Method and system for diagnosing waterproof performance of low-voltage power distribution cabinet

Abstract

The invention provides a method and system for diagnosing the waterproof performance of a low-voltage power distribution cabinet, and the method comprises the steps: collecting a plurality of micro-vibration signals corresponding to the low-voltage power distribution cabinet during operation; enhancement processing is carried out on the collected micro-vibration signals, corresponding characteristic parameters are extracted, and graph structure data are formed by the micro-vibration signals and the corresponding characteristic parameters; inputting the graph structure data into a pre-trained deep learning model to estimate the abnormal probability of the low-voltage power distribution cabinet to obtain a corresponding abnormal probability value; and if the abnormal probability value is greater than a preset abnormal threshold value, judging that the waterproof performance of the low-voltage power distribution cabinet is abnormal. According to the invention, the micro-vibration signals during the operation of the power distribution cabinet are collected in real time, and the graph neural network is utilized to carry out fusion analysis on the data of the multiple sensor nodes, so that the real-time monitoring and accurate diagnosis of the waterproof performance of the power distribution cabinet are realized, and the cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power equipment monitoring, and particularly to a method and system for diagnosing the waterproof performance of a low-voltage power distribution cabinet. Background Art

[0002] A low-voltage power distribution cabinet is a key power transmission and distribution device in the power system, and is widely used in industrial, commercial, and residential places. Its operating environment is complex and it is often exposed to humid, rainy, or corrosive gas environments. The waterproof performance of the power distribution cabinet directly affects its operating reliability and safety. Once the waterproof performance deteriorates, it may cause the internal part of the equipment to be affected by moisture, short circuit, or even serious accidents such as fires.

[0003] Traditional waterproof performance detection mainly relies on manual regular inspections, and inspectors judge the waterproof condition of the power distribution cabinet through visual observation, manual measurement, etc. This method is highly subjective, has low accuracy, low efficiency, and a long inspection cycle, and cannot detect the deterioration of the waterproof performance in time, which may miss the best maintenance opportunity and lead to potential safety hazards.

[0004] Existing monitoring systems usually use environmental parameter sensors such as humidity and temperature, and cannot comprehensively reflect the waterproof performance of the power distribution cabinet. These sensors can only detect changes in the internal environment and cannot directly sense the deterioration process of the waterproof performance, such as seal aging, tiny structural cracks, etc. In addition, humidity sensors may become saturated in high-humidity environments, resulting in monitoring failure. Due to the lack of in-depth analysis of micro-vibration signals, it is difficult for existing technologies to achieve real-time and accurate monitoring of the waterproof performance of the power distribution cabinet. The deterioration of the waterproof performance usually causes tiny deformations and vibration changes in the structure of the power distribution cabinet, and this information cannot be captured by traditional environmental parameter sensors.

[0005] In addition, most existing monitoring systems use simple threshold judgment or experience-based rules, lack advanced intelligent algorithms, resulting in low fault diagnosis accuracy, and cannot predict the deterioration trend of the waterproof performance, affecting the preventive maintenance of the equipment. In the case of multi-sensor deployment, existing methods are difficult to effectively fuse the data of each sensor, cannot make full use of the correlation and complementarity between sensors, resulting in low information utilization rate and affecting the accuracy of fault diagnosis. Due to the limitations of monitoring means, the deterioration of the waterproof performance cannot be detected in time, which may lead to serious problems such as water ingress inside the power distribution cabinet, insulation degradation, and damage to electrical equipment, and even safety accidents such as fires and power outages, affecting the reliable operation of the power system.

[0006] In summary, existing technologies have problems such as poor real-time performance, low accuracy, and insufficient intelligence in waterproof performance monitoring, and there is an urgent need for a new method that can monitor the waterproof performance of the power distribution cabinet in real time and accurately, and has high intelligent diagnosis and prediction capabilities. Summary of the Invention

[0007] The purpose of the present invention is to provide a method and system for diagnosing the waterproof performance of a low-voltage power distribution cabinet, so as to solve the technical problem of how to achieve self-diagnosis of the waterproof performance of the power distribution cabinet.

[0008] On the one hand, a method for diagnosing the waterproof performance of a low-voltage distribution cabinet is provided, comprising:

[0009] Collect multiple micro-vibration signals corresponding to low-voltage distribution cabinets during operation;

[0010] Perform enhancement processing on the collected micro-vibration signal and extract the corresponding characteristic parameters, so that the micro-vibration signal and the corresponding characteristic parameters form graph structure data;

[0011] The graph structure data is input into the pre-trained deep learning model to estimate the abnormal probability of the low-voltage distribution cabinet and obtain the corresponding abnormal probability value; if the abnormal probability value is greater than the preset abnormal threshold, the waterproof performance of the low-voltage distribution cabinet is judged to be abnormal.

[0012] Preferably, the multiple micro-vibration signals during operation include at least the micro-vibration signal on the inside of the cabinet door of the distribution cabinet, the micro-vibration signal in the middle of the left panel, the micro-vibration signal in the middle of the right panel, the micro-vibration signal in the top center and the micro-vibration signal of the bottom base.

[0013] Preferably, the enhancement processing of the collected micro-vibration signal includes amplifying the micro-vibration signal by a preset low-noise operational amplifier; filtering and denoising the amplified micro-vibration signal by a preset filter.

[0014] Preferably, the characteristic parameters include at least time domain characteristics, frequency domain characteristics and time-frequency domain characteristics; wherein the time domain characteristics include mean, standard deviation, kurtosis, skewness, root mean square, and kurtosis; the frequency domain characteristics include main frequency, spectral energy, frequency center, and frequency variance; the time-frequency domain characteristics include time-frequency moments, energy entropy, and singular value decomposition characteristics obtained using short-time Fourier transform.

[0015] Preferably, the deep learning model includes at least a three-layer graph convolutional network, the input is the node and the feature matrix, and the output is the state probability of the waterproof performance of the distribution cabinet; wherein the parameter setting of the deep learning model includes at least the input feature dimension, the number of hidden neurons in the first layer, the number of hidden neurons in the second layer and the number of output categories.

[0016] Preferably, the deep learning model is specifically used to construct the relationship between the sensor nodes and nodes used to collect micro-vibration signals into a weighted undirected graph G = (V, E), where V = {v 1 ,v 2 ,v 3 ,v 4, v 5} represents a sensor node, and E represents the edge between nodes.

[0017] Preferably, it further includes determining the weight of the edge between nodes according to the following formula

[0018]

[0019] where e ij represents the weight of the edge between sensor i and sensor j, represents the physical distance weight between sensor i and sensor j, represents the signal correlation weight between sensor i and sensor j, i represents the sensor serial number, j represents the sensor serial number, D ij represents the physical distance between sensor i and sensor j, ρ ij represents the signal cross-correlation coefficient between sensor i and sensor j.

[0020] Preferably, it further includes generating and outputting a corresponding alarm message when it is determined that the waterproof performance of the low-voltage power distribution cabinet is abnormal, where the alarm message at least includes the faulty equipment number and the time when the abnormality is detected, as well as the power distribution cabinet number, the type of abnormality, the possible cause, and the recommended treatment measures.

[0021] On the other hand, a diagnostic system for the waterproof performance of a low-voltage power distribution cabinet is also provided, which is used to implement the diagnostic method for the waterproof performance of the low-voltage power distribution cabinet, including

[0022] a signal acquisition module, which is used to acquire a plurality of micro-vibration signals during the operation of the corresponding low-voltage power distribution cabinet;

[0023] a signal processing module, which is used to perform enhancement processing on the acquired micro-vibration signals and extract the corresponding characteristic parameters, and the micro-vibration signals and the corresponding characteristic parameters form graph structure data;

[0024] a fault diagnosis module, which is used to input the graph structure data into a pre-trained deep learning model to estimate the abnormality probability of the low-voltage power distribution cabinet, and obtain the corresponding abnormality probability value; if the abnormality probability value is greater than the preset abnormality threshold, it is determined that the waterproof performance of the low-voltage power distribution cabinet is abnormal.

[0025] Preferably, it further includes an alarm module, which is used to generate and output a corresponding alarm message when it is determined that the waterproof performance of the low-voltage power distribution cabinet is abnormal, where the alarm message at least includes the faulty equipment number and the time when the abnormality is detected, as well as the power distribution cabinet number, the type of abnormality, the possible cause, and the recommended treatment measures.

[0026] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0027] The diagnostic method and system for the waterproof performance of a low-voltage power distribution cabinet provided by the present invention can monitor the micro-vibration changes of the power distribution cabinet in real time, detect the deterioration of the waterproof performance in a timely manner, and avoid major safety hazards caused by detection lag. By using a graph neural network to process multi-sensor data, it can effectively fuse the signals of sensors at different positions, analyze the correlations between them, and significantly improve the accuracy and precision of fault diagnosis. Especially in a complex environment, it can still maintain high monitoring performance. It can automatically learn and identify the changing trend of the waterproof performance, not only diagnose existing faults, but also predict possible future problems, facilitating the adoption of preventive maintenance measures, thereby reducing the equipment maintenance cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, obtaining other drawings based on these drawings still belongs to the scope of the present invention.

[0029] Figure 1 It is a schematic diagram of the main process of a diagnostic method for the waterproof performance of a low-voltage power distribution cabinet in an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0030] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings.

[0031] As Figure 1 shown, it is a schematic diagram of an embodiment of a diagnostic method for the waterproof performance of a low-voltage power distribution cabinet provided by the present invention. In this embodiment, the method includes the following steps:

[0032] Step S1, collect multiple micro-vibration signals during the operation of the corresponding low-voltage power distribution cabinet; it can be understood that piezoelectric ceramic vibration sensors are installed at key parts of the low-voltage power distribution cabinet to capture the micro-vibration signals during operation. One piezoelectric ceramic vibration sensor of model PZT-5H is installed at the inner side of the cabinet door (at a height of 1.2 meters), the middle of the left side panel (at a height of 0.8 meters), the middle of the right side panel (at a height of 0.8 meters), the center of the top (at a height of 1.8 meters), and the bottom base (at a height of 0 meters) of the power distribution cabinet. The sensors are fixed on the surface of the power distribution cabinet through high-strength adhesives or bolts to ensure good coupling between the sensors and the power distribution cabinet, so as to improve the accuracy of signal collection.

[0033] In one embodiment, the multiple micro-vibration signals during operation include at least a micro-vibration signal on the inside of a cabinet door of the power distribution cabinet, a micro-vibration signal in the middle of a left side panel, a micro-vibration signal in the middle of a right side panel, a micro-vibration signal in the top center, and a micro-vibration signal at the bottom base.

[0034] Step S2, enhance the collected micro-vibration signal and extract the corresponding characteristic parameters, and form a graph structure data with the micro-vibration signal and the corresponding characteristic parameters; wherein the characteristic parameters at least include time domain characteristics, frequency domain characteristics and time-frequency domain characteristics; wherein the time domain characteristics include mean, standard deviation, kurtosis, skewness, root mean square, kurtosis; the frequency domain characteristics include main frequency, spectrum energy, frequency center, frequency variance; the time-frequency domain characteristics include time-frequency moment, energy entropy and singular value decomposition characteristics obtained by short-time Fourier transform. It can be understood that the collected vibration signal needs to go through a series of preprocessing steps to extract effective features for subsequent analysis.

[0035] In one embodiment, the enhanced processing of the collected micro-vibration signal includes amplifying the micro-vibration signal through a preset low-noise operational amplifier; filtering and denoising the amplified micro-vibration signal through a preset filter. It is understandable that since the vibration signal is weak, a low-noise operational amplifier is used to construct a preamplifier to perform primary amplification of the signal. The gain is set to 1000 times, and the bandwidth range is 10Hz to 5kHz to ensure that the signal is not distorted during the amplification process. The amplified signal is transmitted to the data acquisition module via a shielded twisted pair cable to reduce electromagnetic interference.

[0036] Specifically, first, the signal is filtered using a fourth-order Butterworth bandpass filter with a cutoff frequency set to 50 Hz to 5 kHz to filter out low-frequency environmental noise and high-frequency electromagnetic interference. The transfer function of the filter is:

[0037]

[0038] Among them, ω c is the cut-off frequency, and n=4 is the filter order.

[0039] Next, the signal is denoised using the wavelet denoising method. The Daubechies 8 (db8) wavelet basis is selected, a 6-layer decomposition is performed, and a soft threshold is performed on the high-frequency coefficients. The threshold λ is calculated according to the VisuShrink method:

[0040]

[0041] Where σ is the noise standard deviation and N is the signal length.

[0042] After preprocessing, features are extracted from the vibration signal of each sensor, including time domain features, frequency domain features, and time-frequency domain features, which are used to fully reflect the vibration signal characteristics of the distribution cabinet under different waterproof performance states and provide reliable data support for fault diagnosis. Among them, the time domain features include mean, standard deviation, kurtosis, skewness, root mean square, and kurtosis; the frequency domain features include main frequency, spectrum energy, frequency center, and frequency variance; the time-frequency domain features use the time-frequency moment, energy entropy, and singular value decomposition features obtained by short-time Fourier transform.

[0043] Step S3, input the graph structure data into the pre-trained deep learning model to estimate the abnormal probability of the low-voltage distribution cabinet and obtain the corresponding abnormal probability value; if the abnormal probability value is greater than the preset abnormal threshold, the waterproof performance of the low-voltage distribution cabinet is determined to be abnormal. It can be understood that after the model training is completed and verified, the vibration signal collected in real time is preprocessed and feature extracted to construct the graph structure data and input the trained GCN model. The model outputs the probability value of the waterproof performance of the distribution cabinet being "normal" or "abnormal". When the abnormal probability predicted by the model exceeds the set threshold of 0.8, the system determines that the waterproof performance of the distribution cabinet is abnormal.

[0044] In one embodiment, the deep learning model includes at least a three-layer graph convolutional network, the input is the node and the feature matrix, and the output is the state probability of the waterproof performance of the distribution cabinet; wherein the parameter setting of the deep learning model includes at least the input feature dimension, the number of hidden neurons in the first layer, the number of hidden neurons in the second layer and the number of output categories.

[0045] It should be noted that the model adopts a three-layer GCN structure, the input is the node feature matrix X, and the output is the state probability of the waterproof performance of the distribution cabinet. The specific parameters of the model are set as the input feature dimension F = 20, the number of hidden neurons in the first layer H 1 =64, the number of hidden neurons in the second layer is H 2 =32, the number of output categories C = 2.

[0046] The loss function of the model adopts the cross entropy loss function plus the L2 regularization term. The optimization algorithm adopts the Adam optimizer. The initial learning rate is α=0.001, and the learning rate decay strategy is adopted. The learning rate is multiplied by 0.9 every 10 training rounds.

[0047] During the model training process, we first construct a data set, including data in normal and abnormal states. For normal state data, vibration signals are collected continuously for 180 days under the condition of good waterproof performance, totaling 4,320 hours. Abnormal state data is obtained by simulating the situation of degraded waterproof performance, including gap leakage, seal aging, surface damage, etc., with a total of about 1,440 hours of data collected. The data set is divided into training set, validation set, and test set in a ratio of 7:2:1.

[0048] During the training process, the model is based on the training set data and continuously adjusts the model parameters through forward propagation and backward propagation to minimize the loss function. After each training epoch, the performance of the model is evaluated on the validation set, and the hyperparameters of the model are adjusted, including the hidden layer dimension, learning rate, regularization coefficient, etc. Finally, metrics such as accuracy, precision, recall, F1-score, and AUC of the model are evaluated on the test set to verify the generalization ability of the model.

[0049] In one embodiment, the deep learning model is specifically used to construct a weighted undirected graph G=(V, E) for the sensor nodes used to collect micro-vibration signals and the relationships between the nodes, where V = {v 1 , v 2 , v 3 , v 4 , v 5} represents the sensor nodes, and E represents the edges between the nodes. The weights of the edges between the nodes are determined according to the following formula:

[0050]

[0051] where e ij represents the weight of the edge between sensor i and sensor j, represents the physical distance weight between sensor i and sensor j, represents the signal correlation weight between sensor i and sensor j, i represents the sensor number, j represents the sensor number, D ij represents the physical distance between sensor i and sensor j, and ρ ij represents the signal cross-correlation coefficient between sensor i and sensor j.

[0052] That is, a deep learning model based on a graph convolutional network (GCN) is constructed to make full use of the data of multiple sensor nodes.

[0053] First, the sensor nodes and the relationships between them are constructed into a weighted undirected graph G=(V, E), where V = {v 1 , v 2 , v 3 , v 4 , v 5} represents the sensor nodes, and E represents the edges between the nodes.

[0054] The weight e ij of the edge is jointly determined by the physical distance weight and the signal correlation weight :

[0055]

[0056] Among them, the physical distance weight is based on the physical distance D between sensors i and j ij Calculate as follows:

[0057]

[0058] The signal correlation weight is obtained by calculating the cross-correlation coefficient ρ of the signals of sensors i and j ij , and after taking the absolute value, it is normalized:

[0059]

[0060] An embodiment further includes generating and outputting corresponding alarm information when it is determined that the waterproof performance of the low-voltage power distribution cabinet is abnormal. The alarm information at least includes the faulty equipment number and the time when the abnormality is detected, as well as the power distribution cabinet number, the type of abnormality, the possible cause, and the recommended treatment measures. The waterproof performance failure immediately triggers the alarm mechanism, and the system will pop up the alarm information, displaying the specific faulty equipment number and the time when the abnormality is detected. At the same time, alarm notifications are sent to the maintenance personnel via text message and email, including the power distribution cabinet number, the type of abnormality, the possible cause, and the recommended treatment measures. The alarm information and related data will also be recorded in the log system for subsequent analysis and traceability.

[0061] In addition, the system also provides automated treatment suggestions. According to different types of abnormalities, the system will give corresponding treatment plans, such as immediately dispatching maintenance personnel for inspection, replacing seals, strengthening waterproof measures, etc. By analyzing the accumulated abnormal data, the system can identify common problems and provide a basis for improving the design and maintenance of the power distribution cabinet.

[0062] An embodiment of the present invention further provides a diagnostic system for the waterproof performance of a low-voltage power distribution cabinet, used to implement the diagnostic method for the waterproof performance of the low-voltage power distribution cabinet, including

[0063] A signal acquisition module for acquiring a plurality of micro-vibration signals during the operation of the corresponding low-voltage power distribution cabinet;

[0064] A signal processing module for enhancing the acquired micro-vibration signals and extracting corresponding characteristic parameters, and forming graph structure data from the micro-vibration signals and the corresponding characteristic parameters;

[0065] A fault diagnosis module for inputting the graph structure data into a pre-trained deep learning model to estimate the abnormal probability of the low-voltage power distribution cabinet, obtaining a corresponding abnormal probability value; if the abnormal probability value is greater than a preset abnormal threshold, it is determined that the waterproof performance of the low-voltage power distribution cabinet is abnormal.

[0066] A specific embodiment further includes an alarm module, which is configured to generate and output corresponding alarm information when it is determined that the waterproof performance of the low-voltage power distribution cabinet is abnormal. The alarm information at least includes the fault equipment number, the time when the abnormality is detected, as well as the power distribution cabinet number, the type of abnormality, the possible cause, and the recommended treatment measures.

[0067] It should be noted that the system described in the above embodiment corresponds to the method described in the above embodiment. Therefore, the parts not described in detail in the system described in the above embodiment can be obtained by referring to the content of the method described in the above embodiment, and will not be elaborated here.

[0068] In summary, implementing the embodiments of the present invention has the following beneficial effects:

[0069] The method and system for diagnosing the waterproof performance of the low-voltage power distribution cabinet provided by the present invention can monitor the micro-vibration changes of the power distribution cabinet in real time, timely detect the deterioration of the waterproof performance, and avoid major safety hazards caused by detection lag; by using the graph neural network to process multi-sensor data, the signals of sensors at different positions can be effectively fused, and the correlation between them can be analyzed, significantly improving the accuracy and precision of fault diagnosis, especially maintaining high monitoring performance in complex environments; it can automatically learn and identify the change trend of the waterproof performance, not only diagnose existing faults, but also predict possible future problems, facilitating preventive maintenance measures, thereby reducing the equipment maintenance cost.

[0070] The above-disclosed are only the preferred embodiments of the present invention. Of course, the scope of the rights of the present invention cannot be limited thereby. Therefore, equivalent changes made according to the claims of the present invention still fall within the scope covered by the present invention.

Claims

1. A method for diagnosing the waterproof performance of a low-voltage distribution cabinet, characterized in that: include: Collect multiple micro-vibration signals corresponding to low-voltage distribution cabinets during operation; Perform enhancement processing on the collected micro-vibration signal and extract the corresponding characteristic parameters, so that the micro-vibration signal and the corresponding characteristic parameters form graph structure data; The graph structure data is input into the pre-trained deep learning model to estimate the abnormal probability of the low-voltage distribution cabinet and obtain the corresponding abnormal probability value; if the abnormal probability value is greater than the preset abnormal threshold, the waterproof performance of the low-voltage distribution cabinet is judged to be abnormal.

2. The method according to claim 1, characterized in that The multiple micro-vibration signals during operation include at least a micro-vibration signal on the inside of the cabinet door of the distribution cabinet, a micro-vibration signal in the middle of the left side panel, a micro-vibration signal in the middle of the right side panel, a micro-vibration signal in the top center, and a micro-vibration signal at the bottom base.

3. The method according to claim 1, characterized in that The enhanced processing of the collected micro-vibration signal includes amplifying the micro-vibration signal by a preset low-noise operational amplifier; filtering and denoising the amplified micro-vibration signal by a preset filter.

4. The method according to claim 1, characterized in that The characteristic parameters include at least time domain characteristics, frequency domain characteristics and time-frequency domain characteristics; wherein the time domain characteristics include mean, standard deviation, kurtosis, skewness, root mean square, and kurtosis; the frequency domain characteristics include main frequency, spectrum energy, frequency center, and frequency variance; the time-frequency domain characteristics include time-frequency moments, energy entropy, and singular value decomposition characteristics obtained by short-time Fourier transform.

5. The method according to claim 1, characterized in that The deep learning model includes at least a three-layer graph convolutional network, the input is the node and the feature matrix, and the output is the state probability of the waterproof performance of the distribution cabinet; wherein the parameter setting of the deep learning model includes at least the input feature dimension, the number of hidden neurons in the first layer, the number of hidden neurons in the second layer and the number of output categories.

6. The method according to claim 5, characterized in that The deep learning model is specifically used to construct the relationship between the sensor nodes and nodes used to collect micro-vibration signals into a weighted undirected graph G=(V, E), Where V = {v1,v2,v3,v4,v 5} represents sensor nodes, and E represents the edges between nodes.

7. The method according to claim 6, characterized in that It also includes determining the weight of the edge between nodes according to the following formula: Among them, e ij represents the weight of the edge between sensor i and sensor j, represents the physical distance weight between sensor i and sensor j, represents the correlation weight of the signals of sensor i and sensor j, i represents the sensor number, j represents the sensor number, D ij represents the physical distance between sensor i and sensor j, ρ ij It represents the mutual correlation coefficient between the signals of sensor i and sensor j.

8. The method according to claim 1, characterized in that It also includes, when it is determined that the waterproof performance of the low-voltage distribution cabinet is abnormal, generating and outputting corresponding alarm information, wherein the alarm information at least includes the faulty equipment number and the time when the abnormality is detected, as well as the distribution cabinet number, abnormality type, possible causes and recommended handling measures.

9. A diagnostic system for waterproof performance of a low-voltage power distribution cabinet, used to implement the method according to any one of claims 1 to 8, characterized in that: include, A signal acquisition module is used to collect micro-vibration signals corresponding to multiple low-voltage power distribution cabinets during operation; A signal processing module is used to enhance the collected micro-vibration signal and extract the corresponding characteristic parameters, so that the micro-vibration signal and the corresponding characteristic parameters form a graph structure data; The fault diagnosis module is used to input the graph structure data into the pre-trained deep learning model to estimate the abnormal probability of the low-voltage distribution cabinet and obtain the corresponding abnormal probability value; If the abnormal probability value is greater than the preset abnormal threshold, it is determined that the waterproof performance of the low-voltage distribution cabinet is abnormal.

10. The system according to claim 9, characterized in that It also includes an alarm module, which is used to generate and output corresponding alarm information when it is determined that the waterproof performance of the low-voltage distribution cabinet is abnormal, wherein the alarm information at least includes the faulty equipment number and the time when the abnormality is detected, as well as the distribution cabinet number, abnormality type, possible causes and recommended treatment measures.

Images