perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0158] In the above method for the operation and maintenance of converter station equipment based on spatial computing, when performing operation and maintenance processing on converter station equipment, after obtaining the equipment information, target location information, and environmental information of the equipment to be analyzed in the converter station to be analyzed, through a series of processes such as preprocessing, feature extraction processing, and fusion processing, based on a pre-trained equipment operation and maintenance instruction prediction model, the target equipment operation and maintenance instruction corresponding to the equipment to be analyzed can be quickly and directly obtained, which is beneficial to improving the determination efficiency of the equipment operation and maintenance instructions for converter station equipment, and thus improving the operation and maintenance efficiency of converter station equipment; moreover, the entire process does not require manual intervention, avoiding the defects that the method of manual inspection is relatively cumbersome, requires a large amount of time and manpower, and results in low operation and maintenance efficiency of converter station equipment, further improving the operation and maintenance efficiency of converter station equipment.

[0159] In an exemplary embodiment, in order to more clearly illustrate the method for the operation and maintenance of converter station equipment based on spatial computing provided by the embodiments of the present application, the following uses a specific embodiment to specifically describe the method for the operation and maintenance of converter station equipment based on spatial computing. In one embodiment, the present application also provides an intelligent interaction technology for the operation and maintenance information of converter station equipment. When performing operation and maintenance processing on converter station equipment, first obtain the equipment information, target location information, and environmental information of the equipment to be analyzed in the converter station to be analyzed, respectively preprocess the equipment information, target location information, and environmental information to obtain the preprocessed equipment information, preprocessed location information, and preprocessed environmental information, and perform corresponding operation and maintenance processing on the equipment to be analyzed according to the preprocessed equipment information, preprocessed location information, and preprocessed environmental information. The specific contents are as follows:

[0160] The intelligent interaction platform for the operation and maintenance information of converter station equipment involved in this solution uses spatial computing technology to provide a visual, interactive, and automated operation and maintenance solution for converter station equipment. The intelligent interaction platform mainly consists of three parts: a data acquisition module, a data processing module, and an intelligent interaction module. Its overall structure is as Figure 3 shown.

[0161] (1) Data acquisition module: used to drive sensors and transmit sensor data to the data processing module.

[0162] (2) Data processing module: used to process, fuse the data collected by the data acquisition module, and generate a high-precision model of converter station equipment for browsing.

[0163] (3) Intelligent interaction module: Staff can achieve intelligent operation and maintenance of converter station equipment in the intelligent interaction module through intelligent mobile devices.

[0164] To ensure the 3D reconstruction effect of the converter station, it is necessary to obtain multi-view data of the key equipment in the converter station. Therefore, this solution uses binocular stereo vision with 2 cameras to collect 2D images. Each camera corresponds to a different view to obtain the structural data from that view. Finally, the data obtained by the two cameras are unified into the same coordinate system to generate global 3D point cloud data. The principle of binocular stereo vision triangulation is as Figure 4 shown.

[0165] In Figure 4 , O 1 -X 1 Y 1 Z 1 and O 2 -X 2 Y 2 Z 2 are the left and right camera coordinate systems respectively, and O 1 ′-X 1 Y 1 and O 2 ′-X 2 Y 2 are the left and right image coordinate systems respectively. The origin O 1 and O 2 are the optical centers of the two cameras respectively. O 1 Z 1 and O 2 Z 2 are the optical axes of the two cameras. Assume that any point P on the device to be measured is connected to O 1 , O 2 respectively. The projection coordinates of point P on the planes O 1 ′-X 1 Y 1 and O 2 ′-X 2 Y 2 are P 1 (x 1 , y 1 ) and P 2 (x 2 , y 2 ) respectively. Assume that the world coordinate system coincides with the coordinate system of the left camera, then the right camera can be approximated as a monocular camera model. The rotation matrix R and translation matrix t of the camera can be obtained through camera calibration. Combining with the projection transformation formula, the 3D coordinates of point P in the world coordinate system can be obtained as shown in Equations (1) and (2).

[0166]

[0167] Or

[0168]

[0169] In the formula, f1 and f 2 is the focal length of the camera, and r i (i = 1, 2, …, 9) is the rotation component; t x , t y , t z is the translation component. Therefore, the three-dimensional coordinates of any point in space can be solved through the binocular vision theoretical model.

[0170] This solution makes full use of LiDAR to first scan the key equipment in the converter station to obtain complete scanned point cloud data. The specific operation is that LiDAR rotates at a specific speed and continuously emits infrared lasers while receiving the laser signals of the reflection points, including the distance, time, and horizontal angle of the reflection points, etc. One emitter corresponds to one vertical angle, and the position information of the corresponding reflection points can be obtained using these variable data. The set of all reflection point coordinates collected by LiDAR rotating 360° becomes a point cloud, thereby obtaining all-round environmental information.

[0171] The point cloud data obtained by LiDAR can be represented in the form of spherical coordinates (R, ω, α). Among them, ω is the pitch angle, which is a fixed value, and R and α are the distance and azimuth angle of the LiDAR from the point cloud data respectively. In order to implement subsequent data processing, it is usually necessary to convert these point cloud data from the spherical coordinate system to the space Cartesian coordinate system, and its conversion schematic diagram is as Figure 5 shown.

[0172] The coordinate system conversion formula is:

[0173]

[0174] Through the above coordinate system conversion, the data finally collected by LiDAR can be described in the form of a set of multi-dimensional points, where each point p i can be represented by Equation (4). Among them, x i , y i , z i represent the three-dimensional position coordinates of the point cloud data, and s i represents the reflection intensity.

[0175] p i = {x i , y i , z i , s i}, Equation (4)

[0176] The set of point cloud data P containing n points can be represented by Equation (5).

[0177] P = {p 1 , p 2 , p 3,L,p n}), Equation (5)

[0178] Since the data obtained by the data acquisition module is point cloud data, which samples the objects in the real scene. Inevitably, during the scanning and sampling process, it will be affected by external factors such as the environment and the surface shape and texture of the target object. Errors and noises will inevitably appear in the point cloud data, and there are also other irrelevant point clouds and redundant data. Therefore, it is necessary to preprocess the point cloud data. The preprocessing of point cloud data generally includes the following steps: noise removal, multi-view alignment, data reduction, and surface reconstruction.

[0179] (1) Noise removal refers to removing the data outside the scanned object from the point cloud data. During the scanning process, due to the influence of certain environmental factors, such as the surrounding people being collected by the scanner, these data need to be deleted in the post-processing stage.

[0180] (2) Multi-view alignment means that due to the large size or complex shape of the measured part, it is often impossible to measure all the data in one scan. Instead, multiple scans need to be performed from different positions and multiple perspectives. These point clouds need to be aligned and stitched, which is called multi-view alignment. Point cloud alignment and stitching can be achieved by arranging homologous control points on the object surface.

[0181] (3) Data reduction means that since the point cloud data is massive data, it is necessary to reduce the data without affecting the surface reconstruction and maintaining a certain accuracy.

[0182] (4) Surface reconstruction means representing the scanned data with an accurate surface. After surface reconstruction, 3D modeling can be carried out to restore the true appearance of the scanned target.

[0183] Since the data based on binocular stereo vision triangulation is easily affected by factors such as illumination and background interference, the target detection results of LiDAR are used to correct the collected data to improve the accuracy of model data collection. The specific operation is to establish the relationship between the three-dimensional coordinates of the camera and LiDAR, and realize the conversion of the two three-dimensional coordinates in the joint measurement system.

[0184] Suppose there is a point P in space i (i = 1, 2,..., n). If the LiDAR measurement system is used, P li =(X li , Y li , Z li ) represents the three-dimensional coordinates of the spatial point P i in the coordinate system of the LiDAR measurement system; if the binocular stereo vision system is used, P ci =(X ci , Y ci , Z ci ) represents the three-dimensional coordinates of the spatial point Pi Three-dimensional coordinates in a binocular stereo vision system. When performing data fusion in a joint measurement system, it is necessary to determine the corresponding relationship between the two sensor coordinate systems.

[0185] The coordinate transformation process generally can be divided into three steps:

[0186] (1) Converting from the LiDAR coordinate system to the camera coordinate system can be represented by a rotation transformation matrix R and a translation transformation matrix T. Among them, R is a matrix of size 3×3, representing the rotation of spatial coordinates; T is a matrix of size 3×1, representing the translation of spatial coordinates. The conversion process is shown in Equation (6).

[0187]

[0188] (2) Converting from the camera coordinate system to the image coordinate system is a process of converting from a three-dimensional coordinate system to a two-dimensional coordinate system, belonging to a perspective projection relationship and satisfying the similarity theorem of triangles. The conversion process is shown in Equation (7).

[0189]

[0190] (3) Converting from the image coordinate system to the pixel coordinate system. At this time, there is no rotation transformation, but the position of the coordinate origin is different and the unit length is different, mainly involving scaling transformation and translation transformation. The conversion process is shown in Equation (8).

[0191]

[0192] To sum up, the coordinate conversion relationship between LiDAR and the camera can be expressed as:

[0193]

[0194] Intelligent operation includes remote operation, equipment assembly, operation and maintenance, etc. of converter station equipment. Its task objectives are as follows:

[0195] (1) During equipment assembly and operation and maintenance, through rich and good human-machine interaction, the staff can obtain guidance and help at any time through auxiliary equipment to improve operation efficiency and reliability.

[0196] (2) For remote equipment operation, through the original background discrimination means and intelligent image recognition, infrared sensing, pressure sensing, etc., automatically judge whether the remote operation is in place, and realize the automatic programming of switching operations and unmanned operation on site.

[0197] The overall technical route of the intelligent interaction module is as Figure 6 shown.

[0198] By Figure 6As shown in the figure, the overall technical route of the intelligent interaction module is divided into two levels: operation intelligent assistance and operation result recognition.

[0199] Operation intelligent assistance includes two aspects:

[0200] (1) Workers can obtain device, environment, and location information during the operation in real time by wearing intelligent equipment and operation assistance devices, and the operation process, operation in-place situation, etc. are displayed on the intelligent equipment, providing operation assistance information for the workers.

[0201] (2) Workers can conduct real-time voice and video information interaction with remote collaborative workers through the assistance device to carry out remote operation control, operation guidance, and task collaboration. The specific technical solutions are as Figure 7 shown.

[0202] First of all, the AR technology is used to guide workers to complete the operation and maintenance work in a standardized and normalized manner in real time. During the whole implementation process, workers wear intelligent devices. By identifying the nameplate and pressure plate of the device, the device information is superimposed on the actual vision of the workers, improving work efficiency.

[0203] Secondly, during the inspection process of wearing intelligent devices, the content information of specific operation tasks is obtained from the production management system in automatic and manual ways, and the inspection operation instruction manual is synchronized to the intelligent interaction platform. Workers can receive and execute tasks through intelligent devices, and use the AR technology to project virtual instruction labels in the real scene to provide clear work guidance for the workers, effectively improving work quality.

[0204] During the cross-region and multi-person mobile maintenance operation process, real-time voice and video information interaction is realized by wearing intelligent devices.

[0205] On-site workers wear intelligent devices to detect power equipment. When interacting with the back-end workers, the information of the observed equipment and environment can be transmitted to the back-end workers in the monitoring center in the form of video images. The back-end workers can master the on-site work situation through intelligent devices and superimpose virtual information such as text, images, or animations on them to guide the on-site workers to operate.

[0206] The operation of equipment in traditional converter stations includes multiple links such as preparation before operation, filling out tickets, departure, arriving at the site, receiving orders, simulation rehearsal, formal operation, and completion report. In the intelligent converter station, a series of operations of circuit breakers and disconnectors can be automatically completed according to the operation commands of the automation system and the pre-set operation logic, realizing the one-key completion of complex on-site operations in the remote back-end. The whole operation process does not require additional manual intervention, significantly reducing the labor intensity, greatly improving the operation efficiency, and effectively reducing the operation safety risk.

[0207] The operation result identification technology is mainly applied to the remote operation scenario of converter station equipment to identify the operation conditions of primary equipment and secondary equipment, and judge whether each step of operation is in place. In order to improve the accuracy of operation result identification, it is often necessary to perform double confirmation with at least two non-homologous signals. This article takes the switch equipment in power equipment as a typical application scenario, and its operation result identification technology framework is as Figure 8 shown.

[0208] As Figure 8 can be seen, the real-time monitoring module for on-site information mainly consists of the following parts:

[0209] (1) Infrared sensor, used to monitor the contact temperature signal. The infrared temperature measurement probe is installed on the internal circuit board under the infrared glass. By monitoring the contact temperature, it is possible to timely detect whether the disconnector has completed closing;

[0210] (2) Pressure sensor, used to monitor the sudden change signal of contact pressure. A pressure sensor is installed at the contact position of the disconnector to measure the pressure during opening and closing, and the opening and closing states are reflected by the pressure value;

[0211] (3) The image monitoring device collects the image of the disconnector in real time, monitors the actual mechanical position of the disconnector, and accurately identifies the state of the disconnector.

[0212] Through the above methods, it is possible to intuitively, effectively and truly perceive the operation results of remote operations. It provides technical support for the double confirmation mechanism of non-homologous signals for operation results.

[0213] In the above embodiment, when performing operation and maintenance processing on converter station equipment, after obtaining the equipment information, target position information and environmental information of the equipment to be analyzed in the converter station to be analyzed, through a series of processes such as preprocessing, feature extraction processing, and fusion processing, based on a pre-trained equipment operation and maintenance instruction prediction model, the target equipment operation and maintenance instruction corresponding to the equipment to be analyzed can be quickly and directly obtained, which is beneficial to improving the determination efficiency of the equipment operation and maintenance instructions of converter station equipment, and then improving the operation and maintenance efficiency of converter station equipment; moreover, the entire process does not require manual intervention, avoiding the defects that the method of manual inspection is relatively cumbersome and requires a large amount of time and manpower, resulting in low operation and maintenance efficiency of converter station equipment, and further improving the operation and maintenance efficiency of converter station equipment.

[0214] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are sequentially shown according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0215] Based on the same inventive concept, an embodiment of the present application also provides a device for implementing the above-mentioned method for operation and maintenance of converter station equipment based on spatial computing. The implementation solution provided by this device to solve problems is similar to the implementation solution described in the above method. Therefore, the specific limitations in one or more embodiments of the device for operation and maintenance of converter station equipment based on spatial computing provided below can refer to the limitations on the method for operation and maintenance of converter station equipment based on spatial computing in the above text, and will not be repeated here.

[0216] In an exemplary embodiment, as Figure 9 shown, a device for operation and maintenance of converter station equipment based on spatial computing is provided, including: an information acquisition module 901, an information processing module 902, a feature extraction module 903, a feature fusion module 904, an instruction prediction module 905, and a device operation and maintenance module 906, where:

[0217] The information acquisition module 901 is used to acquire device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed.

[0218] The information processing module 902 is used to preprocess the device information, target location information, and environmental information respectively to obtain preprocessed device information, preprocessed location information, and preprocessed environmental information.

[0219] The feature extraction module 903 is used to perform feature extraction processing on the preprocessed device information, preprocessed location information, and preprocessed environmental information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environmental information.

[0220] The feature fusion module 904 is used to perform fusion processing on the first feature vector, the second feature vector, and the third feature vector to obtain a fusion feature vector corresponding to the device to be analyzed.

[0221] The instruction prediction module 905 is configured to input the fused feature vector into a pre-trained device operation and maintenance instruction prediction model to obtain the target device operation and maintenance instruction corresponding to the device to be analyzed.

[0222] The device operation and maintenance module 906 is configured to perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

[0223] In an exemplary embodiment, the information acquisition module 901 is further configured to acquire the initial position information of the device to be analyzed in the converter station to be analyzed; perform conversion processing on the initial position information according to a preset correspondence relationship to obtain the target position information; the preset correspondence relationship is used to represent the correspondence relationship between the initial position information and the target position information.

[0224] In an exemplary embodiment, the feature extraction module 903 is further configured to use the preprocessed device information as the main data, and the preprocessed position information and the preprocessed environment information as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain a first feature vector; use the preprocessed position information as the main data, and the preprocessed device information and the preprocessed environment information as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain a second feature vector; use the preprocessed environment information as the main data, and the preprocessed device information and the preprocessed position information as auxiliary data, and input them into a feature extraction model for feature extraction processing to obtain a third feature vector.

[0225] In an exemplary embodiment, the feature fusion module 904 is further configured to obtain a first initial weight corresponding to the preprocessed device information, a second initial weight corresponding to the preprocessed position information, and a third initial weight corresponding to the preprocessed environment information; perform normalization processing on the first initial weight, the second initial weight, and the third initial weight to obtain a first target weight corresponding to the preprocessed device information, a second target weight corresponding to the preprocessed position information, and a third target weight corresponding to the preprocessed environment information; perform fusion processing on the first feature vector, the second feature vector, and the third feature vector according to the first target weight, the second target weight, and the third target weight to obtain a fused feature vector.

[0226] In an exemplary embodiment, the converter station device operation and maintenance device based on spatial calculation further includes a converter station determination module, configured to acquire the voltage level information and device type information of the candidate converter stations; screen out the candidate converter stations whose voltage level information and device type information both meet the preset conditions from each candidate converter station as the converter station to be analyzed.

[0227] In an exemplary embodiment, the converter station equipment operation and maintenance device based on spatial computing further includes a model training module, which is configured to obtain sample device information, sample location information, and sample environment information of sample devices in a sample converter station; perform preprocessing on the sample device information, sample location information, and sample environment information respectively to obtain preprocessed sample device information, preprocessed sample location information, and preprocessed sample environment information; perform feature extraction processing on the preprocessed sample device information, preprocessed sample location information, and preprocessed sample environment information respectively to obtain a first sample feature vector corresponding to the preprocessed sample device information, a second sample feature vector corresponding to the preprocessed sample location information, and a third sample feature vector corresponding to the preprocessed sample environment information; perform fusion processing on the first sample feature vector, the second sample feature vector, and the third sample feature vector to obtain a sample fusion feature vector corresponding to the sample device; input the sample fusion feature vector into a device operation and maintenance instruction prediction model to be trained to obtain a predicted device operation and maintenance instruction corresponding to the sample device; obtain the actual device operation and maintenance instruction corresponding to the sample device, and perform iterative training on the device operation and maintenance instruction prediction model to be trained according to the difference between the predicted device operation and maintenance instruction and the actual device operation and maintenance instruction to obtain a pre-trained device operation and maintenance instruction prediction model.

[0228] Each module in the above-mentioned converter station equipment operation and maintenance device based on spatial computing can be implemented in whole or in part by software, hardware, and their combination. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so as to facilitate the processor to call and execute the operations corresponding to the above-mentioned modules.

[0229] In an exemplary embodiment, a computer device is provided. The computer device can be a server, and its internal structure diagram can be as Figure 10As shown in the figure. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as device information, target location information, and environment information. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through a network connection. When the computer program is executed by the processor, it implements a method for operation and maintenance of converter station equipment based on spatial computing.

[0230] Those skilled in the art can understand that Figure 10 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0231] In an exemplary embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0232] In an exemplary embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0233] In an exemplary embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0234] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memories can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memories can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0235] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0236] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A converter station equipment operation and maintenance method based on spatial computing, characterized in that: The method comprises: Obtaining device information, target location information, and environmental information of the device to be analyzed in the converter station to be analyzed; Preprocessing the device information, the target location information and the environment information respectively to obtain preprocessed device information, preprocessed location information and preprocessed environment information; Performing feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environment information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environment information; fusing the first feature vector, the second feature vector and the third feature vector to obtain a fused feature vector corresponding to the device to be analyzed; Inputting the fused feature vector into a pre-trained equipment operation and maintenance instruction prediction model to obtain the target equipment operation and maintenance instruction corresponding to the equipment to be analyzed; According to the target device operation and maintenance instruction, corresponding operation and maintenance processing is performed on the device to be analyzed.

2. The method according to claim 1, characterized in that The obtaining of device information, target location information and environment information of the device to be analyzed in the converter station to be analyzed includes: Obtaining initial position information of the equipment to be analyzed in the converter station to be analyzed; According to a preset corresponding relationship, the initial position information is converted to obtain the target position information; the preset corresponding relationship is used to represent the corresponding relationship between the initial position information and the target position information.

3. The method according to claim 1, characterized in that The performing feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environment information respectively to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environment information includes: The preprocessed device information is used as main data, and the preprocessed location information and the preprocessed environment information are used as auxiliary data, and are input into a feature extraction model for feature extraction processing to obtain the first feature vector; The preprocessed position information is used as main data, and the preprocessed device information and the preprocessed environment information are used as auxiliary data, and are input into the feature extraction model for feature extraction processing to obtain the second feature vector; The preprocessed environmental information is used as main data, and the preprocessed device information and the preprocessed location information are used as auxiliary data, which are input into the feature extraction model for feature extraction processing to obtain the third feature vector.

4. The method according to claim 1, characterized in that: The fusing the first feature vector, the second feature vector and the third feature vector to obtain a fused feature vector corresponding to the device to be analyzed includes: Obtaining a first initial weight corresponding to the preprocessed device information, a second initial weight corresponding to the preprocessed location information, and a third initial weight corresponding to the preprocessed environment information; Normalizing the first initial weight, the second initial weight, and the third initial weight to obtain a first target weight corresponding to the preprocessed device information, a second target weight corresponding to the preprocessed location information, and a third target weight corresponding to the preprocessed environment information; The first feature vector, the second feature vector and the third feature vector are fused according to the first target weight, the second target weight and the third target weight to obtain the fused feature vector.

5. The method according to claim 1, characterized in that Before obtaining the device information, target location information and environment information of the device to be analyzed in the converter station to be analyzed, the following steps are also included: Obtain voltage level information and equipment type information of candidate converter stations; From the candidate converter stations, the candidate converter stations whose voltage level information and equipment type information both meet preset conditions are screened out as the converter stations to be analyzed.

6. According to the method according to any one of claims 1 to 5, the pre-trained equipment operation and maintenance instruction prediction model is trained in the following manner: Acquire sample device information, sample location information and sample environment information of a sample device in a sample converter station; Preprocessing the sample device information, the sample location information and the sample environment information respectively to obtain preprocessed sample device information, preprocessed sample location information and preprocessed sample environment information; Performing feature extraction processing on the preprocessed sample device information, the preprocessed sample position information, and the preprocessed sample environment information respectively to obtain a first sample feature vector corresponding to the preprocessed sample device information, a second sample feature vector corresponding to the preprocessed sample position information, and a third sample feature vector corresponding to the preprocessed sample environment information; Performing fusion processing on the first sample feature vector, the second sample feature vector and the third sample feature vector to obtain a sample fusion feature vector corresponding to the sample device; Inputting the sample fusion feature vector into the equipment operation and maintenance instruction prediction model to be trained to obtain the predicted equipment operation and maintenance instruction corresponding to the sample equipment; The actual equipment operation and maintenance instructions corresponding to the sample equipment are obtained, and according to the difference between the predicted equipment operation and maintenance instructions and the actual equipment operation and maintenance instructions, the equipment operation and maintenance instruction prediction model to be trained is iteratively trained to obtain the pre-trained equipment operation and maintenance instruction prediction model.

7. A converter station equipment operation and maintenance device based on spatial calculation, characterized in that: The device comprises: An information acquisition module, used to acquire device information, target location information and environmental information of the device to be analyzed in the converter station to be analyzed; An information processing module, used to preprocess the device information, the target location information and the environment information respectively to obtain preprocessed device information, preprocessed location information and preprocessed environment information; a feature extraction module, configured to perform feature extraction processing on the preprocessed device information, the preprocessed location information, and the preprocessed environment information, respectively, to obtain a first feature vector corresponding to the preprocessed device information, a second feature vector corresponding to the preprocessed location information, and a third feature vector corresponding to the preprocessed environment information; A feature fusion module, used for fusing the first feature vector, the second feature vector and the third feature vector to obtain a fused feature vector corresponding to the device to be analyzed; An instruction prediction module, used for inputting the fused feature vector into a pre-trained equipment operation and maintenance instruction prediction model to obtain the target equipment operation and maintenance instruction corresponding to the equipment to be analyzed; The device operation and maintenance module is used to perform corresponding operation and maintenance processing on the device to be analyzed according to the target device operation and maintenance instruction.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images