A transformer intelligent rapid acceptance method and system

Abstract

The present application discloses a method and system for intelligent rapid acceptance of transformers, which relate to the technical field of intelligent storage of electrical equipment, and include: establishing a theoretical model of the transformer to be accepted according to a transformer equipment abnormality database and equipment information of the transformer to be accepted; acquiring actual parameters of the transformer to be accepted and establishing an actual model; performing a rapid pre-inspection according to characteristic components and / or characteristic parameters of the transformer to be accepted; comparing the theoretical model with the actual model to obtain a first comparison deviation, and obtaining an abnormal diagnosis result of the transformer to be accepted based on the first comparison deviation; performing a multi-dimensional quantitative evaluation on the abnormal diagnosis result in combination with the transformer equipment abnormality database, the uses of various components of the transformer to be accepted and the abnormal thresholds of the corresponding components, obtaining a quantitative analysis report and outputting corresponding early warning instructions; and updating the transformer equipment abnormality database based on the quantitative analysis report to improve the efficiency and accuracy of power transformer acceptance.

Description

Technical Field

[0001] The present application relates to the technical field of intelligent storage of electrical equipment, and in particular to a method and system for intelligent rapid acceptance of transformers. Background Art

[0002] In the field of power systems, transformers are one of the key equipment, and their acceptance work faces many challenges. The transformers are large in size and complex in structure, which makes it difficult to manually obtain the parameters required for acceptance; there are many manufacturers and models, which makes it difficult for warehouse managers to manually identify; there are complex acceptance situations such as the winning manufacturer's insufficient production capacity, purchasing other manufacturers' products for OEM delivery, and the winning manufacturer changing the design and reducing performance.

[0003] At present, the existing electrical equipment acceptance technology is mostly for the acceptance of overall projects or projects such as power transmission and transformation projects and power transmission and transformation lines. As a key core component of the power transmission and distribution grid, the transformer has a complex structure and many characteristic components. The existing acceptance methods for overall projects or projects of power transmission and transformation projects and power transmission and transformation lines cannot be fully applied. Therefore, it is urgent to improve the acceptance methods for power transformers. Summary of the invention

[0004] The present application provides a transformer intelligent rapid acceptance method and system to solve the problem of how to improve the efficiency and accuracy of power transformer acceptance.

[0005] To solve this technical problem, this application provides the following technical solutions:

[0006] A transformer intelligent rapid acceptance method, comprising:

[0007] According to the transformer equipment anomaly database and the equipment information of the transformer to be accepted, a theoretical model of the transformer to be accepted is established, and the theoretical model is stored in the theoretical model database, wherein the theoretical model includes theoretical two-dimensional drawing information, theoretical three-dimensional point cloud model and theoretical appearance parameters; the theoretical model database is used to obtain the theoretical model of the transformer of the corresponding model or specification or the theoretical model of the common component corresponding to the transformer when accepting transformers of the same model or specification or common components of the transformer;

[0008] Performing laser scanning and image recognition processing on the transformer to be accepted, obtaining actual parameters of the transformer to be accepted, and establishing an actual model, wherein the actual model includes actual two-dimensional drawing information, an actual three-dimensional point cloud model, and actual appearance parameters;

[0009] Perform a quick pre-check based on the characteristic components and / or characteristic parameters of the transformer to be accepted in the theoretical model and the actual model to determine whether the supply is correct;

[0010] If the supply is correct, the theoretical two-dimensional drawing information is compared with the actual two-dimensional drawing information, the theoretical three-dimensional point cloud model is compared with the actual three-dimensional point cloud model, and the theoretical appearance parameters are compared with the actual appearance parameters to obtain a first comparison deviation, and an abnormal diagnosis result of the transformer to be accepted is obtained based on the first comparison deviation;

[0011] Combined with the transformer equipment abnormality database, the purpose of each component of the transformer to be accepted and the abnormal threshold of the corresponding component, a multi-dimensional quantitative evaluation is performed on the abnormal diagnosis result to obtain a quantitative analysis report and output a corresponding early warning instruction;

[0012] Based on the quantitative analysis report, the transformer equipment anomaly database is updated, and according to the updated transformer equipment anomaly database, the modeling parameters of the theoretical model and the actual model, and the characteristic components and / or characteristic parameters of the transformer to be accepted are optimized.

[0013] A transformer intelligent rapid acceptance system, comprising:

[0014] Theoretical model module, used for establishing the theoretical model of the transformer to be accepted according to the transformer equipment abnormality database and the equipment information of the transformer to be accepted, and storing the theoretical model in the theoretical model database, wherein the theoretical model includes theoretical two-dimensional drawing information, theoretical three-dimensional point cloud model and theoretical appearance parameters; the theoretical model database is used for obtaining the theoretical model of the transformer of the corresponding model specification or the theoretical model of the common component corresponding to the transformer when accepting the transformer of the same model specification or the common component of the transformer;

[0015] An actual model module is used to perform laser scanning and image recognition processing on the transformer to be accepted, obtain actual parameters of the transformer to be accepted, and establish an actual model, wherein the actual model includes actual two-dimensional drawing information, an actual three-dimensional point cloud model, and actual appearance parameters;

[0016] A rapid pre-check module, used to perform a rapid pre-check based on the characteristic components and / or characteristic parameters of the transformer to be accepted in the theoretical model and the actual model, to determine whether the supply is correct;

[0017] An abnormality diagnosis module is used to compare the theoretical two-dimensional drawing information with the actual two-dimensional drawing information, the theoretical three-dimensional point cloud model with the actual three-dimensional point cloud model, and the theoretical appearance parameters with the actual appearance parameters respectively if the supply is correct, to obtain a first comparison deviation, and obtain an abnormality diagnosis result of the transformer to be accepted based on the first comparison deviation;

[0018] A quantitative analysis module, which is used to combine the transformer equipment abnormality database, the uses of the components of the transformer to be accepted and the abnormal thresholds of the corresponding components, to perform a multi-dimensional quantitative evaluation on the abnormal diagnosis results, obtain a quantitative analysis report and output a corresponding early warning instruction;

[0019] The abnormality database updating module is used to update the transformer equipment abnormality database based on the quantitative analysis report, and optimize the modeling parameters of the theoretical model and the actual model, the characteristic components and / or characteristic parameters of the transformer to be accepted according to the updated transformer equipment abnormality database.

[0020] At least one of the above technical solutions adopted in this application can achieve the following beneficial effects:

[0021] 1. Based on a large amount of historical acceptance data, a transformer equipment anomaly database is established. The modeling parameters of the theoretical model and the actual model are optimized by using the transformer equipment anomaly database updated in real time, and the characteristic parameters and characteristic components of the rapid pre-inspection are optimized, so as to optimize the acceptance system, make the acceptance process more intelligent, and improve the efficiency and accuracy of the acceptance;

[0022] 2. Use the "characteristic parameter method" to quickly pre-check whether the transformer supply is correct, pre-exclude incorrectly supplied transformers, reduce resource consumption for acceptance testing, and further improve transformer acceptance efficiency;

[0023] 3. Present the acceptance results in the form of multi-dimensional evaluation quantitative reports and early warnings, presenting the acceptance results in a more intuitive and eye-catching way, so as to quickly understand the abnormal conditions, degree and impact of transformer equipment;

[0024] 4. Use laser scanning to obtain point cloud data of the transformer's physical dimensions, and digital cameras and image recognition technology to obtain appearance texture information. This can quickly acquire and process large amounts of data, improving acceptance efficiency; it can also acquire parameters that cannot be measured manually, improving acceptance accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0026] Figure 1 A schematic diagram of a flow chart of a transformer intelligent rapid acceptance method provided in Example 1 of the present application;

[0027] Figure 2 A schematic diagram of a detailed process of the intelligent rapid acceptance method for transformers provided in Example 1 of the present application;

[0028] Figure 3A schematic diagram of a laser scanning coordinate system provided in Example 1 of the present application;

[0029] Figure 4 A step diagram of an image recognition algorithm based on a convolutional neural network provided in Example 1 of the present application;

[0030] Figure 5 A schematic diagram of a transformer intelligent rapid acceptance system provided in Example 2 of the present application;

[0031] Figure 6 This is a schematic diagram of another intelligent rapid acceptance system for transformers provided in Example 3 of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0033] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0034] Example 1

[0035] See also Figure 1 , Figure 2 , Figure 1 A schematic diagram of a flow chart of a transformer intelligent rapid acceptance method provided in an embodiment of the present application, Figure 2 A schematic diagram of a detailed process of a transformer intelligent rapid acceptance method provided in an embodiment of the present application. The method may specifically include the following steps:

[0036] Step 101: Establish a theoretical model of the transformer to be accepted based on the transformer equipment anomaly database and the equipment information of the transformer to be accepted, and store the theoretical model in a theoretical model database, wherein the theoretical model includes theoretical two-dimensional drawing information, theoretical three-dimensional point cloud model and theoretical appearance parameters; the theoretical model database is used to obtain the theoretical model of the transformer of the corresponding model or specification or the theoretical model of the common component corresponding to the transformer when accepting transformers of the same model or specification or transformers with common components.

[0037] It should be noted that a specific implementation of step 101 may be:

[0038] Before electrical equipment is put into storage, a virtual model based on theoretical parameters is established according to the equipment information and the equipment information in the abnormal database. Among them, the transformer equipment information comes from the relevant documents of the supplier qualification and capability verification process, the bidding technical documents (such as technical solutions, technical requirements, technical standards, etc.), the bill of materials and the quotation documents. The equipment information specifically includes: the equipment name, model specifications, manufacturer, technical parameters, component composition and structural dimensions of the transformer. The equipment abnormality data includes four types of data: high-frequency abnormal location, high-frequency abnormal component, abnormal type and abnormal degree.

[0039] The theoretical model establishment includes CAD two-dimensional drawings, three-dimensional models converted from CAD two-dimensional drawings, and other information such as color texture, parameters, etc. are classified into tables. In the process of building the three-dimensional model, according to the high-frequency abnormal positions and high-frequency abnormal components recorded in the transformer equipment abnormality database, the point cloud data volume of the corresponding components and position three-dimensional models is increased. This improves the accuracy of acceptance of abnormal positions and abnormal components.

[0040] In order to improve the efficiency of transformer acceptance, in this embodiment, the established theoretical model is stored in a database, and the relevant parameters of the theoretical model stored in the theoretical model database include: supplier, transformer model specifications, component parameters and corresponding component models, etc. When the next transformer acceptance is carried out, the corresponding theoretical model can be directly obtained according to the supplier of the transformer to be accepted, the model specifications of the transformer, etc.; or when there are common components in the transformer, the theoretical model of the common components of the corresponding transformer can be directly obtained from the database. Based on this, it is avoided to repeatedly establish theoretical models of transformers with the same specifications to be accepted, and the acceptance efficiency of the transformer is improved.

[0041] Step 102: perform laser scanning and image recognition processing on the transformer to be accepted, obtain actual parameters of the transformer to be accepted, and establish an actual model, wherein the actual model includes actual two-dimensional drawing information, an actual three-dimensional point cloud model, and actual appearance parameters.

[0042] It should be noted that a specific implementation of step 102 may be:

[0043] When electrical equipment is put into storage, based on the abnormal data stored in the equipment abnormality database and the actual parameters detected by various sensors, a realistic model based on actual parameters is established. Specifically:

[0044] The establishment of the actual model includes using laser scanning equipment to obtain the equipment's external surface point cloud data, and using digital cameras and image recognition technology to obtain the equipment's overall appearance color, texture, and special markings.

[0045] The laser scanning equipment specifically includes: laser sensor, rotating platform, to obtain distance and angle information. Using the propagation and reflection properties of light, the distance is calculated based on the time it takes for the laser to irradiate the surface of the object and then return and the speed of light, and then the three-dimensional coordinates of the measuring point are calculated using the angle. , get point cloud data.

[0046]

[0047] In the formula, is the distance between the measuring point and a laser source, is the speed of light; It is the time interval between sending and receiving the pulse signal.

[0048]

[0049] In the formula, Figure 3 As shown, is the horizontal rotation angle of the rotating platform; is the scanner pitch angle.

[0050] Furthermore, based on the abnormal data in the transformer equipment abnormality database, the amount of point cloud data of the transformer high-frequency abnormal components and abnormal positions is increased.

[0051] Image recognition technology uses image recognition technology based on convolutional neural networks, specifically the region-based fast convolutional neural network (Faster R-CNN) two-stage target detection algorithm. The main steps of the algorithm are:

[0052] (1) The size of the input image can be freely changed. The image is first input into the convolutional layer (CNN) for feature extraction. The region proposal network (RPN) and the region of interest pooling (RoI Pooling) share a feature map;

[0053] (2) RPN receives the global feature map extracted from CNN and introduces the concept of anchor box to map each point on the feature map back to the original image one by one. Each feature map can generate about 300 suggested windows (anchor boxes);

[0054] (3) The proposal box proposed by RPN is mapped on the global feature map and sent to the RoI Pooling layer, so that each proposal window generates a feature map of fixed size;

[0055] (4) The feature map generated at the end is sent to the classification layer and the bounding box regression layer respectively to perform joint training on target classification and bounding box size regression. The main process of training is as follows: Figure 4 shown.

[0056] Based on this, laser scanning is used to obtain point cloud data of the physical dimensions of the transformer, and digital cameras and image recognition technology are used to obtain appearance texture information. This can quickly acquire and process large amounts of data and improve acceptance efficiency. Laser scanning can obtain parameters that cannot be measured manually, thereby improving acceptance accuracy. At the same time, non-manual acceptance methods can overcome human visual fatigue and subjective factors, thereby improving the reliability of acceptance.

[0057] Step 103: Perform a quick pre-check based on the characteristic components and / or characteristic parameters of the transformer to be accepted in the theoretical model and the actual model to determine whether the supply is correct.

[0058] It should be noted that a specific implementation of step 103 may be:

[0059] The correctness of the supply can be quickly determined based on the characteristic components or characteristic parameters of the transformer, which may include: bushing height, distance between high and low voltage bushings, bushing color and texture, intercept from the front heat sink to the back heat sink, heat sink spacing and quantity, etc.

[0060] The specific method is: use image recognition technology to determine whether the transformer contains characteristic components; if the transformer has characteristic components, determine the point cloud data group of the characteristic components, convert them into two-dimensional parameters, compare them with theoretical parameters, and determine the deviation. If the deviation exceeds the preset threshold, the supply is not in accordance with the version, and a red warning instruction is sent to the background management warning module.

[0061] If the transformer does not contain the characteristic components, it means that the supply is not correct and a red warning instruction is sent to the background management warning module.

[0062] Methods for determining a characteristic component point cloud data group include: distinguishing the position of characteristic components according to the point cloud data features of different components, and determining the position of characteristic components by combining image recognition and laser positioning.

[0063] Based on this, the characteristic parameter method is used to quickly pre-check whether the supply is correct, further improving the acceptance efficiency.

[0064] Step 104: If the supply is correct, the theoretical two-dimensional drawing information is compared with the actual two-dimensional drawing information, the theoretical three-dimensional point cloud model is compared with the actual three-dimensional point cloud model, and the theoretical appearance parameters are compared with the actual appearance parameters to obtain a first comparison deviation, and based on the first comparison deviation, an abnormal diagnosis result of the transformer to be accepted is obtained.

[0065] It should be noted that a specific implementation of step 104 may be:

[0066] The point cloud data obtained by the laser array scanning of the equipment is converted into two-dimensional drawing information and compared with the theoretical two-dimensional drawing information, the theoretical three-dimensional point cloud model of the equipment converted from the theoretical two-dimensional drawing information is collided and compared with the actual three-dimensional point cloud model of the equipment, and the actual appearance parameters (including nameplate parameters, color texture, etc.) obtained by image recognition are compared with the theoretical appearance parameters. If the deviation of each 1,000 point comparison is less than 1.5%, it is normal, and if it is greater than 1.5%, it is abnormal.

[0067] Step 105: Combine the transformer equipment abnormality database, the purpose of each component of the transformer to be accepted and the abnormal threshold of the corresponding component, perform a multi-dimensional quantitative evaluation on the abnormal diagnosis result, obtain a quantitative analysis report and output a corresponding early warning instruction.

[0068] It should be noted that a specific implementation of step 105 may be:

[0069] Combined with the abnormal data in the transformer equipment abnormality database, the purpose of each transformer component and the abnormal threshold of the corresponding component, the acceptance results are quantified and divided into fatal abnormalities, progressive abnormalities and non-functional abnormalities, and quantitative evaluation is performed according to the abnormal situation, and an early warning instruction is sent to the background management early warning module. Among them, fatal abnormalities issue red warnings; progressive abnormalities set different abnormal thresholds according to the purpose of different components, and issue yellow warnings if the threshold is not exceeded and red warnings if the threshold is exceeded; non-functional abnormalities only issue yellow warnings.

[0070] Taking the high-voltage bushing as an example, collision or damage is a critical fault. When the abnormal area determined by the point cloud data deviation exceeds 0.2 cm 2 , a red warning is issued when a critical fault occurs; scratches and contamination on the surface are progressive faults. When the image recognition technology determines that the area of ​​the insulator surface exceeds 0.15 cm 2 A yellow warning is issued when the stain area exceeds 25 cm 2 A red warning is issued when the parameter specifications are unclear or incomplete, which is a non-functional fault. When the image recognition result shows that the parameters on the insulator nameplate are unclear, a yellow warning is issued. When the image recognition technology determines that a position on the insulator surface has an area of ​​more than 0.15 cm 2 At the same time, there is a transformer in the abnormal database where the area of ​​the insulator surface at the same location is larger than 0.15 cm 2 If there are stains on the two transformers, the abnormal data of the two transformers are consistent, indicating that the transformer to be accepted and the transformer that has been accepted in the database may be the same transformer, and a red warning is issued. Therefore, for the same abnormal features in the acceptance results, an emphasis on early warning prompts is made to prevent transformers obtained through abnormal channels from appearing on the market.

[0071] Based on this, the acceptance results are presented in the form of multi-dimensional evaluation quantitative reports and early warnings, and the acceptance results are presented in a more intuitive and eye-catching way, which makes it easy to quickly understand the abnormal conditions, extent and impact of the equipment.

[0072] Step 106: Based on the quantitative analysis report, the transformer equipment abnormality database is updated, and according to the updated transformer equipment abnormality database, the modeling parameters of the theoretical model and the actual model, and the characteristic components and / or characteristic parameters of the transformer to be accepted are optimized.

[0073] It should be noted that a specific implementation of step 106 may be:

[0074] Summarize historical quantitative analysis reports and update the transformer equipment abnormality database in real time based on previous transformer damage conditions. Use updated abnormal data to optimize the modeling strategies of theoretical models and actual models, and optimize parametric modeling. For example, summarize high-frequency abnormal components and increase the overall point cloud data volume of the component during theoretical modeling and actual modeling; summarize high-frequency abnormal locations and increase the point cloud data volume of the location during theoretical modeling and actual modeling.

[0075] Secondly, according to the updated equipment abnormality database, frequently abnormal components, abnormal parameters, etc. are summarized, new transformer characteristic components, characteristic parameters, etc. are added, or the original transformer characteristic components, characteristic parameters, etc. are modified. According to the optimized characteristic components and characteristic parameters, the subsequent transformers to be accepted are quickly pre-checked to determine whether the supply is correct. In this embodiment, according to the parameter information such as the outer contour size and appearance texture of the transformer collected historically, the original characteristic parameters can also be optimized.

[0076] In summary, the present invention can achieve the following beneficial effects:

[0077] 1. Based on a large amount of historical acceptance data, a transformer equipment abnormality database is established. The modeling parameters of the theoretical model and the actual model are optimized by using the transformer equipment abnormality database updated in real time, and the characteristic parameters and characteristic components of the rapid pre-inspection are optimized, so as to optimize the acceptance system, make the acceptance process more intelligent, and improve the efficiency and accuracy of the acceptance;

[0078] 2. Use the "characteristic parameter method" to quickly pre-check whether the transformer supply is correct, pre-exclude incorrectly supplied transformers, reduce the resource consumption of acceptance testing, and further improve the efficiency of transformer acceptance;

[0079] 3. Present the acceptance results in the form of multi-dimensional evaluation quantitative reports and early warnings, presenting the acceptance results in a more intuitive and eye-catching way, so as to quickly understand the abnormal conditions, degree and impact of transformer equipment;

[0080] 4. Use laser scanning to obtain point cloud data of the transformer's physical dimensions, and digital cameras and image recognition technology to obtain appearance texture information. This can quickly acquire and process large amounts of data, improving acceptance efficiency; it can also acquire parameters that cannot be measured manually, improving acceptance accuracy.

[0081] Example 2

[0082] Figure 5 A schematic diagram of a transformer intelligent rapid acceptance system provided in Example 2 of the present application. The system includes:

[0083] Theoretical model module 201 is used to establish a theoretical model of the transformer to be accepted according to the transformer equipment abnormality database and the equipment information of the transformer to be accepted, and store the theoretical model in the theoretical model database, wherein the theoretical model includes theoretical two-dimensional drawing information, theoretical three-dimensional point cloud model and theoretical appearance parameters; the theoretical model database is used to obtain the theoretical model of the transformer of the corresponding model or specification or the theoretical model of the common component corresponding to the transformer when accepting transformers of the same model or specification or common components of the transformer;

[0084] The actual model module 202 is used to perform laser scanning and image recognition processing on the transformer to be accepted, obtain actual parameters of the transformer to be accepted, and establish an actual model, wherein the actual model includes actual two-dimensional drawing information, actual three-dimensional point cloud model and actual appearance parameters;

[0085] A fast pre-check module 203 is used to perform a fast pre-check based on the characteristic components and / or characteristic parameters of the transformer to be accepted in the theoretical model and the actual model to determine whether the supply is correct;

[0086] The abnormality diagnosis module 204 is used to compare the theoretical two-dimensional drawing information with the actual two-dimensional drawing information, the theoretical three-dimensional point cloud model with the actual three-dimensional point cloud model, and the theoretical appearance parameters with the actual appearance parameters respectively if the supply is correct, to obtain a first comparison deviation, and obtain an abnormality diagnosis result of the transformer to be accepted based on the first comparison deviation;

[0087] The quantitative analysis module 205 is used to combine the transformer equipment abnormality database, the purpose of each component of the transformer to be accepted and the abnormal threshold of the corresponding component, perform multi-dimensional quantitative evaluation on the abnormal diagnosis result, obtain a quantitative analysis report and output a corresponding early warning instruction;

[0088] The abnormality database updating module 206 is used to update the transformer equipment abnormality database based on the quantitative analysis report, and optimize the modeling parameters of the theoretical model and the actual model, the characteristic components and / or characteristic parameters of the transformer to be accepted according to the updated transformer equipment abnormality database.

[0089] In summary, the present invention can achieve the following beneficial effects:

[0090] 1. Based on a large amount of historical acceptance data, a transformer equipment abnormality database is established. The modeling parameters of the theoretical model and the actual model are optimized by using the transformer equipment abnormality database updated in real time, and the characteristic parameters and characteristic components of the rapid pre-inspection are optimized, so as to optimize the acceptance system, make the acceptance process more intelligent, and improve the efficiency and accuracy of the acceptance;

[0091] 2. Use the "characteristic parameter method" to quickly pre-check whether the transformer supply is correct, pre-exclude incorrectly supplied transformers, reduce the resource consumption of acceptance testing, and further improve the efficiency of transformer acceptance;

[0092] 3. Present the acceptance results in the form of multi-dimensional evaluation quantitative reports and early warnings, presenting the acceptance results in a more intuitive and eye-catching way, so as to quickly understand the abnormal conditions, degree and impact of transformer equipment;

[0093] 4. Use laser scanning to obtain point cloud data of the transformer's physical dimensions, and digital cameras and image recognition technology to obtain appearance texture information. This can quickly acquire and process large amounts of data, improving acceptance efficiency; it can also acquire parameters that cannot be measured manually, improving acceptance accuracy.

[0094] Example 3

[0095] Figure 6 This is a schematic diagram of another transformer intelligent rapid acceptance system provided in Example 3 of the present application. The system includes:

[0096] Equipment information and abnormality database, used to store theoretical parameters and abnormal conditions of transformers;

[0097] Parametric modeling module, used to model transformers using theoretical and measured parameters;

[0098] A fast pre-check module is used to quickly determine whether the supply is correct based on the transformer's characteristic components and / or characteristic parameters;

[0099] The quantitative analysis module is used to diagnose the damage after determining that the supply is correct and to conduct a multi-dimensional quantitative evaluation of the damage results;

[0100] Artificial intelligence analysis module, used to conduct comprehensive analysis of abnormal information and database information to optimize the system;

[0101] The rapid acceptance alarm module is used to manage the information reception and alarm of each module.

[0102] In summary, the present invention can achieve the following beneficial effects:

[0103] 1. Based on a large amount of historical acceptance data, a transformer equipment abnormality database is established. The modeling parameters of the theoretical model and the actual model are optimized by using the transformer equipment abnormality database updated in real time, and the characteristic parameters and characteristic components of the rapid pre-inspection are optimized, so as to optimize the acceptance system, make the acceptance process more intelligent, and improve the efficiency and accuracy of the acceptance;

[0104] 2. Use the "characteristic parameter method" to quickly pre-check whether the transformer supply is correct, pre-exclude incorrectly supplied transformers, reduce the resource consumption of acceptance testing, and further improve the efficiency of transformer acceptance;

[0105] 3. Present the acceptance results in the form of multi-dimensional evaluation quantitative reports and early warnings, presenting the acceptance results in a more intuitive and eye-catching way, so as to quickly understand the abnormal conditions, degree and impact of transformer equipment;

[0106] 4. Use laser scanning to obtain point cloud data of the transformer's physical dimensions, and digital cameras and image recognition technology to obtain appearance texture information. This can quickly acquire and process large amounts of data, improving acceptance efficiency; it can also acquire parameters that cannot be measured manually, improving acceptance accuracy.

[0107] Example 4

[0108] In another feasible embodiment, this embodiment provides a device for intelligent rapid acceptance of transformers, and the device may specifically include:

[0109] Processor; and

[0110] A memory arranged to store computer executable instructions, which, when executed, cause the processor to perform the steps in any of the above method embodiments.

[0111] Example 5

[0112] In another feasible embodiment, this embodiment provides a storage medium for intelligent rapid acceptance of transformers, and the storage medium may specifically include:

[0113] The storage medium stores a processing program for intelligent rapid acceptance of transformers, and when the processing program for intelligent rapid acceptance of transformers is executed by a processor, the steps in any of the above method embodiments are implemented.

[0114] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A transformer intelligent rapid acceptance method, characterized in that: include: According to the transformer equipment anomaly database and the equipment information of the transformer to be accepted, a theoretical model of the transformer to be accepted is established, and the theoretical model is stored in the theoretical model database, wherein the theoretical model includes theoretical two-dimensional drawing information, theoretical three-dimensional point cloud model and theoretical appearance parameters; The theoretical model database is used to obtain the theoretical model of the transformer of the corresponding model or specification or the theoretical model of the common component corresponding to the transformer when accepting transformers of the same model or specification or transformers with common components; Performing laser scanning and image recognition processing on the transformer to be accepted, obtaining actual parameters of the transformer to be accepted, and establishing an actual model, wherein the actual model includes actual two-dimensional drawing information, an actual three-dimensional point cloud model, and actual appearance parameters; Perform a quick pre-check based on the characteristic components and / or characteristic parameters of the transformer to be accepted in the theoretical model and the actual model to determine whether the supply is correct; If the supply is correct, the theoretical two-dimensional drawing information is compared with the actual two-dimensional drawing information, the theoretical three-dimensional point cloud model is compared with the actual three-dimensional point cloud model, and the theoretical appearance parameters are compared with the actual appearance parameters to obtain a first comparison deviation, and an abnormal diagnosis result of the transformer to be accepted is obtained based on the first comparison deviation; Combined with the transformer equipment abnormality database, the purpose of each component of the transformer to be accepted and the abnormal threshold of the corresponding component, a multi-dimensional quantitative evaluation is performed on the abnormal diagnosis result to obtain a quantitative analysis report and output a corresponding early warning instruction; Based on the quantitative analysis report, the transformer equipment anomaly database is updated, and according to the updated transformer equipment anomaly database, the modeling parameters of the theoretical model and the actual model, and the characteristic components and / or characteristic parameters of the transformer to be accepted are optimized.

2. The method according to claim 1, characterized in that The step of performing laser scanning and image recognition processing on the transformer to be accepted, obtaining actual parameters of the transformer to be accepted, and establishing an actual model includes: Calculate the distance between each measuring point of the transformer to be inspected and the laser source according to the time taken for the laser to irradiate each measuring point on the surface of the transformer to be inspected and then return and the speed of light; Calculate the three-dimensional coordinates of each measuring point of the transformer to be inspected based on the distance between each measuring point of the transformer to be inspected and the laser source, the pitch angle of the laser scanner and the horizontal rotation angle of the rotating platform; Based on the three-dimensional coordinates of each measuring point of the transformer to be accepted, an actual three-dimensional point cloud model is established.

3. The method according to claim 1, characterized in that The step of performing laser scanning and image recognition processing on the transformer to be accepted, obtaining actual parameters of the transformer to be accepted, and establishing an actual model includes: The region-based fast convolutional neural network two-stage target detection algorithm is used to obtain the actual appearance parameters of the transformer to be accepted. The steps include: Obtaining the original image of the transformer to be accepted, inputting it into the convolution layer to extract image features, and obtaining a global feature map of the transformer to be accepted; Receiving the global feature map through a region proposal network, and generating a plurality of suggestion windows based on the global feature map; Mapping the multiple suggestion windows and the global feature map into an interest domain, and generating a feature map of a fixed size based on each suggestion window; Based on the fixed-size feature map, target classification and bounding box size regression are jointly trained to obtain actual appearance parameters of the transformer to be accepted.

4. The method according to claim 1, characterized in that: The method of performing a quick pre-inspection based on the characteristic components of the transformer to be accepted in the theoretical model and the actual model to determine whether the supply is correct includes: Determining whether there are characteristic components in the actual model; If there is a characteristic component, the theoretical two-dimensional drawing information corresponding to the characteristic component is compared with the actual two-dimensional drawing information and / or the theoretical three-dimensional point cloud model is compared with the actual three-dimensional point cloud model to obtain a second comparison deviation degree; If the second comparison deviation exceeds a preset second threshold value, the supply is incorrect and a red warning instruction is output.

5. The method according to claim 1, characterized in that The obtaining of the abnormal diagnosis result of the transformer to be accepted based on the first comparison deviation includes: If the first comparison deviation is less than a preset first threshold, it is diagnosed as no abnormality, otherwise it is diagnosed as abnormality.

6. The method according to claim 5, characterized in that If the first comparison deviation is less than a preset first threshold, then the diagnosis is that there is no abnormality; otherwise, the diagnosis is that there is an abnormality, including: For every 1000 points, when the first comparison deviation is less than 1.5%, it is diagnosed as no abnormality, otherwise it is diagnosed as abnormality.

7. The method according to claim 1, characterized in that The method combines the transformer equipment abnormality database, the uses of the components of the transformer to be accepted, and the abnormal thresholds of the corresponding components to perform a multi-dimensional quantitative evaluation on the abnormality diagnosis result, obtain a quantitative analysis report, and output a corresponding early warning instruction, including: Based on the uses of the components of the transformer to be accepted, the abnormal diagnosis results are quantified to form a quantitative analysis report of the components of the transformer to be accepted, wherein the quantitative analysis report includes fatal abnormalities, progressive abnormalities and non-functional abnormalities; If the quantitative analysis report of a certain component of the transformer to be accepted is a fatal abnormality, a red warning instruction is output; If the quantitative analysis report of a certain component of the transformer to be accepted is a progressive abnormality, a yellow warning instruction or a red warning instruction is output according to the abnormal threshold of the component; If the quantitative analysis report of a certain component of the transformer to be accepted is a non-functional abnormality, a yellow warning instruction is output.

8. The method according to claim 7, characterized in that The method combines the transformer equipment abnormality database, the uses of the components of the transformer to be accepted and the abnormal thresholds of the corresponding components, performs a multi-dimensional quantitative evaluation on the abnormality diagnosis result, obtains a quantitative analysis report and outputs a corresponding early warning instruction, and also includes: The abnormality diagnosis result is compared with the abnormality data stored in the transformer equipment abnormality database. If there is data in the abnormality diagnosis result that is consistent with the abnormality data stored in the transformer equipment abnormality database, a red warning is output.

9. A transformer intelligent rapid acceptance system, characterized in that: include: Theoretical model module, used for establishing a theoretical model of the transformer to be accepted according to the transformer equipment abnormality database and the equipment information of the transformer to be accepted, and storing the theoretical model in the theoretical model database, wherein the theoretical model includes theoretical two-dimensional drawing information, theoretical three-dimensional point cloud model and theoretical appearance parameters; The theoretical model database is used to obtain the theoretical model of the transformer of the corresponding model or specification or the theoretical model of the common component corresponding to the transformer when accepting transformers of the same model or specification or transformers with common components; An actual model module is used to perform laser scanning and image recognition processing on the transformer to be accepted, obtain actual parameters of the transformer to be accepted, and establish an actual model, wherein the actual model includes actual two-dimensional drawing information, an actual three-dimensional point cloud model, and actual appearance parameters; A rapid pre-check module, used to perform a rapid pre-check based on the characteristic components and / or characteristic parameters of the transformer to be accepted in the theoretical model and the actual model, to determine whether the supply is correct; An abnormality diagnosis module is used to compare the theoretical two-dimensional drawing information with the actual two-dimensional drawing information, the theoretical three-dimensional point cloud model with the actual three-dimensional point cloud model, and the theoretical appearance parameters with the actual appearance parameters respectively if the supply is correct, to obtain a first comparison deviation, and obtain an abnormality diagnosis result of the transformer to be accepted based on the first comparison deviation; A quantitative analysis module, which is used to combine the transformer equipment abnormality database, the uses of the components of the transformer to be accepted and the abnormal thresholds of the corresponding components, to perform a multi-dimensional quantitative evaluation on the abnormal diagnosis results, obtain a quantitative analysis report and output a corresponding early warning instruction; The abnormality database updating module is used to update the transformer equipment abnormality database based on the quantitative analysis report, and optimize the modeling parameters of the theoretical model and the actual model, the characteristic components and / or characteristic parameters of the transformer to be accepted according to the updated transformer equipment abnormality database.

Images