Work equipment intelligent identification method and system for power transmission unscheduled operation

Abstract

The present application relates to a work equipment intelligent identification method and system for power transmission unscheduled operation, and relates to the field of image processing, by receiving the image of the work equipment to be identified and the world coordinate information of the positioning patch deployed on the equipment; according to the coordinate information, the work equipment is sorted through a first-level identification model to obtain a model set: when the number of the first-level selected work equipment model set is greater than 1, the first-level selected work equipment model set is traversed to obtain a reference image set that meets the world coordinate information of the first-level selected work equipment model set; the reference image set is subjected to similarity clustering analysis to obtain a reference image clustering result; the reference image clustering result is traversed and subjected to image similarity analysis with the image of the equipment to be identified respectively, and the work equipment model corresponding to the reference image with the maximum image similarity is output, thereby solving the technical problem that the work equipment is difficult to accurately identify and cannot meet the efficient scheduling demand of power transmission unscheduled operation.

Description

Technical Field

[0001] The present invention relates to the field of image processing, and in particular to a method and system for intelligently identifying operating equipment for unplanned power transmission operations. Background Art

[0002] In the field of unplanned power transmission operations, rapid and accurate identification of operating equipment is crucial for ensuring operational safety and improving efficiency. However, traditional equipment identification and scheduling relies primarily on manual guidance, a time-consuming and labor-intensive approach susceptible to human error, resulting in low accuracy. With the rapid development of computer vision and machine learning technologies, intelligent identification methods are increasingly being applied across various fields. However, in unplanned power transmission operations, the types of operating equipment involved are diverse, including spanning gantry cranes, tower cranes, cranes, excavators, pickup trucks, and boom trucks. Their movements are multimodal, often operating in complex and ever-changing environments. This poses significant challenges for intelligent identification, making it difficult to accurately identify equipment and failing to meet the requirements for efficient scheduling of unplanned power transmission operations. Furthermore, existing intelligent identification methods mostly utilize a single recognition model, which often struggles to cope with the diverse and complex characteristics of operating equipment. Furthermore, these methods often employ general feature extraction techniques, lacking refined processing tailored to the characteristics of individual equipment and failing to fully consider the importance of different features. Consequently, the accuracy and reliability of the identification results remain to be improved. Summary of the Invention

[0003] The present invention aims to solve the technical problem in the prior art that operating equipment is difficult to accurately identify and cannot meet the efficient scheduling requirements of unplanned power transmission operations, and provides an intelligent identification method and system for operating equipment for unplanned power transmission operations to solve the problem.

[0004] The technical solution of the present invention to solve the above technical problems is as follows:

[0005] In a first aspect, the present invention provides an intelligent identification method for operating equipment for unplanned power transmission operations, comprising: receiving an image of an operating equipment to be identified, and world coordinate information of a positioning patch deployed on the operating equipment to be identified; sorting the image through a first-level identification model of the operating equipment according to the world coordinate information to obtain a first-level selected operating equipment model set, wherein the first-level identification model of the operating equipment includes multiple sub-channels, and any sub-channel machine learning is obtained based on training of multiple groups of data, and any group of the multiple groups of data includes: the world record coordinates of the positioning patch of a preset operating equipment model, and a label indicating whether it belongs to the preset operating equipment model; when the number of the first-level selected operating equipment model set is greater than 1, traversing the first-level selected operating equipment model set to obtain a reference image set of the first-level selected operating equipment model set that meets the world coordinate information; performing similarity clustering analysis on the reference image set to obtain a reference image clustering result; traversing the reference image clustering result, performing image similarity analysis with the image of the equipment to be identified respectively, and extracting the operating equipment model corresponding to the reference image with the maximum image similarity value for output.

[0006] In a second aspect, the present invention provides an intelligent identification system for operating equipment for unplanned power transmission operations, comprising: an information receiving module for receiving an image of an operating equipment to be identified, and world coordinate information of a positioning patch deployed on the operating equipment to be identified; an intelligent sorting module for sorting the operating equipment through a first-level identification model based on the world coordinate information to obtain a first-level selected operating equipment model set, wherein the first-level identification model for the operating equipment includes multiple sub-channels, and any sub-channel machine learning is obtained based on training of multiple groups of data, and any group of the multiple groups of data includes: the world record coordinates of the positioning patch of a preset operating equipment model, and a label indicating whether it belongs to the preset operating equipment model; an information screening module for traversing the first-level selected operating equipment model set when the number of the first-level selected operating equipment model set is greater than 1, and obtaining a reference image set of the first-level selected operating equipment model set that meets the world coordinate information; a clustering analysis module for performing similarity clustering analysis on the reference image set to obtain a reference image clustering result; a result output module for traversing the reference image clustering result, performing image similarity analysis with the image of the equipment to be identified, and extracting the operating equipment model corresponding to the reference image with the maximum image similarity for output.

[0007] The beneficial effects of the present application are: by receiving the image of the to-be-identified work equipment and the world coordinate information of the positioning patch, a primary sorting is performed by using a first identification model containing multiple sub-channels trained based on multiple groups of data, to obtain a possible work equipment model set; when there are multiple possible models, the most matched work equipment model is determined and output through image similarity analysis, so that the work equipment is quickly and accurately identified, and the technical effects of improving the efficiency and safety of power transmission unplanned work are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0008] Figure 1 The flowchart of the work equipment intelligent identification method for power transmission unplanned work provided by the present application is shown.

[0009] Figure 2 The structure diagram of the work equipment intelligent identification system for power transmission unplanned work provided by the present application is shown.

[0010] The reference signs are explained: information receiving module 11, intelligent sorting module 12, information screening module 13, cluster analysis module 14, result output module 15. DETAILED DESCRIPTION

[0011] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0012] In the description of the present application, the terms "first", "second" are only used for description purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features. Therefore, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "multiple" is two or more, unless otherwise specifically limited.

[0013] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0014] Example 1

[0015] like Figure 1 As shown, an embodiment of the present invention provides an intelligent identification method for operating equipment for unplanned power transmission operations, including:

[0016] S10: Receive an image of the working device to be identified and world coordinate information of a positioning patch deployed on the working device to be identified.

[0017] S20: According to the world coordinate information, sorting is performed through the first-level identification model of the operating equipment to obtain a first-level selected operating equipment model set, wherein the first-level identification model of the operating equipment includes multiple sub-channels, and the machine learning of any sub-channel is obtained based on multiple groups of data training, and any group of the multiple groups of data includes: the world record coordinates of the positioning patch of the preset operating equipment model, and a label indicating whether it belongs to the preset operating equipment model.

[0018] S30: When the number of the first-level selected operating equipment model sets is greater than 1, traverse the first-level selected operating equipment model sets to obtain a reference image set of the first-level selected operating equipment model sets that meets the world coordinate information.

[0019] S40: performing similarity cluster analysis on the reference image set to obtain a reference image clustering result.

[0020] S50: Traversing the clustering results of the reference images, performing image similarity analysis with the images of the equipment to be identified, extracting and outputting the operating equipment model corresponding to the reference image with the maximum image similarity.

[0021] For example, unplanned transmission operations generally refer to transmission operations within the power system that are not planned or scheduled in advance. These operations may be carried out temporarily due to emergencies, equipment failures, or other urgent needs. Because they are not planned and scheduled in detail in advance, they may pose certain challenges to the stable operation of the power system. In reality, due to the existence of various unforeseen factors, unplanned transmission operations are difficult to completely avoid. Intelligent identification of operating equipment refers to the use of advanced technologies such as computer vision and machine learning to automatically identify equipment at the operation site. This identification technology can quickly obtain information such as equipment type, model, and status, thereby providing strong support for subsequent operations. During unplanned transmission operations, intelligent identification of operating equipment can help workers quickly and accurately identify the equipment that needs to be operated, thereby improving work efficiency and safety.

[0022] In this embodiment, when an unplanned power transmission operation occurs, an image of the equipment to be identified at the work site must first be captured through some method (such as camera capture). This image serves as the basis for the subsequent identification process and contains key information such as the equipment's appearance, shape, and color. Furthermore, several positioning patches have been pre-deployed on the equipment. These patches provide a fixed reference point, enabling the system to accurately determine the equipment's position in three-dimensional space. Each positioning patch has a corresponding world coordinate information describing its exact position in space. This world coordinate information is typically obtained using a positioning technology, such as GPS, LiDAR, or measurement using pre-defined fixed reference objects. These technologies can accurately determine the coordinate values ​​of each positioning patch in three-dimensional space. In the actual intelligent identification process, the system first receives an image of the equipment to be identified and then obtains the world coordinate information of the positioning patches deployed on these equipment. This information is then fed into a subsequent processing module. The processing module uses the first-level equipment identification model to sort the equipment based on the world coordinate information of the positioning patches, quickly narrowing the list of equipment to be identified and obtaining a set of potentially matching equipment models.

[0023] Furthermore, once an image of the equipment to be identified at the work site is captured and the world coordinate information of the positioning patches deployed on the equipment is obtained, this information is input into the first-level equipment identification model. This model is a machine learning model designed and trained specifically for quickly and accurately identifying the model of the equipment being identified. The first-level equipment identification model consists of multiple sub-channels, which are not independent but work together to improve recognition accuracy and efficiency. Each sub-channel is trained based on a large amount of data. This data contains the world coordinates of the positioning patches of a predefined equipment model and a label indicating whether these coordinates belong to that specific equipment model. Specifically, each data set in the training dataset includes the world coordinate information of a positioning patch and a corresponding label. The label is a simple indicator that indicates whether the coordinates match a specific equipment model. For example, if the coordinates in a data set match the coordinates of a positioning patch on a specific transformer model, the label indicates that the coordinates belong to that transformer model. During training, each sub-channel learns how to extract useful features from the input world coordinate information and predict the equipment model based on these features. Because each sub-channel is trained on a different dataset, it can capture subtle differences between device models, thereby improving recognition accuracy. When actually used, the input world coordinate information is fed into the first-level identification model for operating equipment. The multiple sub-channels in the model process this information in parallel and predict possible device models based on their respective learned features. Ultimately, the outputs of these sub-channels are integrated to form a first-level set of selected operating equipment models. This set includes all device models that the model believes may match the input world coordinate information. Through this process, the first-level identification model for operating equipment can use world coordinate information to quickly narrow the range of devices to be identified, providing strong support for subsequent, more accurate identification.

[0024] Next, after the first-level equipment recognition model preliminarily screens a set of possible equipment models (i.e., the first-level selected equipment model set) based on the world coordinate information, if the number of equipment models in this set is greater than one, it indicates that there are multiple possible matches. To further determine the most accurate equipment model, the system then performs a more refined matching process. The core of this process involves traversing the first-level selected equipment model set and searching for a corresponding reference image set that satisfies the world coordinate information for each equipment model in the set. The reference image set consists of a series of images of known equipment models, captured under different conditions, each containing the precise location information of the positioning patch on that equipment model. Specifically, the system extracts the first equipment model from the first-level selected equipment model set and searches for it in its corresponding reference image set. The search is based on whether the positioning patch position information in the reference image matches the input world coordinate information. If a match is found, the equipment model represented by the reference image is consistent with the spatial layout of the on-site equipment, and the reference image is added to the set of qualified reference images. This process is repeated until all equipment models in the first-level selected equipment model set have been traversed. Ultimately, the system generates one or more reference image sets that meet the world coordinate information, each corresponding to a possible device model. This step not only further narrows the range of possible device models but also provides a more precise and targeted reference image set for subsequent image similarity analysis, which helps improve recognition accuracy and efficiency, especially in scenarios requiring rapid response, such as unplanned power transmission operations.

[0025] Next, a similarity cluster analysis is performed on all reference images that meet the world coordinate information. This step aims to group visually similar or similar reference images together to form clustering results. Cluster analysis is based on image features such as color, texture, and shape, calculating the similarity between images to create clusters. The result of similarity cluster analysis is one or more clusters, each containing visually similar reference images. These clusters provide a more accurate reference set for subsequent image similarity analysis, helping to reduce computational effort and improve recognition accuracy. Next, the system iterates through these clusters, performing image similarity analysis on the reference images in each cluster with the device image to be identified. Image similarity analysis is a technique that quantitatively compares the degree of similarity between two images. It can be achieved by calculating the distance between image features or a similarity score. During the traversal process, the system retrieves each reference image from each cluster and calculates the similarity score between them and the device image to be identified. This score reflects the degree of visual proximity between the reference image and the device image to be identified. A higher score indicates greater similarity between the two images. After the system traverses the entire clustering result, it will obtain a series of similarity scores. Then, the maximum value among these scores is found. The reference image corresponding to this maximum value is the image that is most similar to the image of the device to be identified. The operating equipment model represented by the reference image corresponding to this maximum similarity value is then extracted and output as the final recognition result. This output result is the operating equipment model determined by the system through a series of intelligent recognition processes based on the input image of the device to be identified and world coordinate information. Through this process, image similarity analysis technology can be used to accurately identify the specific model of on-site operating equipment from multiple possible equipment models, providing key information for the subsequent processing of unplanned power transmission operations, ensuring the efficient scheduling requirements of unplanned power transmission operations, and achieving the technical effect of improving the efficiency and safety of unplanned power transmission operations.

[0026] In a preferred embodiment, the world coordinate information of the positioning patch to be deployed on the to-be-identified job equipment comprises: performing equipment element decomposition on a preset equipment model to obtain an equipment element set; collecting a plurality of equipment job motion monitoring logs of the preset equipment model in a network according to the equipment element set, wherein any one of the plurality of equipment job motion monitoring logs comprises an equipment element motion position time sequence information set; performing motion consistency clustering on the equipment element set in the plurality of equipment job motion monitoring logs respectively according to the equipment element motion position time sequence information set to obtain a plurality of equipment element clustering results; and performing frequency grouping according to the plurality of equipment element clustering results to obtain a plurality of equipment element groups, wherein any one of the plurality of equipment element groups has a frequency of being classified into the same category in the plurality of equipment element clustering results exceeding a selected frequency threshold, and the selected frequency threshold is greater than or equal to 0.9 of the number of equipment job motion monitoring logs; and selecting random elements in the plurality of equipment element groups respectively to obtain a plurality of equipment elements as positioning patch deployment elements of the preset equipment model.

[0027] Optionally, for a preset device model, the device element disassembly work is performed, that is, a complex device is disassembled into a plurality of independent device elements to form a device element set. These elements are the basic constituent units of the device and also the basis for subsequent analysis. Next, network technology is used to collect a plurality of motion monitoring logs of these preset device models in the operation process, and each log records the motion position time sequence information of the device element in a specific time period, that is, the position change of the element at different time points. These log data provide rich materials for subsequent clustering analysis. After obtaining these log data, motion consistency clustering analysis is performed, and each element in the device element set is clustered by traversing each device operation motion monitoring log. The basis for clustering is whether the motion position time sequence information of the element in different logs shows consistency, that is, whether the element shows a similar motion pattern under similar operation conditions. Through such clustering analysis, elements with similar motion characteristics can be classified into the same category. After clustering, the clustering results are further grouped for frequency, and the purpose is to find element combinations that show consistency in all or most logs. Specifically, if a group of device elements is divided into the same category in multiple clustering results with a frequency exceeding a selected frequency threshold (the threshold is usually set to 0.9 or higher of the number of device operation motion monitoring logs), it is considered that the element group has high stability and representativeness. Finally, a plurality of elements are randomly selected from the frequently appearing element combinations as the deployment elements of the positioning patch, and the position information of these elements will be used as the basis for the deployment of the positioning patch to obtain the key world coordinate information. Since these elements are strictly selected, their position information not only has representativeness but also is relatively small in quantity, thereby reducing the number of positioning patches and the consumption of subsequent computing resources. In summary, through this series of steps, the deployment position of the positioning patch can be efficiently and accurately determined, the number of patches is reduced, the efficiency is improved, and strong support is provided for the subsequent intelligent recognition process.

[0028] In a preferred embodiment, the motion consistency clustering of the device element set is performed respectively based on the set of device element motion position time sequence information and the plurality of device operation motion monitoring logs to obtain a plurality of device element clustering results, including: extracting a set of device element motion position time sequence information of a first device operation motion monitoring log based on the plurality of device operation motion monitoring logs; performing motion similarity calculation on the set of device element motion position time sequence information to obtain a plurality of element motion trajectory similarities; based on a motion trajectory similarity threshold, combining the plurality of element motion trajectory similarities to perform clustering analysis on the device element set to obtain a first device element clustering result, and adding the first device element clustering result to the plurality of device element clustering results.

[0029] Furthermore, key information is extracted from multiple equipment motion monitoring logs. This information constitutes a set of equipment component motion position time series information, which details the position changes and their time series during the operation process. First, focus on the first equipment motion monitoring log and extract the corresponding equipment component motion position time series information set from this log. This set contains the position data of all equipment components within a specific operation cycle, as well as their temporal changes. Next, motion similarity is calculated for this time series information set. This step aims to quantify the degree of similarity between the motion trajectories of different equipment components. Appropriate algorithms or methods, such as dynamic time warping (DTW) and Euclidean distance, can be used to calculate the motion trajectory similarity between each pair of components. Since the time series information of the motion trajectories may not be aligned, dynamic time warping is preferred to improve the comparison accuracy. The main output of DTW is a distance value, which represents the minimum cumulative distance between the two time series. The smaller the distance value, the more similar the two time series are after time warping. Specifically, the results can be more intuitively presented by converting the distance to similarity. The calculation formula is: ;in, Is a positive scaling parameter. In this way, the distance is converted to similarity, and the closer the value is to 1, the more similar it is. Set The reason is that by adjusting You can control the sensitivity of the similarity score to changes in the DTW distance. Makes the similarity score more sensitive to distance changes, while smaller This makes the similarity score less sensitive to distance changes, making it easier to perform efficient calculations. For example, assuming that the DTW distance between two time series X and Y is D, their similarity can be calculated as ,if =0.1, distance D=10, then after calculation = ≈0.3679. Through these calculations, we obtain a set of similarities for several component motion trajectories, where each similarity value represents the degree of proximity between a pair of components in their motion trajectories. With these similarity values, we can perform cluster analysis. A trajectory similarity threshold is set to determine which components should be grouped together. Specifically, if the trajectory similarity between two components exceeds this threshold, they are considered to have consistent motion and should be grouped together. Based on this principle, we cluster the set of equipment components using the trajectory similarities of several components, obtaining the equipment component clustering result corresponding to the first equipment operation motion monitoring log. This clustering result includes several categories, each containing a group of equipment components with consistent motion. This result is added to the multiple equipment component clustering results as a preliminary analysis result. The above process is then repeated for the remaining equipment operation motion monitoring logs. Each log contributes a clustering result, which is then added to the multiple equipment component clustering results. The resulting set of clustering results for all logs provides comprehensive information about the motion consistency of device components. This process not only effectively identifies which device components exhibit consistent motion but also provides strong support for subsequent positioning patch deployment and intelligent identification processes.

[0030] In a preferred embodiment, any one of the sub-channels is machine-learned based on multiple sets of data training, wherein any one of the multiple sets of data includes: the world record coordinates of the positioning patch of the preset operation equipment model and a label indicating whether it belongs to the preset operation equipment model, including: constructing a device action recognition loss function: , ,in, Characterizes the loss value of device action recognition, A label indicating whether the device belongs to the preset operation equipment model. 1 indicates yes, and 0 indicates no. Characterize the judgment results of the training output, is the dynamic weight, Characterize the probability of predicting the type of equipment that belongs to the preset operation, Represents the average value of the predicted probability of all samples, A hyperparameter that characterizes the speed at which the weights are controlled change; configuring the same number of input nodes according to the number of world record coordinates of the positioning patches of the preset operating equipment model, constructing a convolutional layer, a pooling layer, and a fully connected layer connected in sequence, and obtaining a sub-channel topology, wherein the fully connected layer includes an output layer, and the output layer includes two output nodes, 1 and 0; using the label indicating whether the identifier belongs to the preset operating equipment model as supervision and the world record coordinates of the positioning patches of the preset operating equipment model as input, the sub-channel topology is trained to generate the sub-channel.

[0031] For example, in machine learning, in order to train a subchannel that can identify the preset operating equipment model, a training method based on multiple sets of data is used. Each set of data contains two key pieces of information: one is the recorded coordinates of the positioning patch of the preset operating equipment model in the world coordinate system, which serve as input features; the other is a label that identifies whether these coordinates belong to the preset operating equipment model. Furthermore, in order to evaluate the accuracy of the model in identifying equipment actions, a device action recognition loss function is constructed: , In this function Characterizes the loss value of device action recognition, A label indicating whether the device belongs to the preset operation equipment model. 1 indicates yes, and 0 indicates no. Characterize the judgment results of the training output, is the dynamic weight, Characterize the probability of predicting the type of equipment that belongs to the preset operation, Represents the average value of the predicted probability of all samples, A hyperparameter that controls the speed of weight change. The key elements of the function are the label value (true value), the training output judgment result (predicted value), the dynamic weight, and the hyperparameter k that controls the speed of weight change. The dynamic weight is calculated based on the probability of belonging to the preset operating device model and the average of the predicted probabilities of all samples. The speed of its change is affected by the value of k. A larger k value causes more drastic weight changes, causing the model to focus more on difficult-to-classify samples; a smaller k value results in a more gradual weight change. For practical applications, a default k value, such as k=2, is often set to balance model sensitivity and stability. Subsequently, the same number of input nodes is configured based on the number of world record coordinates for the positioning patches of the preset operating device model. A neural network structure is then constructed as a subchannel topology, sequentially connecting convolutional layers, pooling layers, and fully connected layers. The number of nodes in the input layer matches the feature dimensions of the equipment motion image—that is, the number of coordinates in the localization patch. The convolutional layer consists of multiple layers of convolution operations, each with a varying number of kernels, typically 3x3 in size, with a stride of 1 and padding of 1, to extract features from the image. For example, the first layer has 64 kernels, the second layer has 128, and so on, up to the fourth layer with 512 kernels. Each convolutional layer is followed by a 2x2 max pooling layer with a stride of 2 to reduce the dimensionality of the feature map, minimizing computational effort while retaining important features. Finally, the fully connected layer consists of two layers, each with 4096 neurons, to integrate the features extracted by the convolutional and pooling layers and perform classification. The final layer is the output layer, consisting of two neurons, one for each of the two classifications: "yes" and "no." The neural network training process uses the ReLU activation function to enhance the model's nonlinear representation capabilities, and dropout layers are added between fully connected layers to reduce the risk of overfitting. During training, the label indicating whether the model belongs to the preset operating equipment is used as supervision information, and the world record coordinates of the positioning patch of the preset operating equipment model are used as input data. The network parameters are continuously adjusted through the back-propagation algorithm, and finally a sub-channel that can accurately identify the preset operating equipment model is generated.

[0032] In a preferred embodiment, the traversing the reference image set, respectively, with the image similarity analysis of the to-be-identified device image, extracting the work device model corresponding to the reference image of the maximum image similarity value and outputting, comprising: obtaining a first reference image of the reference image set; according to the work device model corresponding to the first reference image, activating the associated feature extraction channel to extract the feature, obtaining the first reference image feature extraction result; according to the feature extraction channel, extracting the feature of the to-be-identified device image, obtaining the to-be-identified device image feature extraction result; calculating the first reference image similarity of the first reference image feature extraction result and the to-be-identified device image feature extraction result, adding to the image similarity set; extracting the work device model corresponding to the maximum value of the reference image of the image similarity set and outputting.

[0033] Optionally, a first reference image is obtained from the reference image set, which represents a certain specific work device model. Then, according to the work device model corresponding to the reference image, a feature extraction channel associated with it is activated, which has been pre-trained and optimized and can efficiently extract key feature information from the image. The first reference image is feature-extracted using this channel to obtain the first reference image feature extraction result. Then, the same feature extraction channel is used to feature-extract the to-be-identified device image to obtain the to-be-identified device image feature extraction result. The two feature extraction results contain key information or feature vectors in the image, which are mathematical descriptions of the image in a certain specific representation space. They can capture key visual elements in the image, such as edges, textures, shapes, colors, etc., as well as the spatial relationships and hierarchical structures between these elements, which can be used for subsequent similarity calculation. Next, the similarity between the first reference image feature extraction result and the to-be-identified device image feature extraction result is calculated. This similarity value reflects the distance or similarity between the two images in the feature space. This similarity value is added to the image similarity set for subsequent comparison and analysis. Repeat the above process to traverse all reference images in the reference image set, respectively, with the to-be-identified device image for feature extraction and similarity calculation. Each calculated similarity value is added to the image similarity set. Finally, the maximum value is extracted from the image similarity set, which corresponds to the reference image most similar to the to-be-identified device image. According to the work device model corresponding to the reference image, it is output as the recognition result of the to-be-identified device image. The whole process realizes the accurate recognition of the to-be-identified device image through feature extraction and similarity calculation, and provides strong support for subsequent automation processing and decision-making.

[0034] In a preferred embodiment, the associated feature extraction channel is activated according to the operating equipment model corresponding to the first reference image to perform feature extraction to obtain the first reference image feature extraction result, which includes: training the associated feature extraction channel according to the preset operating equipment model; wherein, when the preset operating equipment model belongs to a spanning frame, the feature extraction dimensions include frame structure features, height features, and connection node features; when the preset operating equipment model belongs to a tower material, the feature extraction dimensions include cross-sectional shape features, surface texture features, length features, and curvature features; when the preset operating equipment model belongs to a crane, the feature extraction dimensions include boom features, body features, and counterweight features; when the preset operating equipment model belongs to an excavator, the feature extraction dimensions include bucket features, arm features, and track features; when the preset operating equipment model belongs to a pickup truck, the feature extraction dimensions include vehicle front features, cargo box features, and tire features; when the preset operating equipment model belongs to a boom truck, the feature extraction dimensions include boom features, operating bucket features, and chassis features.

[0035] Furthermore, to ensure accuracy, different feature extraction dimensions are defined for different types of work equipment to ensure the model captures the most discriminative information. For example, when the pre-set work equipment model is a spanning frame, feature extraction dimensions include frame structure features (such as the frame outline, the number and spacing of vertical and horizontal bars), height features (the overall height of the spanning frame), and connection node features (information such as the shape and location of the connection nodes). For tower materials, feature extraction dimensions focus on cross-sectional shape features (the cross-sectional shape of the tower material), surface texture features (the texture pattern on the tower material's surface), length features, and curvature features (the length of the tower material and whether it is curved). In the case of cranes, feature extraction dimensions include boom features (the length, shape, and tilt angle of the boom), body features (the shape, size, and color of the body), and counterweight features (the shape, location, and number of counterweights). For excavators, feature extraction dimensions focus on bucket features (the shape, size, and capacity of the bucket), boom features (the length, shape, and range of motion of the boom), and track features (the shape, size, and width of the tracks). When the pre-defined working equipment model is a pickup truck, the feature extraction dimensions include front features (front shape, size, and brand logo), cargo box features (cargo box shape, size, and loading capacity), and tire features (tire size, number, and tread pattern). For a boom truck, the feature extraction dimensions include arm features (arm length, shape, and reach), bucket features (bucket size, shape, and load capacity), and chassis features (chassis shape, size, and suspension system). During training, a large amount of annotated image data containing diverse instances of the various working equipment models mentioned above is used. Through supervised learning, the model learns how to map input images onto these pre-defined feature dimensions, accurately extracting key information related to the equipment model. Once trained, these associated feature extraction channels can be used to process new images of equipment to be identified. Based on the working equipment model associated with the first reference image, the corresponding feature extraction channel is activated, and a feature vector for the first reference image is extracted. This feature vector is then compared with the feature vector of the image of the equipment to be identified to calculate their similarity. This entire process not only improves the accuracy and efficiency of equipment identification but also provides strong support for subsequent automated processing and decision-making. By defining specific feature extraction dimensions for different types of operating equipment, we can ensure that the model can still maintain excellent performance when processing complex and changing field images.

[0036] In a preferred embodiment, the calculation of the first reference image similarity between the feature extraction result of the first reference image and the feature extraction result of the image of the device to be identified includes: when the operating equipment type belongs to a spanning frame, the first reference image similarity is a weighted mean calculation result of the frame structure similarity, height similarity, and connection node similarity, and the weighted weight is assigned by the Delphi method; when the operating equipment type belongs to a tower material, the first reference image similarity is a weighted mean calculation result of the cross-sectional shape similarity, surface texture similarity, length and curvature similarity, and the weighted weight is assigned by the Delphi method; when the operating equipment type belongs to a crane, the first reference image similarity is a weighted mean calculation result of the boom similarity, The weighted mean calculation result of the vehicle body similarity and the counterweight block similarity is assigned by the Delphi method; when the operating equipment type is an excavator, the first benchmark image similarity is the weighted mean calculation result of the bucket similarity, the arm similarity, and the track similarity, and the weighted weight is assigned by the Delphi method; when the operating equipment type is a pickup truck, the first benchmark image similarity is the weighted mean calculation result of the front similarity, the cargo box similarity, and the tire similarity, and the weighted weight is assigned by the Delphi method; when the operating equipment type is a boom truck, the first benchmark image similarity is the weighted mean calculation result of the boom similarity, the working bucket similarity, and the chassis similarity, and the weighted weight is assigned by the Delphi method.

[0037] Exemplarily, in order to calculate the similarity between the feature extraction results of the first reference image and the feature extraction results of the image of the device to be identified, different weighted mean calculation methods are used according to the type of operating equipment. A series of feature similarity indicators are defined for each type of operating equipment, which reflect the degree of proximity between the two images in specific feature dimensions. When the operating equipment type is a spanning frame, the frame structure similarity (measuring the similarity of the frame outline, the number and spacing of vertical poles and horizontal poles), the height similarity (measuring the similarity of the spanning frame height) and the connection node similarity (measuring the similarity of the shape and position of the connection node) are calculated. These similarity indicators are weighted by the weighted weights determined by the Delphi method, and then the weighted mean is calculated as the first reference image similarity. The specific calculation formula for the frame structure similarity is: ;in, and are the lengths of the i-th vertical pole or horizontal pole in the real-time image and the reference image, respectively; n is the number of vertical poles or horizontal poles, is the average length. The specific calculation formula for high similarity is: ;in, and are the heights of the supports in the real-time image and the reference image, Is the preset maximum height. The specific calculation formula for the connection node similarity is: ;in, is the number of connected nodes that match between the live image and the reference image, is the total number of connected nodes in the reference image. For tower materials, the cross-sectional shape similarity, surface texture similarity, length and curvature similarity are calculated, and the weighted weights determined by the Delphi method are also used for weighting, and the weighted mean is obtained as the similarity result. The specific calculation formula for cross-sectional shape similarity is: ,in It is the contour distance between the cross-section of the tower material in the real-time image and the reference image, measured by Frechet distance. Frechet distance is often used to measure the similarity between two curves or shapes. The smaller the value, the more similar the cross-section shapes are. The closer it is to 1. The specific calculation formula for surface texture similarity is: ,here and The gray-level co-occurrence matrix is ​​the surface gray-level co-occurrence matrix of the tower material in the real-time image and the reference image, respectively. The gray-level co-occurrence matrix is ​​a statistical method used to describe texture features in an image. This formula measures the similarity of surface textures by calculating the ratio of the dot product of two matrices to the product of their norms. The closer the value is to 1, the more similar the textures are. The specific calculation formulas for length and curvature similarity are: ,in and is the length of the tower material, and is the curvature. This formula takes into account the difference between the tower material length and curvature and the benchmark value to calculate the similarity. The smaller the difference, the The closer to 1. In the case of cranes, consider the boom similarity (including similarity in length, shape, and tilt angle), the body similarity (similarity in shape, size, and color), and the counterweight similarity (similarity in shape, position, and number), and use the corresponding weights to calculate the weighted mean. The specific calculation formula for boom similarity is: ,in, and are the actual angle and reference angle of the boom respectively; and are the actual length of the boom and the reference length respectively. This formula takes into account the cosine value of the boom angle difference and the relative value of the length difference to calculate the boom similarity. The specific calculation formula for the vehicle body similarity is: ,in, and are the width and height of the vehicle body, respectively. The subscript "real" indicates the actual value, and "bench" indicates the benchmark value. This formula measures the similarity of vehicle bodies by calculating the sum of the relevant ratios of the width and height of the vehicle body respectively. The specific calculation formula for the counterweight similarity is: ,in, and They are the actual distance between the counterweight and the vehicle body and the reference distance, is the average distance. This formula calculates the counterweight similarity based on the ratio of the difference in distance between the counterweight and the vehicle body to the average distance. For excavators, bucket similarity, arm similarity, and track similarity are calculated using the same weighted mean method, with the weights determined by the Delphi method. The specific calculation formula for bucket similarity is: ,in, and The actual bucket area and the reference area are respectively. This formula measures the bucket similarity by calculating the ratio of the product of the actual bucket area and the reference bucket area and the sum of the squares of the two. The closer the value is to 0.5, the higher the similarity. The specific calculation formula for boom similarity is: ,in is the boom length, is the boom angle. This formula takes into account the relative difference between the actual value and the reference value of the boom length and angle. The smaller the difference, the The closer it is to 1, the higher the boom similarity. The specific calculation formula for track similarity is: ,in and are the actual width and reference width of the track, Is the average width. This formula calculates the track similarity based on the ratio of the difference between the actual track width and the reference width to the average width. The smaller the difference, the The closer it is to 1, the higher the track similarity. When the operating equipment is a pickup truck, the similarity of the head, cargo box, and tires is evaluated, and the similarity is calculated using the weighted mean and the weights determined by the Delphi method. The specific calculation formula for the head similarity is: ,The similarity is measured by calculating the ratio of the difference in the coordinates of the points on the front contour to the area of ​​the front. The smaller the difference in coordinates, The closer it is to 1, the higher the similarity of the vehicle head. The specific calculation formula for the similarity of the cargo box is: , calculated based on the relationship between the actual volume of the cargo box and the reference volume. The closer the value is to 0.5, the higher the cargo box similarity. The specific calculation formula for tire similarity is: , calculated based on the ratio of the difference between the actual tire radius and the reference radius relative to the average radius. The smaller the radius difference, The closer it is to 1, the higher the tire similarity. Finally, for the boom truck, the boom similarity, working bucket similarity, and chassis similarity are calculated, and the weighted mean and Delphi method weights are also used to determine the similarity. The specific calculation formula for the boom similarity is: , combined with the cosine value product of the bucket arm angle and the relative value of the length difference to calculate, the closer the angle and length are to the reference value, The closer it is to 1, the higher the bucket arm similarity. The specific calculation formula for the bucket similarity is: , calculated based on the relationship between the actual capacity of the working bucket and the benchmark capacity, the closer the value is to 0.5, the higher the similarity of the working bucket. The specific calculation formula for chassis similarity is: , measured by calculating the ratio of the difference between the width of each part of the chassis and the reference width relative to the average width. The smaller the width difference, The closer it is to 1, the higher the chassis similarity. In short, the Delphi method plays a key role in this process. It can reasonably assign the weight of each feature similarity indicator based on expert opinion. This method takes into account the importance of different features in identifying specific operating equipment models, thereby improving the accuracy and reliability of similarity calculations. By calculating the weighted mean, a comprehensive similarity score can be obtained. This score reflects the degree of proximity between the image of the device to be identified and the first reference image in the feature space. This score can then be used to determine the model of the device to be identified, and the device model with the highest similarity to the first reference image is selected as the identification result. In general, this method combines feature extraction and weighted mean similarity calculation to provide an efficient and accurate method for the automatic identification of operating equipment models.

[0038] In a preferred embodiment, the benchmark image clustering results are traversed, and image similarity analysis is performed with the image of the device to be identified respectively, and the operating equipment model corresponding to the benchmark image with the maximum image similarity is extracted and output, including: extracting the first centroid benchmark image of the first category benchmark image set of the benchmark image clustering results, until the Lth centroid benchmark image of the Lth category benchmark image set is extracted; traversing the first centroid benchmark image to the Lth centroid benchmark image, performing image similarity analysis with the image of the device to be identified respectively, and extracting the selected category benchmark image set of the category to which the centroid benchmark image with the maximum image similarity belongs; traversing the selected category benchmark image set, performing image similarity analysis with the image of the device to be identified respectively, and extracting the operating equipment model corresponding to the benchmark image with the maximum image similarity for output.

[0039] Specifically, clustered benchmark images are obtained. These benchmark images represent different types of operating equipment and are divided into L categories. Each category contains a series of benchmark images similar to the equipment models in that category. To simplify computation and improve efficiency, a representative image, the centroid benchmark image, is extracted from the benchmark images in each category. The centroid benchmark image can be considered the average or center of all benchmark images in that category and, to a certain extent, represents the characteristics of the entire category. Next, these centroid benchmark images are iterated over, starting with the first centroid benchmark image in the first category of benchmark image set and continuing to the Lth centroid benchmark image in the Lth category of benchmark image set. Image similarity analysis is performed on each centroid benchmark image with the image of the equipment to be identified. This process is achieved by comparing features (such as color, texture, and shape) between the two images to identify the centroid benchmark image that is most similar to the image of the equipment to be identified. Once the centroid benchmark image with the highest similarity to the image of the equipment to be identified is found, its category is determined, and this is the selected category benchmark image set. This set contains all benchmark images similar to (i.e., belonging to the same category as) the centroid benchmark image. Next, the search proceeds again, this time within the set of reference images for the selected class. Each reference image in the selected class is subjected to image similarity analysis with the image of the device to be identified to determine the one with the highest similarity. This step further refines the matching results, ensuring that the final output of the operating device model closely matches the image of the device to be identified. Finally, the operating device model corresponding to the reference image with the highest similarity to the device to be identified is extracted and output as the identification result. This model is the most likely model of the device to be identified. This entire process achieves accurate identification of the device model by gradually narrowing the search scope and improving matching accuracy.

[0040] In a preferred embodiment, extracting the first centroid reference image of the first category of reference image set of the reference image clustering result includes: extracting the first reference image of the first category of reference image set, and a first group of image similarity sets between the images within the category and the first reference image, performing mean calculation to obtain a first similarity mean; until the Qth similarity mean is calculated; extracting the reference image with the maximum value from the first similarity mean to the Qth similarity mean, and setting it as the first centroid reference image.

[0041] Furthermore, if the clustering results for a single reference image include multiple image categories, the process here focuses on the first category of reference images. Specifically, all reference images belonging to the first category are identified and extracted from the clustering results to form a set. A reference image is randomly selected from this set as the starting point, referred to as the first reference image. The similarity between this first reference image and all other images in the set is then calculated, resulting in a similarity set containing the similarity values ​​between the first reference image and each image in the set. The similarity set is then averaged to obtain a mean similarity value, the first mean similarity value, which reflects the average similarity between the first reference image and the other images in the set. In this case, the above steps are not repeated, but a new reference image in the set that has not been used as a starting point is selected as a new starting point. The similarity between this image and the other images in the set is then calculated again, resulting in a new mean similarity value. This process continues until every reference image in the set has been selected as a starting point at least once and its corresponding mean similarity value has been calculated. Assuming Q such calculations are performed, Q similarity means are obtained, and then similarity means from the first to the Qth are obtained. After obtaining all the similarity means, these means are compared to find the maximum value. The reference image corresponding to the maximum similarity mean is the image with the highest average similarity to the other images in the set among all the reference images. This reference image is set as the centroid reference image of the first category of reference image set. In summary, by calculating the mean similarity between each reference image and the other images in the set and finding the reference image with the maximum mean similarity, a centroid reference image that best represents the image class is determined, improving the accuracy of the comparison baseline and ensuring the precision of subsequent results.

[0042] In a preferred embodiment, the traversal of the benchmark image clustering results, performing image similarity analysis with the device images to be identified respectively, extracting the operating equipment model corresponding to the benchmark image with the maximum image similarity and outputting it, also includes: when the number of operating equipment models corresponding to the benchmark image with the maximum image similarity is greater than 1, analyzing multiple groups of screened operating equipment models of the device images to be identified at multiple moments; taking the operating equipment model row with the highest frequency of occurrence among the multiple groups of screened operating equipment models and outputting it.

[0043] Specifically, when there is not a single reference image with the highest similarity to the device image to be identified, but rather multiple reference images, each corresponding to a different model of the equipment being identified, this means that the features of the device image to be identified share a high degree of similarity with multiple reference images, making it difficult for the system to make a unique judgment. To address this issue, the system further analyzes the device image to be identified at multiple times, performing similarity analysis and filtering the device models for each image at each time. This process yields multiple sets of filtered device model results. Finally, the most frequently appearing device model among these filtered results is counted. Since this model appears in the device images to be identified at multiple times, it is considered the most likely correct answer. This most frequently appearing device model is then output as the final recognition result. This entire process, by traversing the reference image clustering results, performing image similarity analysis, handling cases where the maximum similarity value is not unique, and counting the most frequently appearing device models, achieves accurate recognition of the device image to be identified.

[0044] The method for intelligently identifying operating equipment for unplanned power transmission operations provided by the embodiments of the present invention has at least the following technical effects:

[0045] 1. By combining feature extraction, image similarity analysis, and weighted mean calculation techniques, we can efficiently extract key features from the image of the device to be identified and accurately match them with images in a reference image set. This approach not only improves recognition accuracy but also significantly reduces recognition time, enabling the system to quickly respond and output the correct operating device model.

[0046] 2. The system defines specific feature extraction dimensions and similarity calculation metrics for different types of equipment, and uses the Delphi method to determine appropriate weightings. This flexible design enables the system to adapt to complex and changing field environments, accurately identifying equipment of various shapes, sizes, and colors. Furthermore, when the maximum similarity value is not unique, the system can further analyze image data at multiple times to improve recognition robustness and reliability.

[0047] 3. Automated identification and output of equipment models significantly reduces manual intervention and error rates, improving overall automated processing and decision-making efficiency. This is particularly important for equipment identification tasks that require rapid response and efficient management, helping to reduce operating costs and improve production efficiency and safety.

[0048] Example 2

[0049] like Figure 2As shown, based on the same inventive concept as the method for intelligently identifying operating equipment for unplanned power transmission operations provided in Example 1, an embodiment of the present invention also provides an intelligent identification system for operating equipment for unplanned power transmission operations, the system comprising:

[0050] The information receiving module 11 is configured to receive an image of the operating device to be identified and world coordinate information of a positioning patch deployed on the operating device to be identified.

[0051] The intelligent sorting module 12 performs sorting based on the world coordinate information through the first-level identification model of the operating equipment to obtain a first-level selected operating equipment model set, wherein the first-level identification model of the operating equipment includes multiple sub-channels, and the machine learning of any sub-channel is obtained based on multiple groups of data training, and any group of the multiple groups of data includes: the world record coordinates of the positioning patch of the preset operating equipment model, and a label indicating whether it belongs to the preset operating equipment model.

[0052] The information screening module 13 is used to traverse the first-level selected operating equipment model sets when the number of the first-level selected operating equipment model sets is greater than 1, and obtain a reference image set of the first-level selected operating equipment model sets that meets the world coordinate information.

[0053] A cluster analysis module 14 is configured to perform similarity cluster analysis on the reference image set to obtain a reference image clustering result;

[0054] The result output module 15 is used to traverse the reference image clustering results, perform image similarity analysis with the device images to be identified, extract the operating device model corresponding to the reference image with the maximum image similarity, and output it.

[0055] Furthermore, the information receiving module 11 is further configured to perform the following steps:

[0056] The device element set is obtained by device element splitting of a preset device model, and a plurality of device operation motion monitoring logs of the preset device model are collected in a network according to the device element set, wherein any one of the plurality of device operation motion monitoring logs comprises a device element motion position time sequence information set; the device element set is respectively subjected to motion consistency clustering by traversing the plurality of device operation motion monitoring logs according to the device element motion position time sequence information set, and a plurality of device element clustering results are obtained; a plurality of device element groups are obtained by frequency grouping according to the plurality of device element clustering results, wherein any one of the plurality of device element groups has a frequency of being classified into the same category in the plurality of device element clustering results exceeding a selected frequency threshold, and the selected frequency threshold is greater than or equal to 0.9 of the number of device operation motion monitoring logs; a plurality of device elements are obtained by selecting random elements in the plurality of device element groups, as positioning patch deployment elements of the preset device model.

[0057] Further, the information receiving module 11 is further configured to perform the following steps:

[0058] According to the plurality of device operation motion monitoring logs, the device element motion position time sequence information set of the first device operation motion monitoring log is extracted; the device element motion position time sequence information set is subjected to motion similarity calculation to obtain a plurality of element motion trajectory similarities; based on a motion trajectory similarity threshold, the device element set is subjected to clustering analysis in combination with the plurality of element motion trajectory similarities to obtain a first device element clustering result, which is added to the plurality of device element clustering results.

[0059] Further, the intelligent sorting module 12 is further configured to perform the following steps:

[0060] The device action recognition loss function is constructed: , wherein, represents a device action recognition loss value, represents a label indicating whether it belongs to a preset operation device model, 1 represents that it belongs to, and 0 represents that it does not belong to, represents a judgment result of training output, is a dynamic weight, represents a probability of predicting that it belongs to a preset operation device model, represents the average value of all sample prediction probabilities, A hyperparameter that characterizes the speed at which the weights are controlled change; configuring the same number of input nodes according to the number of world record coordinates of the positioning patches of the preset operating equipment model, constructing a convolutional layer, a pooling layer, and a fully connected layer connected in sequence, and obtaining a sub-channel topology, wherein the fully connected layer includes an output layer, and the output layer includes two output nodes, 1 and 0; using the label indicating whether the identifier belongs to the preset operating equipment model as supervision and the world record coordinates of the positioning patches of the preset operating equipment model as input, the sub-channel topology is trained to generate the sub-channel.

[0061] Furthermore, the result output module 14 is further configured to perform the following steps:

[0062] Obtain a first reference image of the reference image set; activate an associated feature extraction channel to perform feature extraction based on the operating equipment model corresponding to the first reference image to obtain a first reference image feature extraction result; extract features of the image of the device to be identified based on the associated feature extraction channel to obtain a feature extraction result of the image of the device to be identified; calculate the first reference image similarity between the feature extraction result of the first reference image and the feature extraction result of the image of the device to be identified, and add them to the image similarity set; extract the operating equipment model corresponding to the reference image with the maximum value of the image similarity set and output it.

[0063] Furthermore, the result output module 15 is further configured to perform the following steps:

[0064] Extract the first centroid reference image of the first category reference image set of the reference image clustering result, until the Lth centroid reference image of the Lth category reference image set is extracted; traverse the first centroid reference image until the Lth centroid reference image, perform image similarity analysis with the image of the device to be identified, and extract the selected category reference image set of the category to which the centroid reference image with the maximum image similarity belongs; traverse the selected category reference image set, perform image similarity analysis with the image of the device to be identified, and extract the operating equipment model corresponding to the reference image with the maximum image similarity for output.

[0065] Furthermore, the result output module 15 is further configured to perform the following steps:

[0066] Extracting a first reference image from the first class of reference image sets and a first set of image similarities between the images within the class and the first reference image, performing mean calculations to obtain a first similarity mean; and calculating a Qth similarity mean. Extracting a reference image with a maximum value between the first similarity mean and the Qth similarity mean, setting this as the first centroid reference image. Furthermore, the result output module 15 is further configured to perform the following steps:

[0067] When the number of operating equipment models corresponding to the reference image with the maximum image similarity value is greater than 1, multiple groups of filtered operating equipment models of the equipment images to be identified at multiple moments are analyzed; the operating equipment model row with the highest frequency of appearance among the multiple groups of filtered operating equipment models is output.

[0068] Through the above detailed description of the method for intelligently identifying operating equipment for unplanned power transmission operations in this specification, those skilled in the art can clearly understand the intelligent identification system for operating equipment for unplanned power transmission operations in this embodiment. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0069] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for intelligently identifying operating equipment for unplanned power transmission operations, characterized in that: include: Receive an image of the device to be identified and world coordinate information of a positioning patch deployed on the device to be identified; According to the world coordinate information, sorting is performed using a first-level identification model for operating equipment to obtain a first-level selected operating equipment model set, wherein the first-level identification model for operating equipment includes multiple sub-channels, and any sub-channel machine learning is obtained based on multiple sets of data training, and any one of the multiple sets of data includes: the world record coordinates of the positioning patch of a preset operating equipment model, and a label indicating whether it belongs to the preset operating equipment model; When the number of the first-level selected operating equipment model sets is greater than 1, traverse the first-level selected operating equipment model sets to obtain a reference image set of the first-level selected operating equipment model sets that satisfies the world coordinate information; Performing similarity cluster analysis on the reference image set to obtain a reference image clustering result; Traversing the clustering results of the reference images, performing image similarity analysis with the images of the devices to be identified, extracting and outputting the operating device model corresponding to the reference image with the maximum image similarity; Among them, the world coordinate information of the positioning patch deployed on the working equipment to be identified includes: Decompose the device components of the preset device model to obtain a device component set; Based on the set of equipment components, a plurality of equipment operation motion monitoring logs of the preset equipment model are collected through the network, wherein any one of the plurality of equipment operation motion monitoring logs includes a set of equipment component motion position timing information; According to the equipment component motion position timing information set, traversing the plurality of equipment operation motion monitoring logs, respectively clustering the equipment component sets for motion consistency, and obtaining a plurality of equipment component clustering results; performing frequency grouping based on the clustering results of the multiple equipment components to obtain multiple groups of equipment components, wherein the frequency of any group of equipment components in the multiple groups of equipment components being classified as a same type based on the clustering results of the multiple equipment components exceeds a selected frequency threshold, and the selected frequency threshold is greater than or equal to 0.9 of the number of equipment operation motion monitoring logs; Randomly select components from the multiple groups of device components to obtain multiple device components as positioning patch deployment components of the preset device model.

2. The method according to claim 1, wherein According to the set of equipment component motion position timing information, the plurality of equipment operation motion monitoring logs are traversed to perform motion consistency clustering on the set of equipment components respectively, to obtain a plurality of equipment component clustering results, including: Extracting a set of equipment component motion position timing information from a first equipment operation motion monitoring log according to the plurality of equipment operation motion monitoring logs; Performing motion similarity calculation on the set of motion position sequence information of the device components to obtain similarities of motion trajectories of several components; Based on the motion trajectory similarity threshold, cluster analysis is performed on the device component set in combination with the motion trajectory similarities of the plurality of components to obtain a first device component clustering result, which is added to the plurality of device component clustering results.

3. The method according to claim 1, wherein Any sub-channel machine learning is obtained based on multiple sets of data training, and any set of the multiple sets of data includes: the world record coordinates of the positioning patch of the preset operation equipment model and a label indicating whether it belongs to the preset operation equipment model, including: According to the number of world record coordinates of the positioning patch of the preset operation equipment model, the same number of input nodes are configured, and a convolutional layer, a pooling layer, and a fully connected layer are constructed in sequence to obtain a sub-channel topology, wherein the fully connected layer includes an output layer, and the output layer includes two output nodes of 1 and 0; The sub-channel topology is trained by taking a label indicating whether the identifier belongs to a preset operating device model as supervision and the world record coordinates of the positioning patch of the preset operating device model as input to generate the sub-channel.

4. The method according to claim 1, wherein Traversing the reference image set, performing image similarity analysis with the device image to be identified, extracting the operating device model corresponding to the reference image with the maximum image similarity and outputting it, including: obtaining a first reference image of the reference image set; activating an associated feature extraction channel to perform feature extraction based on the operating equipment model corresponding to the first reference image, and obtaining a feature extraction result of the first reference image; Extracting features of the image of the device to be identified according to the associated feature extraction channel to obtain a feature extraction result of the image of the device to be identified; Calculating a first reference image similarity between the first reference image feature extraction result and the image feature extraction result of the device to be identified, and adding the similarity to an image similarity set; The operating equipment model corresponding to the reference image with the maximum value of the image similarity set is extracted and outputted.

5. The method according to claim 1, wherein Traversing the clustering results of the reference images, performing image similarity analysis with the images of the device to be identified, extracting the operating device model corresponding to the reference image with the maximum image similarity and outputting it, including: Extracting a first centroid reference image of a first type of reference image set from the reference image clustering result, until extracting an Lth centroid reference image of an Lth type of reference image set; Traversing the first centroid reference image up to the Lth centroid reference image, performing image similarity analysis with the image of the device to be identified, and extracting a selected class reference image set of the class to which the centroid reference image with the maximum image similarity belongs; The selected class reference image set is traversed, and image similarity analysis is performed with the image of the device to be identified respectively, and the operating device model corresponding to the reference image with the maximum image similarity is extracted and output.

6. The method according to claim 5, wherein Extracting a first centroid reference image of a first type of reference image set from the reference image clustering result, comprising: Extracting a first reference image from the first category of reference image sets and a first set of image similarities between images within the category and the first reference image, performing mean calculation on the similarities to obtain a first similarity mean; Until the Qth similarity mean is calculated; A reference image having a maximum value from the first similarity mean value to the Qth similarity mean value is extracted and set as the first centroid reference image.

7. The method according to claim 1, wherein Traversing the reference image set, performing image similarity analysis with the device image to be identified, extracting the operating device model corresponding to the reference image with the maximum image similarity and outputting it, further comprising: When the number of operating equipment models corresponding to the reference image with the maximum image similarity is greater than 1, analyzing multiple groups of images of the equipment to be identified at multiple moments to filter the operating equipment models; Take the row of the operating equipment model with the highest frequency among the multiple groups of screened operating equipment models and output it.

8. Intelligent identification system for operating equipment in unplanned power transmission operations, characterized in that: A system for implementing the method for intelligently identifying operating equipment for unplanned power transmission operations according to any one of claims 1 to 7, the system comprising: An information receiving module is used to receive an image of the operating device to be identified and world coordinate information of a positioning patch deployed on the operating device to be identified; An intelligent sorting module, which performs sorting based on the world coordinate information using a first-level identification model for operating equipment to obtain a first-level selected operating equipment model set, wherein the first-level identification model for operating equipment includes multiple sub-channels, and any sub-channel machine learning is obtained based on multiple sets of data training, and any set of the multiple sets of data includes: the world record coordinates of the positioning patch of a preset operating equipment model, and a label indicating whether it belongs to the preset operating equipment model; An information screening module is configured to, when the number of the first-level selected operating equipment model sets is greater than 1, traverse the first-level selected operating equipment model sets to obtain a reference image set of the first-level selected operating equipment model sets that satisfies the world coordinate information; A cluster analysis module, configured to perform similarity cluster analysis on the reference image set to obtain a reference image clustering result; The result output module is used to traverse the benchmark image clustering results, perform image similarity analysis with the device image to be identified, extract the operating device model corresponding to the benchmark image with the maximum image similarity, and output it.

Images