A method and device for real-time monitoring and early warning of power distribution cabinet status

Abstract

The present invention discloses a method and device for real-time monitoring and early warning of distribution cabinet status, which relates to the field related to distribution cabinet status monitoring. The method comprises: receiving multi-angle images of distribution cabinet collected by distribution cabinet monitoring camera; parsing multi-angle images of distribution cabinet through spatial positioning channel to obtain control key control mode information; receiving distribution cabinet input voltage; inputting control key control mode information and distribution cabinet input voltage into distribution cabinet electrical state prediction channel, outputting electrical state prediction value matrix; obtaining electrical state monitoring value matrix through electrical state sensor; when electrical state prediction value matrix is ​​inconsistent with electrical state monitoring value matrix, generating distribution cabinet abnormal operation early warning signal, and sending to distribution cabinet control terminal to execute early warning. The method solves the technical problems of low early warning accuracy and efficiency existing in existing real-time monitoring and early warning of distribution cabinet status, and achieves the technical effect of improving the accuracy and efficiency of early warning system.

Description

Technical Field

[0001] The present application relates to the field related to power distribution cabinet status monitoring, and in particular to a real-time monitoring and early warning method and device for power distribution cabinet status. Background Art

[0002] In the power system, the distribution cabinet is a key device for power distribution and control, and its operating status directly affects the stability and safety of the power system. In the traditional distribution cabinet status warning method, in order to realize the key status analysis of the distribution cabinet, complex spatial modeling processing is usually required. This method has problems such as large amount of calculation, high complexity, easy to make errors and low efficiency. These problems not only affect the accuracy and reliability of the warning system, but also limit the wide application of the warning system in the power system.

[0003] In the current related technologies, real-time monitoring and early warning of distribution cabinet status have technical problems such as low accuracy and efficiency of early warning. Summary of the invention

[0004] The present application provides a method and device for real-time monitoring and early warning of the status of a distribution cabinet. The method and device receive multi-angle images collected by a monitoring camera of the distribution cabinet, use a spatial positioning channel to parse the images to obtain control key control mode information, and combine the input voltage of the distribution cabinet to input this information into the electrical status prediction channel of the distribution cabinet. Through the trained prediction channel, an electrical status prediction value matrix is ​​output. At the same time, an electrical status monitoring value matrix is ​​obtained through an electrical status sensor, and the prediction value matrix is ​​compared with the monitoring value matrix. When the two are inconsistent, an abnormal operation early warning signal of the distribution cabinet is generated and sent to the distribution cabinet control terminal to execute early warning, etc., thereby achieving the technical effect of improving the accuracy and efficiency of the early warning system.

[0005] The present application provides a method for real-time monitoring and early warning of a power distribution cabinet state, comprising:

[0006] Receive multi-angle images of the distribution cabinet collected by the distribution cabinet monitoring camera; parse the multi-angle images of the distribution cabinet through the spatial positioning channel to obtain the control key control mode information; receive the input voltage of the distribution cabinet; input the control key control mode information and the input voltage of the distribution cabinet into the electrical state prediction channel of the distribution cabinet, and output the electrical state prediction value matrix; obtain the electrical state monitoring value matrix through the electrical state sensor; when the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, generate an abnormal operation warning signal of the distribution cabinet, and send it to the distribution cabinet control terminal to execute the warning.

[0007] In a possible implementation, a multi-angle image of a power distribution cabinet collected by a monitoring camera of the power distribution cabinet is received, and the following processing is performed:

[0008] The front image of the distribution cabinet is collected through the binocular camera of the first distribution cabinet; the first side image of the distribution cabinet is collected through the binocular camera of the second distribution cabinet; the second side image of the distribution cabinet is collected through the binocular camera of the third distribution cabinet; the back image of the distribution cabinet is collected through the binocular camera of the fourth distribution cabinet; the front image of the distribution cabinet, the first side image of the distribution cabinet, the second side image of the distribution cabinet and the back image of the distribution cabinet are added to the multi-angle image of the distribution cabinet.

[0009] In a possible implementation, the multi-angle image of the power distribution cabinet is parsed through a spatial positioning channel to obtain control key control mode information, and the following processing is performed:

[0010] The multi-angle image of the distribution cabinet has an image acquisition parameter identifier; through the spatial positioning channel, the image acquisition parameter identifier and the multi-angle image of the distribution cabinet are processed to obtain the spatial positioning label of the multi-angle image of the distribution cabinet; the control key identifier of the multi-angle image of the distribution cabinet is identified to obtain the control key area identifier of the multi-angle image of the distribution cabinet; according to the spatial positioning label of the multi-angle image of the distribution cabinet combined with the control key area identifier of the multi-angle image of the distribution cabinet, the control key spatial positioning information is obtained; according to the control key spatial positioning information, based on the control key preset modal library, the control key control modal information is matched.

[0011] In a possible implementation, the image acquisition parameter identifier and the multi-angle image of the power distribution cabinet are processed through the spatial positioning channel to obtain the multi-angle image spatial positioning label of the power distribution cabinet, and the following processing is performed:

[0012] Obtain a multi-angle image set and an image acquisition parameter record data set of a distribution cabinet; construct a spatial coordinate system of the distribution cabinet; based on the spatial coordinate system of the distribution cabinet, locate and identify the multi-angle image set of the distribution cabinet to obtain a multi-angle image location identification data set of the distribution cabinet; input the multi-angle image set of the distribution cabinet and the image acquisition parameter record data set into a convolutional neural network model, and train the spatial positioning channel with the multi-angle image location identification data set of the distribution cabinet as supervision.

[0013] In a possible implementation, the multi-angle image of the power distribution cabinet is subjected to control key identification, the control key area identification of the multi-angle image of the power distribution cabinet is obtained, and the following processing is performed:

[0014] A first control key reference image is obtained, wherein the first control key reference image includes a first control key front image and a first control key side image; based on the first control key front image, semantic segmentation is performed on the distribution cabinet front image and the distribution cabinet back image of the distribution cabinet multi-angle image to obtain a first control key front area identifier; based on the first control key side image, semantic segmentation is performed on the distribution cabinet first side image and the distribution cabinet second side image of the distribution cabinet multi-angle image to obtain a first control key side area identifier; the first control key front area identifier and the first control key side area identifier are stored as the first control key area identifier, and added to the distribution cabinet multi-angle image control key area identifier.

[0015] In a possible implementation, the control key control mode information and the input voltage of the power distribution cabinet are input into the electrical state prediction channel of the power distribution cabinet, an electrical state prediction value matrix is ​​output, and the following processing is performed:

[0016] A three-dimensional model of a power distribution cabinet is obtained and gridded to obtain a three-dimensional grid model of the power distribution cabinet; based on the three-dimensional grid model of the power distribution cabinet, control key control record mode, input record voltage and three-dimensional grid electrical indicator record state value are collected; based on the three-dimensional grid model of the power distribution cabinet, a distribution cabinet graph neural network topology is constructed, wherein the input nodes of the distribution cabinet graph neural network topology are the control key distribution area and the distribution cabinet voltage input area, and the output nodes of the distribution cabinet graph neural network topology are the grid distribution area; the control key control record mode and the input record voltage are input into the input node, the three-dimensional grid electrical indicator record state value is used as the supervision value of the output node, and the distribution cabinet electrical state prediction channel is trained.

[0017] In a possible implementation, the control key control recording mode and the input recording voltage are input into the input node, the three-dimensional grid electrical indicator recording state value is used as the supervision value of the output node, the distribution cabinet electrical state prediction channel is trained, and the following processing is performed:

[0018] Obtain first grid output data, wherein the first grid output data has a first-dimensional electrical state prediction value up to an M-th-dimensional electrical state prediction value; obtain first grid supervision data of the first grid output data, wherein the first grid supervision data has a first-dimensional electrical state supervision value up to an M-th-dimensional electrical state supervision value; calculate twice the product of the first-dimensional electrical state prediction value and the first-dimensional electrical state supervision value, and store it as a first loss parameter; calculate the sum of the square of the first-dimensional electrical state prediction value and the square of the first-dimensional electrical state supervision value, and store it as a second loss parameter; calculate the ratio of the second loss parameter to the first loss parameter, and set it as a first-dimensional loss coefficient; until the M-th-dimensional loss coefficient is obtained, the first-dimensional loss coefficient is counted up to the mean of the M-th-dimensional loss coefficient, and set it as a first grid training loss value; until the Q-th grid training loss value of the same training data is obtained, when the first grid training loss value up to the Q-th grid training loss value are all less than or equal to the convergence loss threshold, it is regarded as training convergence; when the training is continuously trained for several times, the proportion of training convergence times is greater than or equal to the convergence times proportion threshold, and the electrical state prediction channel of the distribution cabinet is generated.

[0019] The present application also provides a real-time monitoring and early warning device for a power distribution cabinet, comprising:

[0020] A multi-angle image receiving module for a distribution cabinet, the multi-angle image receiving module for a distribution cabinet is used to receive multi-angle images of a distribution cabinet collected by a monitoring camera for a distribution cabinet; a control key control modal information acquisition module, the control key control modal information acquisition module is used to parse the multi-angle image of the distribution cabinet through a spatial positioning channel to obtain control key control modal information; a distribution cabinet input voltage receiving module, the distribution cabinet input voltage receiving module is used to receive the input voltage of the distribution cabinet; an electrical state prediction value matrix output module, the electrical state prediction value matrix output module is used to input the control key control modal information and the distribution cabinet input voltage into the electrical state prediction channel of the distribution cabinet, and output the electrical state prediction value matrix; an electrical state monitoring value matrix acquisition module, the electrical state monitoring value matrix acquisition module is used to obtain the electrical state monitoring value matrix through an electrical state sensor; an early warning module, the early warning module is used to generate an abnormal operation early warning signal of the distribution cabinet when the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, and send it to the distribution cabinet control terminal to execute the early warning.

[0021] A method and device for real-time monitoring and early warning of the status of a distribution cabinet proposed in this application first receives multi-angle images of the distribution cabinet collected by a monitoring camera of the distribution cabinet, then parses the multi-angle images of the distribution cabinet through a spatial positioning channel, obtains control key control modal information, and then receives the input voltage of the distribution cabinet, and then inputs the control key control modal information and the input voltage of the distribution cabinet into the electrical state prediction channel of the distribution cabinet, outputs an electrical state prediction value matrix, and then obtains an electrical state monitoring value matrix through an electrical state sensor. When the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, an abnormal operation warning signal of the distribution cabinet is generated and sent to the distribution cabinet control terminal for execution of the warning, thereby achieving the technical effect of improving the accuracy and efficiency of the early warning system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solution of the embodiment of the present invention, the accompanying drawings of the embodiment of the present invention will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the device according to the embodiment of the present application. It should be understood that the previous or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0023] Figure 1 A flow chart of a method for real-time monitoring and early warning of the status of a power distribution cabinet provided in an embodiment of the present application.

[0024] Figure 2 A schematic diagram of the structure of a real-time monitoring and early warning device for the status of a power distribution cabinet provided in an embodiment of the present application.

[0025] Explanation of the accompanying drawings: distribution cabinet multi-angle image receiving module 10, control key control modal information acquisition module 20, distribution cabinet input voltage receiving module 30, electrical state prediction value matrix output module 40, electrical state monitoring value matrix acquisition module 50, early warning module 60. DETAILED DESCRIPTION

[0026] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0027] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0028] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, device, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0029] The present application provides a method for real-time monitoring and early warning of the status of a power distribution cabinet, such as Figure 1 As shown, the method includes:

[0030] Step S100, receiving multi-angle images of the distribution cabinet collected by the distribution cabinet monitoring camera. Specifically, multiple distribution cabinet monitoring cameras (camera devices used to capture multi-angle images of the distribution cabinet) are deployed at key positions of the distribution cabinet. These cameras are used to simultaneously capture images of the front, side, back and other perspectives of the distribution cabinet, thereby providing stereoscopic visual information. Each camera collects multi-angle images of the distribution cabinet according to preset shooting parameters (such as exposure time, white balance, etc.) and trigger mechanisms (such as timing trigger, event trigger) for spatial positioning and extraction of control key modal information.

[0031] In a possible implementation, a multi-angle image of a distribution cabinet captured by a monitoring camera of the distribution cabinet is received, and step S100 further includes step S110, capturing a front image of the distribution cabinet through a first binocular camera of the distribution cabinet; step S120, capturing a first side image of the distribution cabinet through a second binocular camera of the distribution cabinet; step S130, capturing a second side image of the distribution cabinet through a third binocular camera of the distribution cabinet; step S140, capturing a back image of the distribution cabinet through a fourth binocular camera of the distribution cabinet; step S150, adding the front image of the distribution cabinet, the first side image of the distribution cabinet, the second side image of the distribution cabinet and the back image of the distribution cabinet into the multi-angle image of the distribution cabinet.

[0032] Specifically, the binocular camera of the power distribution cabinet is a camera system that can capture two perspective images at the same time, and can provide accurate depth information and spatial positioning through stereo vision technology. Start the binocular camera of the power distribution cabinet, and collect the front image of the power distribution cabinet (including the front panel of the power distribution cabinet and the control keys, indicator lights, etc. on it), the first side image of the power distribution cabinet (including the control keys, wiring ports, etc. on the side of the power distribution cabinet), the second side image of the power distribution cabinet (the other side opposite to the first side) and the back image of the power distribution cabinet (including the rear panel of the power distribution cabinet and the wiring, heat dissipation device, etc. on it) according to the set parameters. Integrate the images of the four directions collected in steps S110-S140 to form a complete multi-angle image set of the power distribution cabinet. This implementation method can comprehensively observe various parts of the power distribution cabinet, including control keys, indicator lights, wiring ports, etc., by collecting the front, side and back images of the power distribution cabinet, thereby realizing comprehensive monitoring of the status of the power distribution cabinet and providing comprehensive visual information support for the early warning of the power distribution cabinet.

[0033] Step S200, through the spatial positioning channel, the multi-angle image of the distribution cabinet is parsed to obtain the control key control mode information. Specifically, the received multi-angle image of the distribution cabinet is preprocessed, including denoising, contrast enhancement, etc., to improve the image quality. The spatial positioning channel is used to match and locate the feature points in the image, and a spatial positioning label of the multi-angle image of the distribution cabinet is generated. The control keys in the image are identified and marked, and the control key area identification of the multi-angle image of the distribution cabinet is generated. According to the spatial positioning label and the control key area identification, combined with the control key preset mode library, the control mode information of the control key is matched. Among them, the spatial positioning channel is a model for processing image spatial information, which can generate a spatial positioning label of the image. The control key control mode information is information describing the current state of the control key (such as on / off, adjustment position, etc.) and possible behaviors (such as about to be pressed, about to be adjusted, etc.).

[0034] In a possible implementation, the multi-angle image of the distribution cabinet is parsed through the spatial positioning channel to obtain the control key control mode information, and step S200 further includes step S210, wherein the multi-angle image of the distribution cabinet has an image acquisition parameter identifier. Specifically, when acquiring the multi-angle image of the distribution cabinet, the camera device will automatically add the image acquisition parameter identifier, and these parameter identifiers include but are not limited to the shooting time, the position, angle, model, focal length, aperture size, shutter speed, etc. of the camera. Step S220, through the spatial positioning channel, the image acquisition parameter identifier and the multi-angle image of the distribution cabinet are processed to obtain the spatial positioning label of the multi-angle image of the distribution cabinet. Specifically, the received multi-angle image of the distribution cabinet is first pre-processed, including denoising, contrast enhancement, etc., to improve the image quality. The model in the spatial positioning channel is used to extract the feature points in the image, wherein the feature points include corner points, edges, textures, etc. Matching and positioning between images are performed according to the image acquisition parameter identifier (such as camera position, focal length, etc.) and the extracted feature points. According to the results of matching and positioning, a spatial positioning label is generated for each multi-angle image of the distribution cabinet, and this label describes the position and posture of the image in three-dimensional space. Step S230, the control key identification is performed on the multi-angle image of the distribution cabinet to obtain the control key area identification of the multi-angle image of the distribution cabinet. Specifically, a template image of the control key is prepared, and a template matching algorithm is used to find an area similar to the control key template in the multi-angle image of the distribution cabinet. According to the results of template matching, the multi-angle image of the distribution cabinet is divided into a control key area and a non-control key area. A unique identification is generated for each control key area, including information such as the position, size, and shape of the control key. Step S240, according to the spatial positioning label of the multi-angle image of the distribution cabinet combined with the control key area identification of the multi-angle image of the distribution cabinet, the control key spatial positioning information is obtained. Specifically, the spatial positioning label generated in step S220 is combined with the control key area identification generated in step S230, and the control key area identification is converted into the control key position information in three-dimensional space according to the information in the spatial positioning label. Step S250, according to the control key spatial positioning information, based on the control key preset modal library, match the control key control modal information. Specifically, the control key preset modal library is a database containing various control keys and their possible control modes. This library includes the shape, color, position of the control key and the corresponding control mode (such as on / off, adjustment, etc.). According to the control key spatial positioning information obtained in step S240, the best matching control key and its control mode are found in the preset modal library, and the matched control key control modal information is output. This implementation method accurately obtains the position and state information of the control key through the combination of the spatial positioning channel and the control key preset modal library, thereby improving the accuracy of electrical state prediction.

[0035] In a possible implementation, the image acquisition parameter identifier and the multi-angle image of the distribution cabinet are processed through the spatial positioning channel to obtain the spatial positioning label of the multi-angle image of the distribution cabinet, and step S220 further includes step S221, obtaining a multi-angle image set of the distribution cabinet and an image acquisition parameter record data set; step S222, constructing a spatial coordinate system of the distribution cabinet. Specifically, the multi-angle image set of the distribution cabinet is used for spatial positioning, the image acquisition parameter record data set is used to provide a correspondence between the image and the actual physical space, the distribution cabinet spatial coordinate system is a three-dimensional coordinate system, which is used to represent the actual spatial position of the distribution cabinet and its internal components, and the distribution cabinet spatial coordinate system is constructed to convert the pixel position in the image into the position in the actual physical space, so as to achieve accurate positioning of the internal components of the distribution cabinet. Step S223, based on the spatial coordinate system of the distribution cabinet, the multi-angle image set of the distribution cabinet is positioned and identified to obtain a multi-angle image positioning identification data set of the distribution cabinet. Specifically, the pixel position in the multi-angle image of the distribution cabinet is corresponded to the position in the spatial coordinate system of the distribution cabinet, and the actual spatial position of each pixel point in each image is obtained. Specifically, a binocular camera system is used to simultaneously capture images of the distribution cabinet from two different angles. In the two captured images, the same feature points (such as corner points, edges, etc.) are extracted through a feature matching algorithm, and the three-dimensional coordinates of each feature point are calculated using the geometric relationship of the binocular camera (such as baseline distance, focal length, etc.) and the matching feature point positions. The calculated three-dimensional coordinates of the feature points are corresponded to the spatial coordinate system of the distribution cabinet, thereby achieving the spatial positioning of the distribution cabinet and its internal components. According to the results of spatial positioning, each pixel point in the multi-angle image of the distribution cabinet is assigned its position information in the spatial coordinate system of the distribution cabinet, that is, the multi-angle image positioning identification data set of the distribution cabinet refers to a set of multi-angle images of the distribution cabinet after positioning identification, in which each pixel point corresponds to its position in the spatial coordinate system of the distribution cabinet.

[0036] Step S224, input the multi-angle image set of the distribution cabinet and the image acquisition parameter record data set into the convolutional neural network model, and use the multi-angle image positioning identification data set of the distribution cabinet as supervision to train the spatial positioning channel. Specifically, the multi-angle image set of the distribution cabinet, the image acquisition parameter record data set, and the multi-angle image positioning identification data set of the distribution cabinet are input into the convolutional neural network model for supervised training, so that the model learns to predict the spatial position information of each pixel in the image from the input image and parameters, and the spatial positioning channel is obtained after the training is completed. This implementation method can automatically learn the features in the image and realize efficient and accurate spatial positioning of the multi-angle image of the distribution cabinet by using the convolutional neural network model for supervised training, thereby improving the computational efficiency and enhancing the robustness and accuracy of positioning.

[0037] In a possible implementation, the control key identification is performed on the multi-angle image of the distribution cabinet to obtain the control key area identification of the multi-angle image of the distribution cabinet, and step S230 further includes step S231, obtaining a first control key reference image, wherein the first control key reference image includes a first control key front image and a first control key side image. Specifically, high-definition images of all control keys on the distribution cabinet are collected, including the front and side perspectives of the control keys. For each control key, the front and side images that best represent its appearance features (such as shape, color, label, etc.) are selected as reference images, that is, the first control key reference image refers to the front and side high-definition images of a specific control key on the distribution cabinet, which are used as a reference for image matching and semantic segmentation. Step S232, based on the first control key front image, semantic segmentation is performed on the distribution cabinet front image and the distribution cabinet back image of the multi-angle image of the distribution cabinet to obtain the first control key front area identification. Specifically, a deep learning model is used to perform semantic segmentation on the front image and the back image in the multi-angle image of the distribution cabinet. During the segmentation process, the area similar to the front image of the first control key is identified by image similarity matching (including comparison of features such as shape, color, and texture). When a matching area is found, the area is selected using an anchor frame as the first control key front area identifier. Step S233, based on the first control key side image, the first side image of the distribution cabinet and the second side image of the distribution cabinet in the multi-angle image of the distribution cabinet are semantically segmented to obtain the first control key side area identifier. Specifically, similar to step 232, the first side and second side images of the distribution cabinet are semantically segmented. Using the first control key side image as a reference, the area similar to the side image is identified and marked as the first control key side area, and the anchor frame is also used to select these areas as the first control key side area identifier. Step S234, the first control key front area identifier and the first control key side area identifier are stored as the first control key area identifier, and added to the control key area identifier of the multi-angle image of the distribution cabinet. Specifically, the front area identifier and the side area identifier of the first control key obtained in steps 232 and 233 are combined to form a first control key area identifier, and this identifier is added to the database of control key area identifiers of the multi-angle image of the distribution cabinet, and this database is used to store the area identifiers of all control keys in the multi-angle image of the distribution cabinet. This implementation method not only considers the front image of the control key, but also the side image, monitors the status of the control key from multiple angles, and improves the comprehensiveness and accuracy of the monitoring.

[0038] Step S300, receiving the input voltage of the power distribution cabinet. Specifically, the input voltage of the power distribution cabinet, that is, the power supply voltage received by the power distribution cabinet, is collected through a voltage sensor or a measuring instrument.

[0039] Step S400, input the control key control modal information and the distribution cabinet input voltage into the distribution cabinet electrical state prediction channel, and output the electrical state prediction value matrix. Specifically, the control key control modal information and the distribution cabinet input voltage are integrated into input data, and the input data is processed using the distribution cabinet electrical state prediction channel to output the electrical state prediction value matrix. Among them, the distribution cabinet electrical state prediction channel is a model for predicting the electrical state of the distribution cabinet, and the electrical state prediction value matrix is ​​a matrix containing the electrical state prediction values ​​of the distribution cabinet, which is used for comparison with the monitoring value.

[0040] In a possible implementation, the control key control mode information and the input voltage of the distribution cabinet are input into the distribution cabinet electrical state prediction channel, and the electrical state prediction value matrix is ​​output. Step S400 further includes step S410, obtaining the three-dimensional model of the distribution cabinet for grid processing, and obtaining the three-dimensional grid model of the distribution cabinet. Specifically, the three-dimensional model of the distribution cabinet is obtained by computer-aided design (CAD) software, and the three-dimensional model is grid processed, that is, the continuous three-dimensional space is divided into a series of small, interconnected grid units (such as triangles, etc.) for numerical analysis and calculation, and each grid unit contains information about its position, shape and size. Finally, the three-dimensional grid model of the distribution cabinet is output. Step S420, based on the three-dimensional grid model of the distribution cabinet, the control key control record mode, input record voltage and three-dimensional grid electrical indicator record state value are collected. Specifically, according to the three-dimensional grid model of the distribution cabinet, the position of the control key and the voltage input point is determined. The control record mode of the control key, that is, a series of state changes of the control key during the operation process, including information such as the pressing, lifting, pressing time and lifting time of the control key. Collect the input recorded voltage of the distribution cabinet (the voltage value of the distribution cabinet input and its change record), and at the same time, collect the electrical indicator record state value of each grid unit in the three-dimensional grid, such as current, voltage, resistance, etc. Step S430, based on the three-dimensional grid model of the distribution cabinet, construct the distribution cabinet graph neural network topology, wherein the input nodes of the distribution cabinet graph neural network topology are the control key distribution area and the distribution cabinet voltage input area, and the output nodes of the distribution cabinet graph neural network topology are the grid distribution area. Specifically, determine the input nodes and output nodes of the distribution cabinet graph neural network topology, wherein the input nodes correspond to the control key distribution area and the distribution cabinet voltage input area, and the output nodes correspond to the grid distribution area. According to the connection relationship of the three-dimensional grid model of the distribution cabinet, construct the edges and nodes in the graph neural network. The edges represent the connection relationship between the grid units, and the nodes represent the grid units themselves. Step S440, input the control key control record mode and the input record voltage into the input node, use the three-dimensional grid electrical indicator record state value as the supervision value of the output node, and train the distribution cabinet electrical state prediction channel. Specifically, the collected control key control record mode and input record voltage are used as the input data of the input node, and the collected three-dimensional grid electrical indicator record state value is used as the supervision value (i.e., target value) of the output node. The back propagation algorithm is used to train the graph neural network so that it can predict the electrical state of the distribution cabinet. During the training process, the parameters of the graph neural network are continuously adjusted to minimize the difference between the predicted value and the actual value. After the training is completed, the electrical state prediction channel of the distribution cabinet is obtained. This implementation method uses the graph neural network topology to construct the electrical state prediction channel of the distribution cabinet. The graph neural network can capture the connection relationship and mutual influence between the various components in the distribution cabinet, thereby improving the accuracy of the electrical state prediction.

[0041] In a possible implementation, the control key control record mode and the input record voltage are input into the input node, the three-dimensional grid electrical indicator record state value is used as the supervision value of the output node, and the electrical state prediction channel of the distribution cabinet is trained. Step S440 further includes step S441, obtaining the first grid output data, wherein the first grid output data has the first dimension electrical state prediction value up to the Mth dimension electrical state prediction value. Specifically, in the forward propagation process of the graph neural network, the control key control record mode and the input record voltage are input into the input node, and after the network calculation, the output data of the first grid is obtained from the output node. This output data is a multi-dimensional vector, in which each dimension represents a predicted value of an electrical state, from the first dimension electrical state prediction value to the Mth dimension electrical state prediction value. Step S442, obtain the first grid supervision data of the first grid output data, wherein the first grid supervision data has the first dimension electrical state supervision value up to the Mth dimension electrical state supervision value. Specifically, for the first grid, corresponding supervision data is obtained from the state value recorded in the three-dimensional grid electrical indicator. The first grid supervision data is the real electrical state data used to compare with the network output data during the training process, and is also a multidimensional vector, from the first dimension electrical state supervision value to the Mth dimension electrical state supervision value.

[0042] Step S443, calculate twice the product of the first-dimensional electrical state prediction value and the first-dimensional electrical state supervision value, and store it as the first loss parameter; step S444, calculate the sum of the square of the first-dimensional electrical state prediction value and the square of the first-dimensional electrical state supervision value, and store it as the second loss parameter; step S445, calculate the ratio of the second loss parameter to the first loss parameter, and set it as the first-dimensional loss coefficient. Specifically, calculate twice the product of the first-dimensional electrical state prediction value and the supervision value as the first loss parameter. Calculate the sum of the square of the first-dimensional electrical state prediction value and the square of the supervision value as the second loss parameter. The ratio of the second loss parameter to the first loss parameter is used as the loss coefficient of the first dimension. Among them, the loss parameter is a parameter used to measure the difference between the network prediction value and the actual value, and the loss coefficient is used to evaluate the performance of the network in a certain dimension of electrical state prediction.

[0043] Step S446, until the M-th dimension loss coefficient is obtained, the first dimension loss coefficient is counted until the mean value of the M-th dimension loss coefficient is set as the first grid training loss value. Specifically, the loss coefficients of each dimension are counted, their mean values ​​are calculated, and the training loss value of the first grid is obtained. This loss value reflects the overall prediction performance of the graph neural network on the first grid. Step S447, until the Q-th grid training loss value of the same training data is obtained, when the first grid training loss value until the Q-th grid training loss value is less than or equal to the convergence loss threshold, the training is considered to have converged. Specifically, for the same training data, the above steps are repeated to obtain the training loss values ​​of multiple grids. If the training loss values ​​of all grids are less than or equal to the set convergence loss threshold (the threshold used to determine whether the training has converged), it is considered that the training has converged, the performance of the graph neural network is stable during the training process, and the loss value no longer decreases significantly. Step S448, when the training is continuously trained for several times, the proportion of training convergence times is greater than or equal to the convergence times proportion threshold, and the electrical state prediction channel of the distribution cabinet is generated. Specifically, if during multiple consecutive training sessions, the percentage of training convergence times is greater than or equal to the set convergence times percentage threshold (the threshold used to determine whether the graph neural network is stable), the graph neural network is considered to be stable, and a distribution cabinet electrical state prediction channel is generated. This implementation method uses multi-dimensional loss coefficients for training, which can capture the complex relationship between the electrical states of the distribution cabinet. Through multiple training and convergence judgments, it ensures that the graph neural network can work stably in a variety of situations, improves the accuracy of prediction, and enhances the robustness of the graph neural network.

[0044] Step S500, obtaining an electrical state monitoring value matrix through an electrical state sensor. Specifically, electrical state sensors are deployed on key electrical components of the power distribution cabinet. The electrical state sensors collect electrical state data in real time, such as current, voltage, temperature, etc., and organize the collected data into an electrical state monitoring value matrix. The electrical state monitoring value matrix is ​​a matrix containing the electrical state monitoring values ​​of the power distribution cabinet, which is used for comparison with the predicted values.

[0045] Step S600, when the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, an abnormal operation warning signal of the distribution cabinet is generated and sent to the distribution cabinet control terminal for early warning. Specifically, the electrical state prediction value matrix and the electrical state monitoring value matrix are compared to determine whether there is an inconsistency. When an inconsistency is found, an abnormal operation warning signal of the distribution cabinet is generated. The abnormal operation warning signal of the distribution cabinet is used to indicate that the distribution cabinet may be operating abnormally. The abnormal operation warning signal of the distribution cabinet is sent to the distribution cabinet control terminal to trigger an early warning mechanism, such as an audible and visual alarm, SMS notification, etc. The embodiment of the present application adopts a method of receiving multi-angle images collected by a monitoring camera of a distribution cabinet, and uses a spatial positioning channel to parse the images to obtain control key control modal information, and combines this information with the input voltage of the distribution cabinet to input the information into the electrical state prediction channel of the distribution cabinet. Through the trained prediction channel, an electrical state prediction value matrix is ​​output. At the same time, an electrical state monitoring value matrix is ​​obtained through an electrical state sensor, and the prediction value matrix is ​​compared with the monitoring value matrix. When the two are inconsistent, an abnormal operation warning signal of the distribution cabinet is generated and sent to the distribution cabinet control terminal to execute the warning and other technical means, thereby achieving the technical effect of improving the accuracy and efficiency of the warning system.

[0046] In the above, refer to Figure 1 A method for real-time monitoring and early warning of a power distribution cabinet state according to an embodiment of the present invention is described in detail. Figure 2 A real-time monitoring and early warning device for the status of a power distribution cabinet according to an embodiment of the present invention is described.

[0047] A real-time monitoring and early warning device for the state of a power distribution cabinet according to an embodiment of the present invention is used to solve the technical problems of low early warning accuracy and efficiency in the existing real-time monitoring and early warning of the state of a power distribution cabinet, and achieve the technical effect of improving the accuracy and efficiency of the early warning system. A real-time monitoring and early warning device for the state of a power distribution cabinet comprises: a multi-angle image receiving module 10 for a power distribution cabinet, a control key control mode information acquisition module 20, a power distribution cabinet input voltage receiving module 30, an electrical state prediction value matrix output module 40, an electrical state monitoring value matrix acquisition module 50, and an early warning module 60.

[0048] The multi-angle image receiving module 10 of the distribution cabinet is used to receive the multi-angle image of the distribution cabinet collected by the distribution cabinet monitoring camera; the control key control modal information acquisition module 20 is used to parse the multi-angle image of the distribution cabinet through the spatial positioning channel to obtain the control key control modal information; the distribution cabinet input voltage receiving module 30 is used to receive the input voltage of the distribution cabinet; the electrical state prediction value matrix output module 40 is used to input the control key control modal information and the distribution cabinet input voltage into the electrical state prediction channel of the distribution cabinet, and output the electrical state prediction value matrix; the electrical state monitoring value matrix acquisition module 50 is used to obtain the electrical state monitoring value matrix through the electrical state sensor; the early warning module 60 is used to generate an abnormal operation early warning signal of the distribution cabinet when the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, and send it to the distribution cabinet control terminal to execute the early warning.

[0049] The specific configuration of the multi-angle image receiving module 10 of the distribution cabinet will be described in detail below. As described above, the multi-angle image of the distribution cabinet collected by the distribution cabinet monitoring camera is received, and the multi-angle image receiving module 10 of the distribution cabinet may further include: a distribution cabinet front image acquisition unit for collecting the front image of the distribution cabinet through the first distribution cabinet binocular camera; a distribution cabinet first side image acquisition unit for collecting the first side image of the distribution cabinet through the second distribution cabinet binocular camera; a distribution cabinet second side image acquisition unit for collecting the second side image of the distribution cabinet through the third distribution cabinet binocular camera; a distribution cabinet back image acquisition unit for collecting the back image of the distribution cabinet through the fourth distribution cabinet binocular camera; a distribution cabinet multi-angle image acquisition unit for adding the distribution cabinet front image, the distribution cabinet first side image, the distribution cabinet second side image and the distribution cabinet back image into the distribution cabinet multi-angle image.

[0050] The specific configuration of the control key control modal information acquisition module 20 will be described in detail below. As described above, the multi-angle image of the distribution cabinet is parsed through the spatial positioning channel to obtain the control key control modal information. The control key control modal information acquisition module 20 may further include: a spatial positioning unit for the multi-angle image of the distribution cabinet having an image acquisition parameter identifier, and the image acquisition parameter identifier and the multi-angle image of the distribution cabinet are processed through the spatial positioning channel to obtain the multi-angle image spatial positioning label of the distribution cabinet; a control key identification unit is used to identify the control key of the multi-angle image of the distribution cabinet to obtain the control key area identifier of the multi-angle image of the distribution cabinet; a control key spatial positioning information acquisition unit is used to obtain the control key spatial positioning information according to the multi-angle image spatial positioning label of the distribution cabinet combined with the multi-angle image control key area identifier of the distribution cabinet; a control key control modal information matching unit is used to match the control key control modal information according to the control key spatial positioning information and based on the control key preset modal library.

[0051] Among them, through the spatial positioning channel, the image acquisition parameter identification and the multi-angle image of the distribution cabinet are processed to obtain the spatial positioning label of the multi-angle image of the distribution cabinet, and the spatial positioning unit may further include: an image information acquisition subunit is used to obtain a multi-angle image set of the distribution cabinet and an image acquisition parameter record data set; a distribution cabinet spatial coordinate system construction subunit is used to construct a distribution cabinet spatial coordinate system; a distribution cabinet multi-angle image positioning identification data set acquisition subunit is used to locate and identify the multi-angle image set of the distribution cabinet based on the distribution cabinet spatial coordinate system, and obtain a distribution cabinet multi-angle image positioning identification data set; a spatial positioning channel training subunit is used to input the multi-angle image set of the distribution cabinet and the image acquisition parameter record data set into a convolutional neural network model, and train the spatial positioning channel with the distribution cabinet multi-angle image positioning identification data set as supervision.

[0052] Among them, control key identification is performed on the multi-angle image of the distribution cabinet to obtain the control key area identification of the multi-angle image of the distribution cabinet, and the control key identification unit may further include: a first control key reference image acquisition subunit is used to obtain the first control key reference image, wherein the first control key reference image includes a first control key front image and a first control key side image; a semantic segmentation subunit is used to perform semantic segmentation on the distribution cabinet front image and the distribution cabinet back image of the multi-angle image of the distribution cabinet according to the first control key front image to obtain the first control key front area identification, and perform semantic segmentation on the distribution cabinet first side image and the distribution cabinet second side image of the multi-angle image of the distribution cabinet according to the first control key side image to obtain the first control key side area identification; the first control key area identification generation subunit is used to store the first control key front area identification and the first control key side area identification as the first control key area identification, and add them into the control key area identification of the multi-angle image of the distribution cabinet.

[0053] Next, the specific configuration of the electrical state prediction value matrix output module 40 will be described in detail. As described above, the control key control modal information and the distribution cabinet input voltage are input into the distribution cabinet electrical state prediction channel, and the electrical state prediction value matrix is ​​output. The electrical state prediction value matrix output module 40 may further include: a distribution cabinet three-dimensional grid model acquisition unit is used to obtain the distribution cabinet three-dimensional model for grid processing to obtain the distribution cabinet three-dimensional grid model; a data acquisition unit is used to collect the control key control record modality, input record voltage and three-dimensional grid electrical indicator record state value based on the distribution cabinet three-dimensional grid model; a distribution cabinet graph neural network topology construction unit is used to construct a distribution cabinet graph neural network topology based on the distribution cabinet three-dimensional grid model, wherein the input nodes of the distribution cabinet graph neural network topology are the control key distribution area and the distribution cabinet voltage input area, and the output nodes of the distribution cabinet graph neural network topology are the grid distribution area; a distribution cabinet electrical state prediction channel training unit is used to input the control key control record modality and the input record voltage into the input node, and use the three-dimensional grid electrical indicator record state value as the supervision value of the output node to train the distribution cabinet electrical state prediction channel.

[0054] Wherein, the control key control record mode and the input record voltage are input into the input node, the three-dimensional grid electrical indicator record state value is used as the supervision value of the output node, and the distribution cabinet electrical state prediction channel is trained. The distribution cabinet electrical state prediction channel training unit may further include: a first grid output data acquisition subunit is used to obtain the first grid output data, wherein the first grid output data has a first-dimensional electrical state prediction value up to the M-th dimension electrical state prediction value; a first grid supervision data acquisition subunit is used to obtain the first grid supervision data of the first grid output data, wherein the first grid supervision data has a first-dimensional electrical state supervision value up to the M-th dimension electrical state supervision value; a first loss parameter acquisition subunit is used to calculate twice the product of the first-dimensional electrical state prediction value and the first-dimensional electrical state supervision value, and store it as a first loss parameter; a second loss parameter acquisition subunit is used to obtain the first grid supervision data of the first grid output data, wherein the first grid supervision data has a first-dimensional electrical state supervision value up to the M-th dimension electrical state supervision value. The parameter acquisition subunit is used to calculate the sum of the square of the first-dimensional electrical state prediction value and the square of the first-dimensional electrical state supervision value, and store it as the second loss parameter; the first-dimensional loss coefficient acquisition subunit is used to calculate the ratio of the second loss parameter to the first loss parameter, and set it as the first-dimensional loss coefficient; the first-grid training loss value acquisition subunit is used to obtain the M-th dimension loss coefficient, count the first-dimensional loss coefficient until the average of the M-th dimension loss coefficient, and set it as the first grid training loss value; the distribution cabinet electrical state prediction channel generation subunit is used to obtain the Q-th grid training loss value of the same training data, when the first grid training loss value until the Q-th grid training loss value are all less than or equal to the convergence loss threshold, it is regarded as training convergence, and when the training is trained several times in succession, the proportion of training convergence times is greater than or equal to the convergence times proportion threshold, the distribution cabinet electrical state prediction channel is generated.

[0055] A real-time monitoring and early warning device for the status of a power distribution cabinet provided in an embodiment of the present invention can execute a real-time monitoring and early warning method for the status of a power distribution cabinet provided in any embodiment of the present invention, and has functional modules and beneficial effects corresponding to the execution method.

[0056] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0057] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A real-time monitoring and early warning method for power distribution cabinet status, characterized in that: include: Receive multi-angle images of the distribution cabinet collected by the distribution cabinet monitoring camera; Through the spatial positioning channel, the multi-angle image of the power distribution cabinet is analyzed to obtain the control mode information of the control key; Receive input voltage from power distribution cabinet; Input the control key control mode information and the distribution cabinet input voltage into the distribution cabinet electrical state prediction channel, and output the electrical state prediction value matrix; Obtaining an electrical state monitoring value matrix through an electrical state sensor; When the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, an abnormal operation warning signal of the power distribution cabinet is generated and sent to the power distribution cabinet control terminal for execution of the warning; Through the spatial positioning channel, the multi-angle image of the power distribution cabinet is analyzed to obtain the control key control mode information, including: The multi-angle image of the power distribution cabinet has an image acquisition parameter identifier; The image acquisition parameter identifier and the multi-angle image of the power distribution cabinet are processed through the spatial positioning channel to obtain a spatial positioning label of the multi-angle image of the power distribution cabinet; Performing control key marking on the multi-angle image of the power distribution cabinet to obtain control key area markings of the multi-angle image of the power distribution cabinet; Obtaining control key spatial positioning information according to the multi-angle image spatial positioning label of the power distribution cabinet and the multi-angle image control key area identifier of the power distribution cabinet; According to the control key spatial positioning information, based on a control key preset modality library, matching the control key control modality information; Inputting the control key control mode information and the input voltage of the power distribution cabinet into the electrical state prediction channel of the power distribution cabinet, and outputting the electrical state prediction value matrix, including: Obtain a three-dimensional model of the power distribution cabinet and perform mesh processing to obtain a three-dimensional mesh model of the power distribution cabinet; Based on the three-dimensional grid model of the power distribution cabinet, the control key control recording mode, the input recording voltage and the three-dimensional grid electrical index recording state value are collected; Based on the three-dimensional grid model of the power distribution cabinet, a power distribution cabinet graph neural network topology is constructed, wherein the input nodes of the power distribution cabinet graph neural network topology are the control key distribution area and the power distribution cabinet voltage input area, and the output nodes of the power distribution cabinet graph neural network topology are the grid distribution area; Input the control key control record mode and the input record voltage into the input node, use the three-dimensional grid electrical index record state value as the supervision value of the output node, and train the distribution cabinet electrical state prediction channel; The control key control recording mode and the input recording voltage are input into the input node, the three-dimensional grid electrical index recording state value is used as the supervision value of the output node, and the electrical state prediction channel of the power distribution cabinet is trained, including: Obtaining first grid output data, wherein the first grid output data has a first-dimensional electrical state prediction value up to an M-th-dimensional electrical state prediction value; Obtaining first grid supervision data of the first grid output data, wherein the first grid supervision data has a first dimension electrical state supervision value up to an Mth dimension electrical state supervision value; Calculate twice the product of the first-dimensional electrical state prediction value and the first-dimensional electrical state supervision value, and store the result as a first loss parameter; Calculate the sum of the square of the first-dimensional electrical state prediction value and the square of the first-dimensional electrical state supervision value, and store the sum as a second loss parameter; Calculate the ratio of the second loss parameter to the first loss parameter and set it as the first-dimensional loss coefficient; Until the M-th dimension loss coefficient is obtained, the first dimension loss coefficient is counted until the average value of the M-th dimension loss coefficient is set as the first grid training loss value; Until the Qth grid training loss value of the same training data is obtained, when the first grid training loss value until the Qth grid training loss value are all less than or equal to the convergence loss threshold, the training is considered to be converged; When the training is performed continuously for several times and the proportion of the training convergence times is greater than or equal to the convergence times proportion threshold, the electrical state prediction channel of the distribution cabinet is generated.

2. A method for real-time monitoring and early warning of power distribution cabinet status according to claim 1, characterized in that: Receive multi-angle images of the distribution cabinet collected by the distribution cabinet monitoring camera, including: The front image of the power distribution cabinet is collected by using the binocular camera of the first power distribution cabinet; Using the binocular camera of the second distribution cabinet, a first side image of the distribution cabinet is collected; The second side image of the power distribution cabinet is collected by using the binocular camera of the third power distribution cabinet; The image of the back of the power distribution cabinet is collected by using the binocular camera of the fourth power distribution cabinet; The front image of the power distribution cabinet, the first side image of the power distribution cabinet, the second side image of the power distribution cabinet and the back image of the power distribution cabinet are added to the multi-angle image of the power distribution cabinet.

3. A method for real-time monitoring and early warning of power distribution cabinet status as claimed in claim 1, characterized in that: The image acquisition parameter identifier and the multi-angle image of the power distribution cabinet are processed through the spatial positioning channel to obtain the multi-angle image spatial positioning label of the power distribution cabinet, including: Obtain a multi-angle image set of a power distribution cabinet and a data set of image acquisition parameter records; Construct the space coordinate system of the power distribution cabinet; Based on the spatial coordinate system of the power distribution cabinet, the multi-angle image set of the power distribution cabinet is positioned and identified to obtain a multi-angle image positioning and identification data set of the power distribution cabinet; The multi-angle image set of the power distribution cabinet and the image acquisition parameter record data set are input into a convolutional neural network model, and the spatial positioning channel is trained with the multi-angle image positioning identification data set of the power distribution cabinet as supervision.

4. A method for real-time monitoring and early warning of power distribution cabinet status according to claim 1, characterized in that: Performing control key identification on the multi-angle image of the power distribution cabinet to obtain control key area identification of the multi-angle image of the power distribution cabinet includes: Obtaining a first control key reference image, wherein the first control key reference image includes a first control key front image and a first control key side image; According to the first control key front image, semantic segmentation is performed on the power distribution cabinet front image and the power distribution cabinet back image of the power distribution cabinet multi-angle image to obtain a first control key front area identifier; According to the first control key side image, semantic segmentation is performed on the first side image of the power distribution cabinet and the second side image of the power distribution cabinet in the multi-angle image of the power distribution cabinet to obtain a first control key side area identifier; The first control key front area identifier and the first control key side area identifier are stored as the first control key area identifier, and added into the distribution cabinet multi-angle image control key area identifier.

5. A real-time monitoring and early warning device for the status of a power distribution cabinet, characterized in that: The device is used to implement a real-time monitoring and early warning method for a power distribution cabinet state according to any one of claims 1 to 4, and the device comprises: A multi-angle image receiving module for a power distribution cabinet, wherein the multi-angle image receiving module for a power distribution cabinet is used to receive multi-angle images of the power distribution cabinet collected by a monitoring camera of the power distribution cabinet; A control key control modal information acquisition module, the control key control modal information acquisition module is used to analyze the multi-angle image of the power distribution cabinet through a spatial positioning channel to obtain the control key control modal information; A power distribution cabinet input voltage receiving module, wherein the power distribution cabinet input voltage receiving module is used to receive the power distribution cabinet input voltage; An electrical state prediction value matrix output module, the electrical state prediction value matrix output module is used to input the control key control mode information and the distribution cabinet input voltage into the distribution cabinet electrical state prediction channel, and output the electrical state prediction value matrix; An electrical state monitoring value matrix acquisition module, wherein the electrical state monitoring value matrix acquisition module is used to obtain an electrical state monitoring value matrix through an electrical state sensor; The early warning module is used to generate an abnormal operation early warning signal of the distribution cabinet when the electrical state prediction value matrix is ​​inconsistent with the electrical state monitoring value matrix, and send it to the distribution cabinet control terminal to execute the early warning.

Images