A fire protection monitoring method and system for circuit protection of electrical cabinet

Abstract

The present invention relates to the technical field of design of circuit protection and fire prevention monitoring schemes for electrical cabinets, and specifically to a circuit protection and fire prevention monitoring method and system for electrical cabinets. The present invention measures the current and temperature of the main circuit, control circuit, and power supply end of the measuring circuit of the electrical cabinet, and then trains an automatic recognition model for abnormal states of the electrical cabinet, and then determines whether the electrical cabinet is in an abnormal state. If so, the on-site staff conducts an on-site inspection of the electrical cabinet and uploads the inspection results. The automatic recognition model for abnormal states of the electrical cabinet can be automatically retrained based on the inspection results, and then the automatic recognition model for abnormal states of the electrical cabinet can be automatically optimized, realizing an overall evaluation and judgment of the overall operation data of the electrical cabinet over a period of time, and more accurately judging whether the overall circuit operation data of the electrical cabinet over a period of time is in an abnormal state, which greatly improves the safety, reliability, and intelligence of circuit protection and fire prevention monitoring of the electrical cabinet.

Description

Technical Field

[0001] The present invention relates to the technical field of design of fire protection monitoring schemes for circuit protection of electrical cabinets, and in particular to a method and system for fire protection monitoring of circuit protection of electrical cabinets. Background Art

[0002] The circuit system of the electrical cabinet tends to be complicated. However, in the prior art, the circuit system of the electrical cabinet is mostly monitored by setting a single or multiple thresholds. Fire supervision personnel and maintenance personnel usually detect the operating status of the circuit system of the electrical cabinet based on experience. The coupling relationship between the current and temperature of each circuit in the circuit system of the electrical cabinet often hides problems that are difficult to be directly discovered by people, such as aging of circuit components, excessive leakage current of circuit components, etc. If this problem is to be manually checked, it requires tedious manual testing and measuring each part of the circuit one by one, which takes a lot of time and may not be able to be checked, because during the inspection process, it is often necessary to turn off the power for inspection, or the detection is Components are missing, and directly replacing the main control circuit is often costly. Therefore, the existing technology lacks an effective evaluation scheme for the coupling state between the current and temperature of each circuit in the circuit system of the electrical cabinet. The traditional fault diagnosis mode is heavily dependent on the technical level and experience accumulation of supervisors / maintenance personnel. On the one hand, the technical threshold is high. On the other hand, it may be effective for the circuit system of a simple electrical cabinet, but for the circuit system of the current electrical cabinet with increasingly complex structures, the traditional mode is not only inefficient, but also inaccurate. Early detection of problems such as aging of circuit components and excessive leakage current of circuit components can enable safer, more reliable and more timely monitoring of the electrical cabinet.

[0003] In the prior art, whether an electrical cabinet is abnormal is often monitored by setting a single or multiple thresholds of multiple sensors. However, since the thresholds are all obtained through manual experience, there are huge errors. That is, the setting of a single or multiple thresholds cannot accurately determine whether there is an abnormality in the circuit of the electrical cabinet, and it is easy to be misjudged by the system. It is difficult to comprehensively evaluate and judge the overall operating data of the circuit of the electrical cabinet over a period of time, that is, it is difficult to accurately determine whether the overall operating data of the electrical cabinet over a period of time is in an abnormal state.

[0004] Therefore, the prior art needs to be further developed. Summary of the invention

[0005] The purpose of the present invention is to overcome the above technical deficiencies and provide an electrical cabinet circuit protection fire monitoring method and system to solve the problems existing in the prior art.

[0006] To achieve the above technical purpose, according to a first aspect of the present invention, the present invention provides a method for monitoring fire protection of circuits in an electrical cabinet, wherein the electrical cabinet comprises a cabinet body, wherein a main circuit, a control circuit, and a measuring circuit are arranged in the cabinet body, wherein the main circuit, the control circuit, and the measuring circuit all comprise power supply ends, wherein the power supply ends of the main circuit, the control circuit, and the measuring circuit are respectively connected to current sensors and thermistors for measuring the current and temperature of the power supply ends of the main circuit, the control circuit, and the measuring circuit, wherein the method comprises:

[0007] S100, after the electrical cabinet is installed and before it is put into use, within a first preset time, the current and temperature data of the power supply end of the main circuit, the control circuit, and the measuring circuit collected at the same time point are recorded as a first current and temperature data set, and multiple first current and temperature data sets collected at multiple time points are obtained at preset time intervals, and the multiple first current and temperature data sets are respectively divided into a training data set and a verification data set according to a first preset ratio, and the training data set is used to train the automatic recognition model of the abnormal state of the electrical cabinet; the verification data set is used to optimize the trained automatic recognition model of the abnormal state of the electrical cabinet, optimize the model parameters, and complete the training of the automatic recognition model of the abnormal state of the electrical cabinet;

[0008] S200, after the target electrical cabinet is put into use, within a first preset time, the current and temperature data of the power supply end of the main circuit, the control circuit, and the measuring circuit collected at the same time point are recorded as a second current and temperature data set, and multiple second current and temperature data sets collected at multiple time points are obtained at preset time intervals, and the multiple second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, and the recognition results are output, and then it is determined whether the electrical cabinet is in an abnormal state, and whether to output an alarm signal related to the abnormal state of the electrical cabinet according to the judgment result;

[0009] S300. If so, the on-site staff will conduct an on-site inspection of the electrical cabinet to confirm whether the electrical cabinet is in an abnormal state, and use the cloud server to record and transmit the on-site inspection result data, obtain the on-site inspection result data, and determine based on the on-site inspection result data whether to combine the multiple second current and temperature data sets as new training data sets, retrain the automatic recognition model for abnormal states of the electrical cabinet, and then automatically optimize the automatic recognition model for abnormal states of the electrical cabinet.

[0010] Specifically, the method further includes:

[0011] After the electrical cabinet is installed and before it is put into use, it is necessary to test the installed electrical cabinet to ensure that there are no abnormalities in the current and temperature data at the power supply end of the main circuit, control circuit, and measurement circuit.

[0012] Specifically, the method of using the training data set to train the automatic recognition model for abnormal state of the electrical cabinet; using the verification data set to optimize the trained automatic recognition model for abnormal state of the electrical cabinet, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal state of the electrical cabinet, includes:

[0013] The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer. The Transformer network layer is used to forward propagate the current and temperature data sets of the power supply end of the main circuit, control circuit and measurement circuit collected at the current time point to predict the current and temperature data sets of the power supply end of the main circuit, control circuit and measurement circuit at the next time point, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization to determine the direction in which the parameter gradient of the automatic recognition model of the abnormal state of the electrical cabinet decreases fastest; the automatic recognition model of the abnormal state of the electrical cabinet performs back propagation according to the loss value and the parameter gradient of the model to optimize the parameters of the automatic recognition model of the abnormal state of the electrical cabinet.

[0014] Specifically, the method of using the training data set to train the automatic recognition model for abnormal state of the electrical cabinet; using the verification data set to optimize the trained automatic recognition model for abnormal state of the electrical cabinet, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal state of the electrical cabinet, further includes:

[0015] After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

[0016] Specifically, the determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes:

[0017] If the loss value meets the first preset condition, the training is terminated and the parameters of the current electrical cabinet abnormal state automatic identification model are output; if the loss value does not meet the first preset condition, the training is continued.

[0018] Specifically, the first preset condition is:

[0019] After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

[0020] Specifically, the recognition result is that the electrical cabinet is in an abnormal state or the electrical cabinet is not in an abnormal state, the plurality of second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, the recognition result is output, and then it is determined whether the electrical cabinet is in an abnormal state, and whether to output an alarm signal about the electrical cabinet being in an abnormal state according to the determination result, including:

[0021] If the output result is that the electrical cabinet is in an abnormal state, it is determined that an alarm signal related to the abnormal state of the electrical cabinet is output;

[0022] If the output result is that there is no abnormal state in the electrical cabinet, it is determined that no alarm signal regarding the abnormality of the electrical cabinet is output.

[0023] Specifically, the field inspection result data includes whether an abnormal state actually exists or does not actually exist, and the determination based on the field inspection result data whether to combine the plurality of second current and temperature data sets as a newly added training data set, retrain the electrical cabinet abnormal state automatic recognition model, and then automatically optimize the electrical cabinet abnormal state automatic recognition model includes:

[0024] If the field inspection result data indicates that an abnormal state actually exists, the multiple second current and temperature data sets are combined as a new training data set, and the automatic recognition model for abnormal states of electrical cabinets is retrained, thereby automatically optimizing the automatic recognition model for abnormal states of electrical cabinets.

[0025] Specifically, the field inspection result data includes whether an abnormal state actually exists or does not actually exist, and the method of determining whether to combine the plurality of second current and temperature data sets as a newly added training data set, retraining the electrical cabinet abnormal state automatic recognition model, and then automatically optimizing the electrical cabinet abnormal state automatic recognition model according to the field inspection result data also includes:

[0026] If the field inspection result data shows that there is actually no abnormal state, the multiple second current and temperature data sets are not combined as a new training data set, the electrical cabinet abnormal state automatic recognition model is not retrained, and the electrical cabinet abnormal state automatic recognition model is not automatically optimized.

[0027] According to a second aspect of the present invention, there is provided an electrical cabinet circuit protection fire monitoring system, wherein the electrical cabinet comprises a cabinet body, wherein a main circuit, a control circuit, and a measuring circuit are arranged in the cabinet body, wherein the main circuit, the control circuit, and the measuring circuit all comprise a power supply end, including:

[0028] An acquisition module, the acquisition module includes a current sensor and a thermistor respectively connected to the power supply ends of the main circuit, the control circuit, and the measurement circuit, and is used to measure the current and temperature of the power supply ends of the main circuit, the control circuit, and the measurement circuit;

[0029] A control module is used to record the current and temperature data of the power supply end of the main circuit, control circuit, and measurement circuit collected at the same time point within a first preset time after the electrical cabinet is installed and before it is put into use as a first current and temperature data set, obtain multiple first current and temperature data sets collected at multiple time points according to preset time intervals, divide the multiple first current and temperature data sets into training data sets and verification data sets according to first preset ratios, use the training data sets to train the automatic recognition model for abnormal states of the electrical cabinet; use the verification data sets to optimize the trained automatic recognition model for abnormal states of the electrical cabinet, optimize the model parameters, and complete the training of the automatic recognition model for abnormal states of the electrical cabinet; and use the verification data sets to record the current and temperature data of the power supply end of the main circuit, control circuit, and measurement circuit collected at the same time point within a first preset time after the target electrical cabinet is put into use. The current and temperature data are recorded as a second current and temperature data set. Multiple second current and temperature data sets collected at multiple time points are obtained according to a preset time interval. The multiple second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, and the recognition result is output to determine whether the electrical cabinet is in an abnormal state. According to the judgment result, it is determined whether to output an alarm signal regarding the abnormal state of the electrical cabinet; if so, the on-site staff conducts an on-site inspection of the electrical cabinet to confirm whether the electrical cabinet is in an abnormal state, and uses a cloud server to record and transmit the on-site inspection result data, obtain the on-site inspection result data, and determine whether to combine the multiple second current and temperature data sets as a new training data set based on the on-site inspection result data, retrain the electrical cabinet abnormal state automatic recognition model, and then automatically optimize the electrical cabinet abnormal state automatic recognition model.

[0030] Beneficial effects:

[0031] The present invention utilizes current sensors and thermistors respectively connected to the power supply ends of the main circuit, control circuit, and measuring circuit to measure the current and temperature of the power supply ends of the main circuit, control circuit, and measuring circuit, and then trains an automatic recognition model for abnormal states of an electrical cabinet, and then determines whether the electrical cabinet is in an abnormal state. If so, on-site staff conducts an on-site inspection of the electrical cabinet and uploads the inspection results. The automatic recognition model for abnormal states of the electrical cabinet can be automatically retrained based on the inspection results, and then the automatic recognition model for abnormal states of the electrical cabinet is automatically optimized, thereby realizing an overall evaluation and judgment of the overall operation data of the electrical cabinet over a period of time, and more accurately judging whether the overall circuit operation data of the electrical cabinet over a period of time is in an abnormal state, thereby greatly improving the safety, reliability, and intelligence of circuit protection and fire prevention monitoring of the electrical cabinet, greatly reducing the safety hazard of fire in the electrical cabinet, and solving the problem that the setting of a single or multiple thresholds cannot accurately judge whether the electrical cabinet is in an abnormal state, thereby greatly improving the intelligence and usability of the present invention, and greatly expanding the application scenarios of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a flow chart of a method for monitoring fire prevention of an electrical cabinet circuit protection provided in a specific embodiment of the present invention;

[0033] Figure 2 Schematic diagram of the system composition of the electrical cabinet circuit protection fire monitoring system provided in a specific embodiment of the present invention;

[0034] Figure 3 It is a structural diagram of an automatic identification model for abnormal status of an electrical cabinet provided in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention is clearly and completely described below in conjunction with the accompanying drawings of the present invention. Based on the embodiments in this application, other similar embodiments obtained by ordinary technicians in this field without making creative work should all fall within the scope of protection of this application. In addition, the directional words mentioned in the following embodiments, such as "up", "down", "left", "right", etc., are only reference to the directions of the accompanying drawings. Therefore, the directional words used are used to illustrate rather than limit the invention.

[0036] The present invention will be further described below in conjunction with the accompanying drawings and preferred embodiments.

[0037] See also Figure 1-Figure 3 The present invention provides a method for monitoring fire prevention of circuit protection of an electrical cabinet, comprising:

[0038] The electrical cabinet comprises a cabinet body, a main circuit, a control circuit, and a measuring circuit are arranged in the cabinet body, the main circuit, the control circuit, and the measuring circuit all comprise power supply ends, the power supply ends of the main circuit, the control circuit, and the measuring circuit are respectively connected to current sensors and thermistors for measuring current and temperature of the power supply ends of the main circuit, the control circuit, and the measuring circuit, and the method comprises:

[0039] S100. After the electrical cabinet is installed and before it is put into use, within a first preset time, the current and temperature data of the power supply ends of the main circuit, the control circuit, and the measuring circuit collected at the same time point are recorded as a first current and temperature data set, and multiple first current and temperature data sets collected at multiple time points are obtained at preset time intervals. The multiple first current and temperature data sets are respectively divided into training data sets and verification data sets according to a first preset ratio, and the training data sets are used to train an automatic recognition model for abnormal states of the electrical cabinet; the verification data sets are used to optimize the trained automatic recognition model for abnormal states of the electrical cabinet, optimize model parameters, and complete the training of the automatic recognition model for abnormal states of the electrical cabinet.

[0040] It should be noted that the step before step S100 includes:

[0041] A first preset time, a preset time interval, a first preset ratio, a first preset round, and a second preset round are preset in the control module.

[0042] It can be understood that the first preset time, preset time interval, first preset ratio, first preset round, and second preset round can be specifically set according to the actual needs of the user of the present invention. The present invention does not limit the specific values ​​of the above parameters, and there is no regularity in setting the specific values ​​of the above parameters, as long as it is applicable to the electrical cabinet circuit protection fire monitoring method proposed in the present invention.

[0043] Preferably, the present invention sets the first preset time to 12 days, the present invention sets the preset time interval to 1 minute, the present invention sets the first preset ratio to 8:2, the present invention sets the first preset round to 20,000 times, and the present invention sets the second preset round to 200. The above settings can further improve the performance of the model and the efficiency of model training, and greatly improve the accuracy and reliability of the prediction of the automatic identification model of the abnormal state of the electrical cabinet of the present invention.

[0044] It can be understood that when the electrical cabinet is in a normal state, there is a certain coupling law between the current and temperature of the power supply ends of the main circuit, control circuit and measurement circuit of the electrical cabinet. For example, there is a certain dynamic balance relationship between the current of the power supply ends of the main circuit, control circuit and measurement circuit and their corresponding temperatures. When the electrical cabinet is in an abnormal state, this balance relationship will be broken. Based on this, the present invention designs and trains an automatic recognition model for abnormal states of electrical cabinets.

[0045] Specifically, the method further includes:

[0046] After the electrical cabinet is installed and before it is put into use, it is necessary to test the installed electrical cabinet to ensure that there are no abnormalities in the current and temperature data at the power supply end of the main circuit, control circuit, and measurement circuit.

[0047] It is understandable that how to ensure that there are no abnormalities in the current and temperature data at the power supply end of the main circuit, control circuit, and measurement circuit can be checked by relevant staff according to industry standards, and the present invention will not go into details here.

[0048] S200. After the target electrical cabinet is put into use, within a first preset time, the current and temperature data of the power supply ends of the main circuit, the control circuit, and the measuring circuit collected at the same time point are recorded as a second current and temperature data set, and multiple second current and temperature data sets collected at multiple time points are obtained at preset time intervals. The multiple second current and temperature data sets are input into the automatic recognition model of the abnormal state of the electrical cabinet, and the recognition results are output to determine whether the electrical cabinet is in an abnormal state, and whether to output an alarm signal regarding the abnormal state of the electrical cabinet based on the judgment result.

[0049] Specifically, the method of using the training data set to train the automatic recognition model for abnormal state of the electrical cabinet; using the verification data set to optimize the trained automatic recognition model for abnormal state of the electrical cabinet, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal state of the electrical cabinet, includes:

[0050] The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer. The Transformer network layer is used to forward propagate the current and temperature data sets of the power supply end of the main circuit, control circuit and measurement circuit collected at the current time point to predict the current and temperature data sets of the power supply end of the main circuit, control circuit and measurement circuit at the next time point, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization to determine the direction in which the parameter gradient of the automatic recognition model of the abnormal state of the electrical cabinet decreases fastest; the automatic recognition model of the abnormal state of the electrical cabinet performs back propagation according to the loss value and the parameter gradient of the model to optimize the parameters of the automatic recognition model of the abnormal state of the electrical cabinet.

[0051] It can be understood that Transformer is a deep learning model based on the self-attention mechanism, which consists of a multi-head attention mechanism and a feedforward neural network. It can process sequence data and capture long-distance dependencies. The Transformer model does not contain traditional convolutional layers, but processes each element in the input sequence through an attention layer. The typical structure of Transformer includes an encoder and a decoder, wherein the encoder is composed of multiple identical layers stacked together, and the decoder adds a self-attention layer on the basis of the encoder to process sequence generation tasks. The present invention uses Transformer to supplement global context information, enhances the model's capabilities in feature extraction and global understanding, greatly improves the performance of the model, and greatly reduces the computational cost.

[0052] Specifically, the preset network layer also includes a ResNet network layer, and the ResNet network layer is used to classify and predict the output results based on the collected current and temperature data sets, and the classification results include whether the electrical cabinet is in an abnormal state or whether the electrical cabinet does not have an abnormal state.

[0053] It should be noted here that ResNet is a deep convolutional neural network that solves the gradient vanishing problem in deep network training by introducing residual connections. The core component of ResNet is the residual block, which usually contains two or three convolutional layers and a jump connection, which allows the gradient to bypass these convolutional layers directly. The network structure of ResNet can be very deep. These networks extract the features of pressure data by stacking multiple residual blocks, and classify and predict the output results through global average pooling and fully connected layers. The classification results include whether the electrical cabinet is in an abnormal state or not.

[0054] See also Figure 3 The present invention combines the ResNet network layer and the Transformer network layer. ResNet is good at extracting local features of pressure data in the training data set, while Transformer can supplement global context information. This combination can enhance the model's capabilities in feature extraction and global understanding, greatly improve the performance of the model, and greatly reduce the computational cost.

[0055] It can be understood that the ResNet network layer is used to classify and predict the output results through the current and temperature data of the power supply end of the main circuit, control circuit, and measurement circuit collected in the training data set and the verification data set. The specific steps are:

[0056] (1) First, put the training data set into the Backbones network layer of the model. Backbone refers to a series of convolutional layers that constitute the skeleton of the neural network. The main function of these layers is to extract the features of the input data. The Backbone network is usually composed of multiple convolutional layers, pooling layers, and activation functions. It can extract meaningful feature representations from the original data. Backbones is composed of multiple Backbones, and the core network layer in Backbone is the Resnet network layer. After the data set is input into the model, the first current and data set will be input into Backbones for training;

[0057] (2) When constructing Backbone, circuit name encoding and Resnet network layers were added. After the training data set was input, Resnet began to train layer by layer to extract features from the current and temperature data sets, assisting Transformer in predicting whether the electrical cabinet is in an abnormal state. The extracted features were used with the help of circuit name encoding to obtain the circuit name information described in the current and temperature data. The circuit name information includes the main circuit, control circuit, and measurement circuit.

[0058] Specifically, the main circuit includes: a switch, a fuse, a contactor, and a relay.

[0059] Specifically, the control circuit includes: a switch, a button, and an MCU control chip.

[0060] The measuring circuit comprises: a voltmeter, an ammeter and an electric energy meter.

[0061] It can be understood that the composition of the main circuit, control circuit and measurement circuit can be specifically arranged according to actual needs or different functions of different electrical cabinets. The present invention does not limit the composition of the main circuit, control circuit and measurement circuit, as long as it is applicable to the electrical cabinet circuit protection fire monitoring method proposed in the present invention.

[0062] Specifically, the method of using the training data set to train the automatic recognition model for abnormal state of the electrical cabinet; using the verification data set to optimize the trained automatic recognition model for abnormal state of the electrical cabinet, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal state of the electrical cabinet, further includes:

[0063] After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

[0064] Specifically, after each training, the verification data set is input into the preset network layer of the previous training in batches to verify the model parameters, and the cyclic training is performed, including the following methods:

[0065] When the number of training times is not enough and the model needs to be optimized, the training can only be performed from the beginning. Therefore, the present invention develops a method for continuing the following training based on the previously trained model parameter file;

[0066] (1) Provide custom parameters, which can be set according to user needs to continue training based on any training parameters;

[0067] (2) Load the parameters of the custom model file;

[0068] (3) Put the verification data set into the custom parameter network layer for verification, obtain the loss value and record the model parameters of the model corresponding to the minimum loss value during the training process;

[0069] (4) Based on this set of model parameters, forward propagation training and back propagation are performed, and the optimizer optimizes the parameters to obtain a new set of data;

[0070] (5) Put the verification data into the new data network layer for inference to obtain the loss value;

[0071] (6) Perform cyclic training to obtain the best model and complete model optimization.

[0072] It is understandable that the training method adopted by the present invention is batch training, which means that each time the model parameters are updated, only a part of the samples in the validation data set is used, which is called a batch. The advantages of batch training are that it can reduce memory consumption, speed up training, increase randomness, and facilitate the generalization of the model.

[0073] It should be noted here that, in the present invention, batch_size=4, and the batch_size is the batch size. For example, when batch_size=4, it means that four first pressure data sets are selected from the verification data set each time and put into the model for verification.

[0074] Specifically, the determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes:

[0075] If the loss value meets the first preset condition, the training is terminated and the parameters of the current electrical cabinet abnormal state automatic identification model are output; if the loss value does not meet the first preset condition, the training is continued.

[0076] Specifically, the first preset condition is:

[0077] After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

[0078] It can be understood that the present invention preferably sets the maximum round of cyclic training to 20,000 times, which can effectively ensure the accuracy of the model prediction of the present invention and record the loss value of the model training. The present invention sets the second preset round to 200, that is, when the loss value of the training no longer decreases within 200 rounds of training, the round with the smallest loss value in the 200 rounds of training is saved as the best verification round parameter to generate the best model file. The above setting makes the training rounds no longer rely on manual judgment, but uses the deep learning model for automatic training, which improves the accuracy of the model prediction, and automatically ends the training when the accuracy meets the requirements, which greatly saves the training time, and can also effectively prevent overfitting, which greatly improves the intelligence level of the present invention and the efficiency of model training.

[0079] Specifically, the recognition result is that the electrical cabinet is in an abnormal state or the electrical cabinet is not in an abnormal state, the plurality of second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, the recognition result is output, and then it is determined whether the electrical cabinet is in an abnormal state, and whether to output an alarm signal about the electrical cabinet being in an abnormal state according to the determination result, including:

[0080] If the output result is that the electrical cabinet is in an abnormal state, it is determined that an alarm signal related to the abnormal state of the electrical cabinet is output;

[0081] If the output result is that there is no abnormal state in the electrical cabinet, it is determined that no alarm signal regarding the abnormality of the electrical cabinet is output.

[0082] Furthermore, the specific process of the electrical cabinet abnormal state automatic identification model of the present invention is as follows:

[0083] (1) Obtain the potential vector mean mu and logarithmic variance logvar in the forward propagation process of the automatic recognition model of the abnormal state of the electrical cabinet;

[0084] (2) Calculate a reasonable loss value representing the training effect:

[0085] Calculate the divergence: k1=-0.5*(1+logvar-mu^2-(e^logvar));

[0086] Calculate the loss between the current and temperature data in the validation data set and the predicted value:

[0087] loss=L1+k1*k1_weight;

[0088] Among them, loss is the loss value, k1 is the divergence, k1_weight is the proportion of k1, the value is 10, L1 is the absolute difference between the predicted value and the true value, and the calculation formula of L1 is as follows:

[0089] ;

[0090] Among them, A is the L1 loss in pytorch, which calculates the average or sum of the absolute differences between two tensors; A has a shape of N*C*H*W, which indicates that it is a four-dimensional tensor, N represents the number of samples (Batch Size), indicating that there are N data points, C represents the number of channels (Channels), H and W represent the height and width respectively, which are usually the two-dimensional spatial dimensions of the input features; P represents the padding value, which is a Boolean type, and the shape of P is c*h*w*1. The dimensions of P are the same as those of A in terms of channel, height, and width, but there is no sample number N, which means that P is a mask or padding information for a feature map of each sample, and the Boolean value of P may be used to mark whether the elements in A are filled (True) or valid (False).

[0091] (3) The loss value when using the validation data set to validate the current model network;

[0092] (4) If the validation loss of this round is less than that of the previous round, the best model parameters are replaced with the training model of this round;

[0093] (5) Put the rounds and loss values ​​into the dictionary one by one;

[0094] (6) Compare the loss value of the current round with the loss values ​​of the 200 rounds before this round (parameters are provided and can be set according to actual needs, the default is 200), that is, traverse the loss values ​​from epoch-200 to epoch-1 (epoch is the training round, and the comparison is performed when the epoch is greater than 200). If the loss values ​​of the current round of epoch are all greater than the loss values ​​from (epoch-200) to (epoch-1), then stop training.

[0095] (7) Save the best model parameters in the cyclic verification as a ckpt model file.

[0096] S300. If so, the on-site staff will conduct an on-site inspection of the electrical cabinet to confirm whether the electrical cabinet is in an abnormal state, and use the cloud server to record and transmit the on-site inspection result data, obtain the on-site inspection result data, and determine based on the on-site inspection result data whether to combine the multiple second current and temperature data sets as new training data sets, retrain the automatic recognition model for abnormal states of the electrical cabinet, and then automatically optimize the automatic recognition model for abnormal states of the electrical cabinet.

[0097] Specifically, the field inspection result data includes whether an abnormal state actually exists or does not actually exist, and the determination based on the field inspection result data whether to combine the plurality of second current and temperature data sets as a newly added training data set, retrain the electrical cabinet abnormal state automatic recognition model, and then automatically optimize the electrical cabinet abnormal state automatic recognition model includes:

[0098] If the field inspection result data indicates that an abnormal state actually exists, the multiple second current and temperature data sets are combined as a new training data set, and the automatic recognition model for abnormal states of electrical cabinets is retrained, thereby automatically optimizing the automatic recognition model for abnormal states of electrical cabinets.

[0099] Specifically, the field inspection result data includes whether an abnormal state actually exists or does not actually exist, and the method of determining whether to combine the plurality of second current and temperature data sets as a newly added training data set, retraining the electrical cabinet abnormal state automatic recognition model, and then automatically optimizing the electrical cabinet abnormal state automatic recognition model according to the field inspection result data also includes:

[0100] If the field inspection result data shows that there is actually no abnormal state, the multiple second current and temperature data sets are not combined as a new training data set, the electrical cabinet abnormal state automatic recognition model is not retrained, and the electrical cabinet abnormal state automatic recognition model is not automatically optimized.

[0101] It is understandable that the prediction result of the electrical cabinet abnormal state automatic identification model of the present invention only includes the presence of an abnormal state or the absence of an abnormal state. When an abnormal state exists, the corresponding predicted value and true value are 1, and when there is no abnormal state, the corresponding predicted value and true value are 0.

[0102] It can be understood that the present invention utilizes the current sensors and thermistors respectively connected to the power supply ends of the main circuit, control circuit, and measuring circuit to measure the current and temperature of the power supply ends of the main circuit, control circuit, and measuring circuit, and then trains the automatic recognition model of the abnormal state of the electrical cabinet, and then determines whether the electrical cabinet is in an abnormal state. If so, the on-site staff conducts an on-site inspection of the electrical cabinet and uploads the inspection results. The automatic recognition model of the abnormal state of the electrical cabinet can be automatically retrained according to the inspection results, and then the automatic recognition model of the abnormal state of the electrical cabinet is automatically optimized, realizing an overall evaluation and judgment of the overall operation data of the electrical cabinet over a period of time, and more accurately judging whether the overall circuit operation data of the electrical cabinet over a period of time is in an abnormal state, which greatly improves the safety, reliability and intelligence of the circuit protection and fire prevention monitoring of the electrical cabinet, greatly reduces the safety hazard of fire in the electrical cabinet, and solves the problem that the setting of a single or multiple thresholds cannot accurately judge whether the electrical cabinet is in an abnormal state, which greatly improves the intelligence and usability of the present invention and greatly expands the application scenarios of the present invention.

[0103] See also Figure 2The present invention provides another embodiment, which provides an electrical cabinet circuit protection fire monitoring system, wherein the electrical cabinet includes a cabinet body, wherein a main circuit, a control circuit, and a measuring circuit are arranged in the cabinet body, wherein the main circuit, the control circuit, and the measuring circuit all include a power supply end, and the system includes:

[0104] An acquisition module 100, comprising a current sensor and a thermistor connected to the power supply ends of the main circuit, the control circuit, and the measurement circuit, respectively, for measuring the current and temperature of the power supply ends of the main circuit, the control circuit, and the measurement circuit;

[0105] The control module 200 is used to record the current and temperature data of the power supply end of the main circuit, the control circuit, and the measuring circuit collected at the same time point as a first current and temperature data set within a first preset time after the electrical cabinet is installed and before it is put into use, obtain multiple first current and temperature data sets collected at multiple time points according to preset time intervals, divide the multiple first current and temperature data sets into training data sets and verification data sets according to a first preset ratio, use the training data set to train the automatic recognition model for abnormal states of the electrical cabinet; use the verification data set to optimize the trained automatic recognition model for abnormal states of the electrical cabinet, optimize the model parameters, and complete the training of the automatic recognition model for abnormal states of the electrical cabinet; and use the verification data set to record the current and temperature data of the power supply end of the main circuit, the control circuit, and the measuring circuit collected at the same time point within a first preset time after the target electrical cabinet is put into use. The current and temperature data are recorded as the second current and temperature data set, and multiple second current and temperature data sets collected at multiple time points are obtained according to a preset time interval, and the multiple second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, and the recognition result is output, so as to judge whether the electrical cabinet is in an abnormal state, and determine whether to output an alarm signal about the abnormal state of the electrical cabinet according to the judgment result; if so, the on-site staff conducts an on-site inspection of the electrical cabinet to confirm whether the electrical cabinet is in an abnormal state, and uses the cloud server to record and transmit the on-site inspection result data, obtain the on-site inspection result data, and determine whether to combine the multiple second current and temperature data sets as a new training data set according to the on-site inspection result data, retrain the electrical cabinet abnormal state automatic recognition model, and then automatically optimize the electrical cabinet abnormal state automatic recognition model.

[0106] It should be noted here that the present invention utilizes current sensors and thermistors respectively connected to the power supply ends of the main circuit, control circuit, and measuring circuit to measure the current and temperature of the power supply ends of the main circuit, control circuit, and measuring circuit, and then trains an automatic recognition model for abnormal conditions of the electrical cabinet, and then determines whether the electrical cabinet is in an abnormal state. If so, the on-site staff conducts an on-site inspection of the electrical cabinet and uploads the inspection results. The automatic recognition model for abnormal conditions of the electrical cabinet can be automatically retrained based on the inspection results, and then the automatic recognition model for abnormal conditions of the electrical cabinet is automatically optimized, thereby realizing an overall evaluation and judgment of the overall operating data of the electrical cabinet over a period of time, and more accurately judging whether the overall circuit operation data of the electrical cabinet over a period of time is in an abnormal state, thereby greatly improving the safety, reliability, and intelligence of the circuit protection and fire prevention monitoring of the electrical cabinet, greatly reducing the safety hazard of fire in the electrical cabinet, and solving the problem that the setting of a single or multiple thresholds cannot accurately judge whether the electrical cabinet is in an abnormal state, thereby greatly improving the intelligence and usability of the present invention, and greatly expanding the application scenarios of the present invention.

[0107] In a preferred embodiment, the present application further provides an electronic device, the electronic device comprising:

[0108] A memory; and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the electrical cabinet circuit protection fire monitoring method is implemented. The computer device can be broadly a server, a terminal, or any other electronic device with necessary computing and / or processing capabilities. In one embodiment, the computer device may include a processor, a memory, a network interface, a communication interface, etc. connected via a system bus. The processor of the computer device can be used to provide necessary computing, processing and / or control capabilities. The memory of the computer device may include a non-volatile storage medium and an internal memory. An operating system, a computer program, etc. may be stored in or on the non-volatile storage medium. The internal memory can provide an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface and the communication interface of the computer device can be used to connect and communicate with external devices via a network. When the computer program is executed by the processor, the steps of the method of the present invention are performed.

[0109] The present invention may be implemented as a computer-readable storage medium having a computer program stored thereon, which causes the steps of the method of an embodiment of the present invention to be executed when executed by a processor. In one embodiment, the computer program is distributed on a plurality of computer devices or processors coupled to a network so that the computer program is stored, accessed and executed in a distributed manner by one or more computer devices or processors. A single method step / operation, or two or more method steps / operations, may be performed by a single computer device or processor or by two or more computer devices or processors. One or more method steps / operations may be performed by one or more computer devices or processors, and one or more other method steps / operations may be performed by one or more other computer devices or processors. One or more computer devices or processors may perform a single method step / operation, or perform two or more method steps / operations.

[0110] It will be appreciated by a person skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as a computer device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present invention to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, database, or other media herein may include non-volatile and / or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0111] It can be understood that the present invention utilizes the current sensors and thermistors respectively connected to the power supply ends of the main circuit, control circuit, and measuring circuit to measure the current and temperature of the power supply ends of the main circuit, control circuit, and measuring circuit, and then trains the automatic recognition model of the abnormal state of the electrical cabinet, and then determines whether the electrical cabinet is in an abnormal state. If so, the on-site staff conducts an on-site inspection of the electrical cabinet and uploads the inspection results. The automatic recognition model of the abnormal state of the electrical cabinet can be automatically retrained according to the inspection results, and then the automatic recognition model of the abnormal state of the electrical cabinet is automatically optimized, realizing an overall evaluation and judgment of the overall operation data of the electrical cabinet over a period of time, and more accurately judging whether the overall circuit operation data of the electrical cabinet over a period of time is in an abnormal state, which greatly improves the safety, reliability and intelligence of the circuit protection and fire prevention monitoring of the electrical cabinet, greatly reduces the safety hazard of fire in the electrical cabinet, and solves the problem that the setting of a single or multiple thresholds cannot accurately judge whether the electrical cabinet is in an abnormal state, which greatly improves the intelligence and usability of the present invention and greatly expands the application scenarios of the present invention.

[0112] The various technical features described above can be combined arbitrarily. Although all possible combinations of these technical features are not described, any combination of these technical features should be considered to be covered by this specification as long as there is no contradiction in such combination.

[0113] The specific implementation of the present invention described above does not constitute a limitation on the protection scope of the present invention. Any other corresponding changes and modifications made based on the technical concept of the present invention should be included in the protection scope of the claims of the present invention.

Claims

1. A method for monitoring fire protection of circuit protection of an electrical cabinet, characterized in that: The electrical cabinet comprises a cabinet body, a main circuit, a control circuit, and a measuring circuit are arranged in the cabinet body, the main circuit, the control circuit, and the measuring circuit all comprise power supply ends, the power supply ends of the main circuit, the control circuit, and the measuring circuit are respectively connected to current sensors and thermistors for measuring current and temperature of the power supply ends of the main circuit, the control circuit, and the measuring circuit, and the method comprises: S100, after the electrical cabinet is installed and before it is put into use, within a first preset time, the current and temperature data of the power supply end of the main circuit, the control circuit, and the measuring circuit collected at the same time point are recorded as a first current and temperature data set, and multiple first current and temperature data sets collected at multiple time points are obtained at preset time intervals, and the multiple first current and temperature data sets are respectively divided into a training data set and a verification data set according to a first preset ratio, and the training data set is used to train the automatic recognition model of the abnormal state of the electrical cabinet; the verification data set is used to optimize the trained automatic recognition model of the abnormal state of the electrical cabinet, optimize the model parameters, and complete the training of the automatic recognition model of the abnormal state of the electrical cabinet; S200, after the target electrical cabinet is put into use, within a first preset time, the current and temperature data of the power supply end of the main circuit, the control circuit, and the measuring circuit collected at the same time point are recorded as a second current and temperature data set, and multiple second current and temperature data sets collected at multiple time points are obtained at preset time intervals, and the multiple second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, and the recognition results are output, and then it is determined whether the electrical cabinet is in an abnormal state, and whether to output an alarm signal related to the abnormal state of the electrical cabinet according to the judgment result; S300, if yes, the on-site staff conducts an on-site inspection of the electrical cabinet to confirm whether the electrical cabinet is in an abnormal state, and uses the cloud server to record and transmit the on-site inspection result data, obtain the on-site inspection result data, and determine whether to combine the multiple second current and temperature data sets as a new training data set based on the on-site inspection result data, retrain the electrical cabinet abnormal state automatic recognition model, and then automatically optimize the electrical cabinet abnormal state automatic recognition model; The method of using the training data set to train the automatic recognition model for abnormal state of the electrical cabinet; and using the verification data set to optimize the trained automatic recognition model for abnormal state of the electrical cabinet, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal state of the electrical cabinet, includes: The training data set is input into the preset network layer in batches for training, and the preset network layer includes a Transformer network layer, and the Transformer network layer is used to forward propagate the current and temperature data sets of the power supply end of the main circuit, control circuit, and measurement circuit collected at the current time point to predict the current and temperature data sets of the power supply end of the main circuit, control circuit, and measurement circuit at the next time point, and then obtain the predicted loss value; the loss value of the preset network layer is calculated and input into the optimizer for optimization, and the direction in which the parameter gradient of the automatic recognition model of the abnormal state of the electrical cabinet is fastest to decrease; the automatic recognition model of the abnormal state of the electrical cabinet is back-propagated according to the loss value and the parameter gradient of the model to optimize the parameters of the automatic recognition model of the abnormal state of the electrical cabinet; Specifically, the preset network layer further includes a ResNet network layer, and the ResNet network layer is used to classify and predict the output result according to the collected current and temperature data set, and the classification result includes whether the electrical cabinet is in an abnormal state or whether the electrical cabinet does not have an abnormal state; The ResNet network layer is used to classify and predict the output results through the current and temperature data of the power supply end of the main circuit, control circuit, and measurement circuit collected in the training data set and the verification data set. The specific steps are as follows: (1) First, put the training data set into the Backbones network layer of the model. Backbone refers to a series of convolutional layers that constitute the skeleton of the neural network. The main function of these layers is to extract the features of the input data. The Backbone network is usually composed of multiple convolutional layers, pooling layers, and activation functions. It can extract meaningful feature representations from the original data. Backbones is composed of multiple Backbones, and the core network layer in Backbone is the Resnet network layer. After the data set is input into the model, the first current and data set will be input into Backbones for training; (2) When constructing Backbone, circuit name encoding and Resnet network layers are added. After the training data set is input, Resnet starts training layer by layer to extract features from the current and temperature data sets, assisting Transformer in predicting whether the electrical cabinet is in an abnormal state. The extracted features are used with the help of circuit name encoding to obtain the circuit name information described in the current and temperature data. The circuit name information includes the main circuit, control circuit, and measurement circuit. Specifically, the main circuit includes: a switch, a fuse, a contactor, and a relay; Specifically, the control circuit includes: a switch, a button, and an MCU control chip; The measuring circuit comprises: a voltmeter, an ammeter and an electric energy meter.

2. The electrical cabinet circuit protection fire prevention monitoring method according to claim 1, characterized in that: The method further comprises: After the electrical cabinet is installed and before it is put into use, it is necessary to test the installed electrical cabinet to ensure that there are no abnormalities in the current and temperature data at the power supply end of the main circuit, control circuit, and measurement circuit.

3. The method for monitoring fire prevention of circuit protection of an electrical cabinet according to claim 1, characterized in that: The method of using the training data set to train the automatic recognition model for abnormal state of the electrical cabinet; using the verification data set to optimize the trained automatic recognition model for abnormal state of the electrical cabinet, optimizing the model parameters, and completing the training of the automatic recognition model for abnormal state of the electrical cabinet, further includes: After each training, the verification data set is input into the preset network layer of the previous training in batches for model parameter verification, and cyclic training is performed, with the total number of training rounds set to the first preset round; the loss value of the preset network layer is recorded, and it is determined whether the loss value meets the first preset condition, and whether to end the training and output the model parameters based on the judgment result.

4. The method for monitoring fire prevention of circuit protection of an electrical cabinet according to claim 3, characterized in that: The determining whether the loss value satisfies the first preset condition, and determining whether to end the training and output the model parameters according to the determination result, includes: If the loss value meets the first preset condition, the training is terminated and the parameters of the current electrical cabinet abnormal state automatic identification model are output; if the loss value does not meet the first preset condition, the training is continued.

5. The electrical cabinet circuit protection fire prevention monitoring method according to claim 4, characterized in that: The first preset condition is: After the second preset round of training, the loss values ​​obtained in the next round of training are greater than or equal to the loss values ​​that occurred during the second preset round of training.

6. The method for monitoring fire prevention of circuit protection of an electrical cabinet according to claim 1, characterized in that: The recognition result is that the electrical cabinet is in an abnormal state or the electrical cabinet is not in an abnormal state, the plurality of second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, the recognition result is output, and then it is determined whether the electrical cabinet is in an abnormal state, and whether to output an alarm signal related to the abnormal state of the electrical cabinet according to the determination result, including: If the output result is that the electrical cabinet is in an abnormal state, it is determined that an alarm signal related to the abnormal state of the electrical cabinet is output; If the output result is that there is no abnormal state in the electrical cabinet, it is determined that no alarm signal regarding the abnormality of the electrical cabinet is output.

7. The method for monitoring fire prevention of circuit protection of an electrical cabinet according to claim 1, characterized in that: The field inspection result data includes whether an abnormal state actually exists or does not actually exist. The determining, based on the field inspection result data, whether to combine the plurality of second current and temperature data sets as a newly added training data set, retrain the automatic recognition model for abnormal state of the electrical cabinet, and then automatically optimize the automatic recognition model for abnormal state of the electrical cabinet includes: If the field inspection result data indicates that an abnormal state actually exists, the multiple second current and temperature data sets are combined as a new training data set, and the automatic recognition model for abnormal states of electrical cabinets is retrained, thereby automatically optimizing the automatic recognition model for abnormal states of electrical cabinets.

8. The method for monitoring fire prevention of circuit protection of an electrical cabinet according to claim 1, characterized in that: The field inspection result data includes whether an abnormal state actually exists or does not actually exist, and the method of determining whether to combine the plurality of second current and temperature data sets as a newly added training data set, retraining the automatic recognition model for abnormal state of the electrical cabinet, and then automatically optimizing the automatic recognition model for abnormal state of the electrical cabinet according to the field inspection result data also includes: If the field inspection result data shows that there is actually no abnormal state, the multiple second current and temperature data sets are not combined as a new training data set, the electrical cabinet abnormal state automatic recognition model is not retrained, and the electrical cabinet abnormal state automatic recognition model is not automatically optimized.

9. An electrical cabinet circuit protection fire monitoring system, characterized in that: The electrical cabinet circuit protection fire monitoring method according to any one of claims 1 to 8 is adopted, wherein the electrical cabinet comprises a cabinet body, wherein a main circuit, a control circuit, and a measuring circuit are arranged in the cabinet body, wherein the main circuit, the control circuit, and the measuring circuit all comprise a power supply end, and the system comprises: An acquisition module, the acquisition module includes a current sensor and a thermistor respectively connected to the power supply ends of the main circuit, the control circuit, and the measurement circuit, and is used to measure the current and temperature of the power supply ends of the main circuit, the control circuit, and the measurement circuit; A control module is used to record the current and temperature data of the power supply end of the main circuit, control circuit, and measurement circuit collected at the same time point within a first preset time after the electrical cabinet is installed and before it is put into use as a first current and temperature data set, obtain multiple first current and temperature data sets collected at multiple time points according to preset time intervals, divide the multiple first current and temperature data sets into training data sets and verification data sets according to first preset ratios, use the training data sets to train the automatic recognition model for abnormal states of the electrical cabinet; use the verification data sets to optimize the trained automatic recognition model for abnormal states of the electrical cabinet, optimize the model parameters, and complete the training of the automatic recognition model for abnormal states of the electrical cabinet; and use the verification data sets to record the current and temperature data of the power supply end of the main circuit, control circuit, and measurement circuit collected at the same time point within a first preset time after the target electrical cabinet is put into use. The current and temperature data are recorded as a second current and temperature data set. Multiple second current and temperature data sets collected at multiple time points are obtained according to a preset time interval. The multiple second current and temperature data sets are input into the electrical cabinet abnormal state automatic recognition model, and the recognition result is output to determine whether the electrical cabinet is in an abnormal state. According to the judgment result, it is determined whether to output an alarm signal regarding the abnormal state of the electrical cabinet; if so, the on-site staff conducts an on-site inspection of the electrical cabinet to confirm whether the electrical cabinet is in an abnormal state, and uses a cloud server to record and transmit the on-site inspection result data, obtain the on-site inspection result data, and determine whether to combine the multiple second current and temperature data sets as a new training data set based on the on-site inspection result data, retrain the electrical cabinet abnormal state automatic recognition model, and then automatically optimize the electrical cabinet abnormal state automatic recognition model.

Images