Electrical automation control cabinet and use method

Abstract

The invention belongs to the technical field of electrical automation, and discloses an electrical automation control cabinet and a use method thereof, and the method comprises the steps: placing an electrical part on a supporting mechanism; the heat absorption box and the S-shaped bent pipe absorb heat and cool the interior of the cabinet body; the data collection module collects real-time data and historical data; the image acquisition module acquires a high-definition image; the image preprocessing module carries out preprocessing; the feature extraction module performs feature extraction; the image analysis module identifies abnormal conditions; the comprehensive evaluation module performs comprehensive evaluation, and the risk prediction module predicts potential problem risks; the air inlet fan and the exhaust fan perform enhanced cooling treatment, and the oil pump improves the flowing speed of the heat-conducting oil and enhances the cooling and heat dissipation effects. Due to the sliding design of the supporting mechanism, an operator can quickly pull out the supporting plate, and the maintenance operation convenience is improved; passive heat dissipation and active heat dissipation cooperate to improve the heat dissipation and cooling effect; and intelligent assessment and risk early warning are carried out to predict potential risks such as aging trend and poor contact of the electrical parts in advance.

Description

Technical Field

[0001] This application relates to the field of electrical automation technology, and more specifically, to an electrical automation control cabinet and its usage method. Background Technology

[0002] A control cabinet is an assembly of switching equipment, measuring instruments, protective electrical devices, and auxiliary equipment in a closed or semi-closed metal cabinet or panel, arranged according to electrical wiring requirements. Its layout should meet the requirements for normal operation of the power system, facilitate maintenance, and not endanger personnel or surrounding equipment. The control cabinet is the "nerve center" of the power system and automation equipment; its design and application must consider safety, reliability, and maintainability. Through reasonable component selection, structured layout, and standardized construction, it can meet the needs of stable power system operation while providing convenience for subsequent maintenance and upgrades.

[0003] The prior art publication CN114389179A provides an electrical automation control cabinet, including a control cabinet body and an adjustment assembly disposed on the surface of the control cabinet body. The adjustment assembly includes a U-shaped bracket, a threaded post, a blocking ring, and casters. The U-shaped bracket is disposed on the surface of the control cabinet body, and a threaded hole is formed on the upper surface of the U-shaped bracket. The threaded post is screwed into the inner wall of the threaded hole. A circular hole is formed on the upper surface of the control cabinet body, and the threaded post is located inside the circular hole. This design facilitates the movement of the control cabinet. Once the control cabinet is moved to a suitable position, it helps to fix the position of the control cabinet, preventing the control cabinet from moving due to collisions. The airflow inside the control cabinet body helps to reduce the heat generated by the control cabinet body during operation, improving the heat dissipation performance of the control cabinet body.

[0004] While the existing technical solutions described above can achieve the relevant beneficial effects through their structure, they still have the following drawbacks: 1. Heat dissipation problem: Electrical automation control cabinets typically house a large number of electrical devices, such as frequency converters, PLCs, and relays, which generate heat during operation. Poor heat dissipation can lead to excessively high temperatures inside the cabinet, affecting the performance and lifespan of the equipment, and may even cause malfunctions. 2. Inability to predict potential problems and risks during the operation of electrical components within the control cabinet in a timely manner.

[0005] In view of this, we propose an electrical automation control cabinet and its usage method. Summary of the Invention

[0006] 1. Technical problems to be solved The purpose of this application is to provide an electrical automation control cabinet and its usage method, which solves the technical problems mentioned in the background art. It realizes that through the sliding design of the support mechanism, the operator can quickly pull out the support plate, improving the convenience of maintenance operations; it realizes the synergy of passive heat dissipation and active heat dissipation to improve the heat dissipation and cooling effect; it can form a dual active heat dissipation of "air convection + liquid heat conduction" to improve the cooling rate; and it realizes intelligent assessment and risk warning. Based on the time series model trained by historical data, it can predict the aging trend of electrical components, poor contact and other potential risks in advance.

[0007] 2. Technical Solution This application provides an electrical automation control cabinet, including: cabinet body, cabinet door, support mechanism, S-shaped bend, air intake fan, exhaust fan, oil supply assembly, heat absorption box and monitoring mechanism; The cabinet is equipped with rotatable doors; The cabinet is equipped with multiple horizontal and vertical partitions that are fixed vertically in a crisscross pattern. Each of the transverse partitions is equipped with a sliding support mechanism, which is used to place electrical components. Multiple horizontal and vertical partitions divide the cabinet's interior space into several electrical component storage areas; different electrical components can be placed separately, while avoiding cable clutter.

[0008] An intake fan is fixedly installed on the lower part of one side of the cabinet, and an exhaust fan is fixedly installed on the lower part of the other side of the cabinet. The intake and exhaust fans are used to cool the inside of the cabinet.

[0009] Several S-shaped bends are installed inside the cabinet; an oil supply component is fixedly installed on the outside of the cabinet, and the oil supply component is connected to each S-shaped bend. The heat transfer oil flows in the S-shaped bends through the oil supply component to absorb heat and cool the internal space of the cabinet.

[0010] Several heat-absorbing boxes are fixedly installed inside the cabinet, and the heat-absorbing boxes are filled with phase change wax; the heat-absorbing boxes absorb heat and cool down the inside of the cabinet.

[0011] The cabinet is equipped with a monitoring system that monitors the operation of the electrical components inside the cabinet, promptly detects any abnormalities, and issues an alarm.

[0012] Using the above technical solution, opening the cabinet door allows different electrical components to be placed and fixed on different support mechanisms. Under normal circumstances, the heat absorption box and S-shaped bend pipe absorb heat and cool the inside of the cabinet. The monitoring mechanism monitors the operation of the electrical components inside the cabinet, and when the temperature inside the cabinet reaches its maximum value, the intake and exhaust fans are used to enhance cooling.

[0013] As an optional embodiment of the present invention, the support mechanism includes a support plate, a trapezoidal rod, a handle, a support rod, and a slider; Two guide rails are fixedly installed in parallel on the partition of each electrical component storage area, and the guide rails are provided with trapezoidal grooves; two U-shaped slide rails are symmetrically fixed below the partition of each electrical component storage area. Two trapezoidal rods are fixedly installed parallel to each other below the support plate; the trapezoidal rods slide in conjunction with the corresponding guide rails; A handle is fixedly installed on the outer side of the trapezoidal rod; a support rod is fixedly installed at the lower end of the handle; a slider is fixedly installed at the rear end of the support rod; the slider is slidably mounted on a U-shaped slide rail.

[0014] The support plate is equipped with adjustable clamps made of high-strength plastic and covered with anti-slip rubber pads. The clamps are adjusted using screws and nuts, allowing operators to adjust the spacing of the clamps according to the size of the electrical components, thus ensuring a secure fixation for electrical components of different sizes.

[0015] Using the above technical solution, the electrical components are installed onto the support plate by pulling it outward with the handle; then the support plate and electrical components are pushed into the cabinet. The trapezoidal rod sliding with the corresponding guide rail and the slider slidably mounted on the U-shaped slide rail effectively support the support plate and electrical components, improving the stability of the support mechanism.

[0016] As an optional embodiment of the present invention, the oil supply assembly includes an oil storage tank, an oil pump, an oil supply pipe, and an oil return pipe; An oil storage tank is fixedly installed on the outside of the cabinet, and an oil pump is fixedly installed on the oil storage tank. The input end of the oil pump extends into the oil storage tank, and the output end of the oil pump is connected to the oil supply pipe. A return oil pipe is installed on the oil storage tank. After several S-shaped bends are connected end to end, both ends are connected to the oil supply pipe and the oil return pipe, respectively.

[0017] As an optional embodiment of the present invention, a copper wire mesh is fixedly installed inside the heat absorption box. The copper wire mesh is located at the bottom of the heat absorption box and is connected to an S-shaped bend via copper sheets. The phase change wax is filled to three-quarters of the volume of the heat absorption box, leaving sufficient margin.

[0018] The above technical solution uses phase change wax in the heat absorption box to absorb heat and cool the inside of the cabinet.

[0019] As an optional embodiment of the present invention, the monitoring mechanism includes: Data collection module: collects real-time and historical data of electrical components; including temperature and humidity sensors, current sensors, point pressure sensors and smoke sensors; Image acquisition module: includes multiple high-definition cameras and LED lights, to acquire high-definition images of the interior of cabinet 1 and electrical components; Image preprocessing module: preprocesses the acquired images, including noise reduction, grayscale conversion, normalization, and image enhancement; Feature extraction module: Extracts features from the preprocessed image, including color, texture, and shape; Image analysis module: Analyzes and identifies images after feature extraction, promptly detecting anomalies; Comprehensive evaluation module: Based on the image analysis module, combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, the module comprehensively evaluates the operation of electrical components and promptly detects abnormalities. Risk prediction module: Based on the image analysis module, combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, and referring to historical data, predict potential problems and risks in the operation of electrical components; Alarm module: includes an alarm that promptly issues an alert when abnormal situations or potential risks are detected; Control Center: A programmable PLC control module; network-connected to data collection module, image acquisition module, image preprocessing module, feature extraction module, image analysis module, comprehensive evaluation module, risk prediction module, and alarm module.

[0020] As an optional embodiment of the present invention, the image analysis module analyzes and identifies the image after feature extraction to promptly detect anomalies; including the following steps: 1. Model Training and Deployment: A Convolutional Neural Network (CNN) model was selected and trained using a large number of normal and abnormal image samples of electrical components. Hyperparameters such as the learning rate and number of iterations were adjusted to optimize the network structure and achieve optimal performance. After training, the validated CNN model was deployed to the computing device of the image analysis module to provide algorithmic support for subsequent image analysis. 2. Image Input: The feature extraction module transmits the processed image data to the image analysis module. The image data contains feature information such as color, texture, and shape, and enters the module in a specific data format to await analysis. 3. Feature Map Extraction: The input image enters the convolutional layer of the CNN model, and different convolutional kernels slide across the image to perform convolution operations, generating multiple feature maps. 4. Feature Dimensionality Reduction and Filtering: The feature maps extracted by the convolutional layers are then processed by the pooling layers. The pooling layers downsample the feature maps using methods such as max pooling or average pooling, reducing data dimensionality and computational cost, while retaining the most critical feature information in the image and filtering out representative features.

[0021] 5. Fully connected layer processing: The pooled feature data is flattened and input into the fully connected layer. The fully connected layer fuses all the features and maps the feature data to a specific output dimension through neuron calculation, outputting the probability values ​​of different states (normal, abnormal) of the corresponding electrical components.

[0022] 6. Abnormal Situation Judgment: The probability value output by the fully connected layer is compared with a preset threshold. If the probability value of a certain type of abnormal situation exceeds the preset threshold, the electrical component is determined to have that type of abnormality, such as component damage, loose connection, or surface overheating. 7. Output Results: The results of the identified anomalies will be output in structured data format. At the same time, the anomaly information will be fed back to the comprehensive evaluation module and the alarm module for subsequent comprehensive evaluation and alarm triggering.

[0023] As an optional embodiment of the present invention, the comprehensive evaluation module, based on the image analysis module and combined with the monitoring results of temperature, humidity, voltage, current, and smoke sensors, comprehensively evaluates the operating status of electrical components and promptly detects abnormalities; including the following steps: 1. Data Collection and Integration: Receives anomaly identification results from the image analysis module; acquires real-time temperature, humidity, voltage, and current data transmitted from the data collection module; integrates monitoring data from the smoke sensor; and unifies the format and synchronizes the time of multi-source heterogeneous data. 2. Data preprocessing: Check the integrity of sensor data, mark or complete missing values; perform data smoothing to eliminate interference from random fluctuations; verify the rationality of the data and remove outliers that are significantly outside the normal range; calculate the real-time rate of change of each parameter. 3. Threshold Judgment: Compare real-time sensor data with preset safety thresholds; generate a single anomaly identifier for each monitored parameter; record the name of the parameter exceeding the threshold, the degree of exceedance, and the duration. 4. Multidimensional feature correlation analysis: Establish correlation models between parameters; identify abnormal linkages between parameters; construct a fault feature vector space and map the current state to known fault modes; 5. Image analysis result fusion: Cross-validate the monitoring results of the image analysis module with temperature sensor data; analyze the potential impact of mechanical structure anomalies on electrical parameters; and determine the fault development trend by combining image time-series changes. 6. Risk Level Assessment: Calculate the weights of each indicator based on the analytic hierarchy process; construct a fuzzy comprehensive evaluation model to calculate the comprehensive risk score; classify risks into different levels; generate a risk heat map to visually display key risk points; 7. Fault Type Identification: Fault type classification is performed based on decision tree algorithm; typical fault modes such as short circuit, overload, and poor contact are identified; the probability confidence of each fault type is calculated; 8. Assessment Results Output: Generate a structured assessment report, including abnormal parameters, risk levels, and possible causes. Output a visual dashboard to display the changing trends of key indicators.

[0024] As an optional embodiment of the present invention, the risk prediction module, based on the image analysis module and combined with the monitoring results of temperature, humidity, voltage, current, and smoke sensors, and referring to historical data, predicts potential problems and risks in the operation of electrical components; including the following steps: 1. Data Collection and Integration: Acquire multi-source data, collecting real-time data from temperature, humidity, voltage, current, and smoke sensors, recording timestamps, parameter values, and other information. Receive results from the image analysis module. Extract historical monitoring data from the database. Link historical fault records, marking the fault occurrence time, type, and related parameter anomalies. Perform format conversion on multi-source heterogeneous data. Align data based on timestamps. 2. Data preprocessing: This includes missing value handling, outlier removal, data smoothing, and feature engineering; feature engineering calculates derived features. Data is then normalized or standardized.

[0025] 3. Historical data analysis: 3.1 Fault Mode Mining: Cluster analysis is performed on historical fault data, and algorithms such as K-means and DBSCAN are used to identify common fault modes. 3.2 Association rule mining: Use the Apriori algorithm or FP-growth algorithm to mine the association relationships between different parameters.

[0026] 3.3 Time Series Feature Extraction: Fourier transform and wavelet transform are performed on the time series data to extract frequency domain features and analyze the periodicity and volatility of the data. A sliding window method is used to extract statistical features within a fixed time window.

[0027] 4. Predictive Model Construction: Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) models are selected for joint prediction using multi-source data. The preprocessed data is divided into training, validation, and test sets. The model is trained using the training set, and its performance is optimized on the validation set by adjusting model parameters (such as the depth of the decision tree and the number of hidden neurons in the LSTM) to avoid overfitting or underfitting. Cross-validation (such as K-fold cross-validation) is employed to improve the reliability of model evaluation. The model is evaluated using metrics such as accuracy, recall, F1 score, mean squared error (MSE), and mean absolute error (MAE).

[0028] 5. Real-time Risk Prediction: Real-time collected and preprocessed sensor data and image analysis results are input into a trained prediction model. The Support Vector Machine (SVM) classification model outputs the probability of different types of failures occurring within a future timeframe. The Long Short-Term Memory (LSTM) network model predicts future values ​​of key parameters. Potential risks are assessed by combining the prediction results with preset risk thresholds. Risks are categorized into different levels and assigned corresponding color codes or warning signals.

[0029] 6. Forecast Results Output: Generate and output a structured report containing forecast results, risk level, possible causes, and recommended measures.

[0030] This invention provides an electrical automation control method, comprising the following steps: S1. Open the cabinet door to place different electrical components onto different support mechanisms for fixation; close the cabinet door. S2. Under normal circumstances, the heat absorption box and S-shaped bend pipe are used to absorb heat and cool the inside of the cabinet; the data collection module of the monitoring mechanism collects real-time and historical data of electrical components; S3, the image acquisition module uses multiple high-definition cameras to capture high-definition images of the cabinet interior and electrical components; S4. The image preprocessing module preprocesses the acquired images, including filtering and denoising, grayscale conversion, normalization, and image enhancement. S5. The feature extraction module extracts features from the preprocessed image. The extracted features include color, texture, and shape. S6. The image analysis module analyzes and identifies the images after feature extraction, and promptly identifies abnormal situations. S7. The comprehensive evaluation module, based on the image analysis module and combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, comprehensively evaluates the operation of electrical components and promptly detects abnormalities. S8. The risk prediction module, based on the image analysis module and combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, and referring to historical data, predicts potential problems and risks in the operation of electrical components. S9. When an abnormal situation or potential risk is detected, the alarm module will issue an alarm in a timely manner. S10. When the temperature inside the cabinet reaches its maximum value, enhanced cooling is implemented using the intake and exhaust fans. Simultaneously, the oil pump is activated to increase the flow rate of the heat transfer oil, further enhancing the cooling effect.

[0031] 3. Beneficial effects One or more technical solutions provided in this application have at least the following technical effects or advantages: 1. Through the sliding design of the support mechanism, the operator can quickly pull out the support plate to realize "plug and play" of electrical components, improve installation efficiency, and avoid the space limitations of traditional fixed installation.

[0032] 2. Through the regional storage design, electrical components can be arranged according to their functions and power consumption levels, reducing cable clutter and allowing for quick location of faulty components during later maintenance, thus improving work efficiency.

[0033] 3. It achieves a synergistic effect of passive and active heat dissipation to improve the cooling effect; it can form a dual active heat dissipation of "air convection + liquid heat conduction" to improve the cooling rate.

[0034] 4. Intelligent assessment and risk warning: The comprehensive assessment model combines sensor data and image analysis results, using fuzzy logic algorithms to classify anomalies and avoid misjudgments based on single parameters. A time-series model trained on historical data can predict potential risks such as aging trends of electrical components and poor contact in advance. Attached Figure Description

[0035] Figure 1 This is an overall schematic diagram of an electrical automation control cabinet disclosed in a preferred embodiment of this application; Figure 2 This is a schematic diagram of the top structure inside an electrical automation control cabinet according to a preferred embodiment of this application; Figure 3 This is a schematic diagram of the closed state of the cabinet door of the electrical automation control cabinet disclosed in a preferred embodiment of this application; Figure 4 This is a schematic diagram of the support mechanism of an electrical automation control cabinet disclosed in a preferred embodiment of this application.

[0036] Reference numerals in the attached diagram: 1. Cabinet body; 2. Cabinet door; 3. Support mechanism; 4. S-shaped bend; 5. Intake fan; 6. Exhaust fan; 7. Oil supply assembly; 8. Heat absorption box; 11. Horizontal partition; 12. Vertical partition; 13. Guide rail; 14. U-shaped slide rail; 31. Support plate; 32. Trapezoidal rod; 33. Handle; 34. Support rod; 35. Slider; 71. Oil storage tank; 72. Oil pump; 73. Oil supply pipe; 74. Oil return pipe. Detailed Implementation

[0037] The present application will be further described in detail below with reference to the accompanying drawings.

[0038] Reference Figure 1 and Figure 2 This application provides an electrical automation control cabinet, including: cabinet body 1, cabinet door 2, support mechanism 3, S-shaped bend 4, air intake fan 5, exhaust fan 6, oil supply assembly 7, heat absorption box 8, and monitoring mechanism; Cabinet 1 is equipped with a rotatable door 2. Cabinet 1 is made of high-quality cold-rolled steel plate with an electrostatic spray coating, providing excellent rust resistance and wear resistance. Cabinet door 2 is connected to cabinet 1 via stainless steel hinges, ensuring it will not rust or jam over long-term use. Cabinet door 2 is equipped with a sealing strip, effectively preventing dust and moisture from entering the interior of cabinet 1 when closed, thus achieving an IP54 protection rating. Cabinet door 2 also has an observation window made of explosion-proof glass, allowing operators to observe the operating status of electrical components inside cabinet 1 without opening cabinet door 2.

[0039] Multiple horizontal partitions 11 and vertical partitions 12 are vertically and crosswise fixed inside the cabinet 1. Both the horizontal partitions 11 and vertical partitions 12 are made of high-strength aluminum alloy and have heat dissipation grooves on their surfaces, which helps to improve the heat dissipation effect inside the cabinet 1.

[0040] Each of the transverse partitions 11 is slidably provided with a support mechanism 3, which is used to place electrical components; Multiple horizontal partitions 11 and vertical partitions 12 divide the interior space of the cabinet 1 into multiple electrical component storage areas; different electrical components can be placed separately, while avoiding cable clutter.

[0041] An intake fan 5 is fixedly installed on the lower part of one side of cabinet 1, and an exhaust fan 6 is fixedly installed on the lower part of the other side of cabinet 1. The intake fan 5 and exhaust fan 6 are used to cool the interior of cabinet 1. Both intake fans 5 and exhaust fans 6 are axial flow fans, characterized by high airflow and low noise. The fan blades of intake fans 5 and exhaust fans 6 are made of aviation-grade aluminum alloy, which is high in strength and lightweight, effectively improving the operating efficiency of the fans. The fan motors are high-temperature resistant and moisture-proof, suitable for the working environment inside cabinet 1. Additionally, protective nets and activated carbon adsorption boxes are installed on the exterior of intake fans 5 and exhaust fans 6. The protective nets prevent foreign objects from entering the fan and damaging it; the activated carbon adsorption boxes contain activated carbon to absorb moisture.

[0042] Several S-shaped bends 4 are installed inside the cabinet 1. An oil supply assembly 7 is fixedly installed on the outside of the cabinet 1, and the oil supply assembly 7 is connected to each S-shaped bend 4. The oil supply assembly 7 allows heat transfer oil to flow inside the S-shaped bends 4, thereby absorbing heat and cooling the internal space of the cabinet 1. The S-shaped bends 4 are made of copper with excellent thermal conductivity, ensuring that the heat transfer oil can quickly absorb the heat inside the cabinet 1 when flowing inside the pipes. The S-shaped bends 4 are distributed in a staggered manner inside the cabinet 1, covering the storage areas of various electrical components, maximizing the heat dissipation area.

[0043] Several heat-absorbing boxes 8 are fixedly installed inside the cabinet 1. Each heat-absorbing box 8 contains phase change wax; the heat-absorbing boxes 8 absorb heat and cool the interior of the cabinet 1. The heat-absorbing boxes 8 are made of aluminum alloy and are rectangular in shape. The interior of each heat-absorbing box 8 is filled with high latent heat phase change wax with a phase change temperature of 45℃, which can rapidly absorb heat when the internal temperature of the cabinet 1 rises, achieving passive heat dissipation. The surface of each heat-absorbing box 8 is equipped with heat dissipation fins with a wave-shaped design, which effectively increases the heat dissipation area and improves heat dissipation efficiency. The heat-absorbing boxes 8 are fixed inside the cabinet 1 with bolts, making installation and disassembly convenient.

[0044] A monitoring mechanism is installed inside cabinet 1 to monitor the operation of electrical components inside cabinet 1, detect abnormalities in a timely manner, and issue alarms promptly.

[0045] In this technical solution, opening cabinet door 2 allows different electrical components to be placed and fixed on different support mechanisms 3. Under normal circumstances, the heat absorption box 8 and S-shaped bend 4 absorb heat and cool the interior of cabinet 1. The monitoring mechanism monitors the operation of the electrical components inside cabinet 1. When the temperature inside cabinet 1 reaches its maximum value, the intake fan 5 and exhaust fan 6 are used to enhance cooling.

[0046] Reference Figure 1 and Figure 4 The support mechanism 3 includes a support plate 31, a trapezoidal rod 32, a handle 33, a support rod 34, and a slider 35; Two guide rails 13 are fixedly installed in parallel on the partition 11 of each electrical component storage area. The guide rails 13 are provided with trapezoidal grooves. Two U-shaped slide rails 14 are symmetrically fixed below the partition 11 of each electrical component storage area. Two trapezoidal rods 32 are fixedly mounted parallel to each other below the support plate 31. The trapezoidal rods 32 slide in contact with the corresponding guide rails 13. The support plate 31 is made of high-strength aluminum alloy sheet with anodized surface treatment, ensuring both lightweight construction and enhanced wear and corrosion resistance. The edges of the sheet are rounded to prevent sharp corners from scratching operators or electrical cables. The trapezoidal rods 32 are made of high-quality 45# carbon structural steel, quenched and tempered to ensure good wear resistance and deformation resistance when mating with the trapezoidal grooves of the guide rails 13. The trapezoidal rods 32 are connected to the support plate 31 by welding, and the weld seams are ground to ensure a smooth surface and prevent interference with the sliding of the guide rails 13.

[0047] A handle 33 is fixedly installed on the outer side of the trapezoidal rod 32; a support rod 34 is fixedly installed at the lower end of the handle 33; a slider 35 is fixedly installed at the rear end of the support rod 34; the two support rods 34 are arranged in a figure-eight shape; the slider 35 is slidably installed on the U-shaped slide rail 14.

[0048] The support plate 31 is equipped with adjustable fixing clips (not shown in the figure). The fixing clips are made of high-strength plastic and covered with anti-slip rubber pads. The fixing clips are adjusted by screws and nuts, allowing operators to adjust the spacing of the clips according to the size of the electrical components, thus ensuring a secure fixation for electrical components of different sizes. The support mechanism 3 also features vibration damping pads made of rubber, which effectively reduce vibrations generated during the operation of the electrical components, protecting their normal operation. The vibration damping pads employ a "honeycomb rubber + spring composite structure." The upper layer is a honeycomb rubber layer, which effectively absorbs high-frequency vibrations; the lower layer is a miniature helical spring, which absorbs low-frequency vibrations. The combination of these two elements achieves full-frequency vibration isolation.

[0049] In this technical solution, the support plate 31 is pulled outward by grasping the handle 33 to install the electrical components onto the support plate 31; then the support plate 31 and the electrical components are pushed into the cabinet 1. The trapezoidal rod 32 slides against the corresponding guide rail 13, and the slider 35 is slidably mounted on the U-shaped slide rail 14, effectively supporting the support plate 31 and the electrical components and improving the stability of the support mechanism 3. The figure-eight structure design, with a 60° angle between the two rods, provides more stable support when the slider 35 slides on the U-shaped slide rail 14. The U-shaped slide rail 14 is made of cold-rolled steel, which can fully accommodate the slider 35 and provide stable guidance. Limit blocks are provided at both ends of the U-shaped slide rail 14 to prevent the slider 35 from sliding off the rail, ensuring the safety of the support mechanism 3 during sliding.

[0050] Reference Figure 3 The oil supply assembly 7 includes an oil reservoir 71, an oil pump 72, an oil supply pipe 73, and an oil return pipe 74; An oil storage tank 71 is fixedly installed on the outside of the cabinet 1. An oil pump 72 is fixedly installed on the oil storage tank 71. The input end of the oil pump 72 extends into the oil storage tank 71. The output end of the oil pump 72 is connected to the oil supply pipe 73. An oil return pipe 74 is installed on the oil storage tank 71. After several S-shaped bends 4 are connected end to end, both ends are connected to the oil supply pipe 73 and the oil return pipe 74 respectively.

[0051] In this technical solution, the oil storage tank 71 is welded from 304 stainless steel plate with a wall thickness of 3mm and polished inner and outer surfaces. Two baffles are installed inside to form a "U-shaped" flow channel, extending the residence time of the heat transfer oil and promoting heat dissipation. An oil inlet (with filter screen) and a breather valve (with dust cover) are located at the top, and a liquid level observation window is located on the side. The oil pump 72 is a magnetically driven centrifugal pump. The oil supply pipe 73 / return pipe 74 is connected to the S-shaped bend 4 via flanges for easy disassembly and maintenance. A Y-type filter with an 80-mesh filter screen is installed at the inlet of the oil pump 72. The filter is equipped with a differential pressure alarm device (alarm when differential pressure ≥ 0.1MPa).

[0052] Furthermore, a copper wire mesh is fixedly installed inside the heat absorption box 8. The copper wire mesh is located at the bottom of the heat absorption box 8 and is connected to the S-shaped bend 4 through copper sheets. The phase change wax is filled into the heat absorption box 8 to three-quarters of its volume, leaving sufficient margin.

[0053] In this technical solution, heat is absorbed by the phase change wax inside the heat-absorbing box 8 to dissipate heat and cool the interior of the cabinet 1. The main body of the heat-absorbing box 8 is made of 6063 aluminum alloy profile with an anodized surface. A 5mm high support boss is set at the bottom of the heat-absorbing box 8 to fix the copper wire mesh. The top is designed with a removable cover plate, using a snap-fit ​​+ sealing strip structure. The copper wire mesh is made of T2 copper and is stacked in three layers. One end of the copper sheet is welded to the copper wire mesh (using silver-based brazing, welding strength ≥150MPa), and the other end is tightly connected to the S-shaped bend 4 through a clamp. Thermal grease is applied to the contact surface.

[0054] Phase change waxes are selected from composite paraffin waxes; composite paraffin waxes include base paraffin wax, nucleating agents, additives, and thickeners; Additive 5%: Expanded graphite (5% wt): Increases thermal conductivity to 1.2 W / m·K; Nucleating agent 0.5% (nano-alumina): Reduces supercooling to ≤2℃; Thickener 0.3% (fumed silica): to prevent phase separation; Base paraffin 94%-94.2%: Provides the core function of phase change.

[0055] First, heat the paraffin wax to 70℃ to melt it → add nano-alumina (disperse at high speed for 30 minutes) → add fumed silica (continue dispersing for 20 minutes) → finally add expanded graphite (stir at low speed for 15 minutes). Avoid adding expanded graphite too early, which could damage the structure. After mixing, let it stand for ten minutes under a -0.09MPa vacuum to remove air bubbles and prevent affecting thermal conductivity. Pour the mixture into the molten state (60-65℃), filling 75% of the absorber box volume, leaving space for expansion.

[0056] Furthermore, an intelligent frame circuit breaker (such as the Schneider MT series) equipped with an electronic trip unit (MIC5.0A) is installed at the main power inlet of cabinet 1. An intelligent miniature circuit breaker (such as the Siemens 5SX series) is installed in the power supply circuit of each partition 11, and a current sensor (Hall effect type) and a voltage sensor (resistive voltage divider type) are integrated at the power input terminal of the support mechanism 3.

[0057] In this technical solution, when a sudden voltage drop (e.g., <90% of rated voltage) is detected and the current continues to rise, it is preferentially determined to be a motor stall fault, and the branch circuit trip is immediately triggered to avoid the delay of simple current monitoring. When a voltage fluctuation exceeding ±10% is detected, the intake fan 5 and exhaust fan 6 are automatically started (regardless of whether the temperature meets the standard) to prevent the equipment from overheating due to abnormal voltage.

[0058] Furthermore, the monitoring agencies include: Data Collection Module: Collects real-time and historical data from electrical components; including temperature and humidity sensors, current sensors, point pressure sensors, and smoke sensors. The temperature and humidity sensors utilize high-precision digital sensors, capable of accurately sensing subtle changes in ambient temperature and humidity, providing precise data support for the operating environment of electrical components. The current sensor, based on the Hall effect principle, can quickly respond to current changes and monitor the current parameters of electrical components in real time. The voltage sensor uses an isolated voltage transformer, effectively isolating high-voltage and low-voltage currents, and safely and reliably collecting voltage data from electrical components. The smoke sensor uses a photoelectric sensor, which can trigger an alarm signal within a short time when smoke particles are detected, with a response time of less than ten seconds. Each sensor transmits data to the control center via RS485 or Modbus communication protocols, ensuring the stability and accuracy of data transmission. Simultaneously, the data collection module is equipped with a large-capacity storage device, which can store historical data for a long time, providing a data foundation for subsequent analysis and prediction. This is existing technology and will not be elaborated further.

[0059] Image acquisition module: includes multiple high-definition cameras and LED lights, to acquire high-definition images of the interior of cabinet 1 and electrical components; Image preprocessing module: This module preprocesses the acquired images, including denoising, grayscale conversion, normalization, and image enhancement. Denoising employs algorithms such as median filtering and Gaussian filtering to effectively remove salt-and-pepper noise and Gaussian noise, improving image clarity. Grayscale conversion transforms the color image into a grayscale image, reducing data volume while highlighting the grayscale features for easier subsequent analysis. Normalization standardizes the pixel values, giving the image data a uniform scale and range, improving the algorithm's stability and generalization ability. Image enhancement uses techniques such as histogram equalization and adaptive histogram equalization to enhance image contrast and detail, making the features of electrical components more prominent.

[0060] The feature extraction module extracts features from the preprocessed image, including color, texture, and shape. For color feature extraction, the HSV color space model is used to decompose the image's color information into three components: hue, saturation, and lightness. Features such as color histograms and color moments are extracted to accurately describe the distribution and statistical characteristics of colors in the image. Texture feature extraction employs algorithms such as gray-level co-occurrence matrix and local binary mode to extract features such as texture direction, thickness, and contrast from the image, used to identify texture variations on the surface of electrical components. Shape feature extraction utilizes methods such as contour extraction and Fourier descriptors to extract the contour shape and geometric parameters of electrical components, achieving a precise description of their shape. These feature extraction methods comprehensively and accurately extract key features from the image, providing rich information for subsequent image analysis.

[0061] Image analysis module: Analyzes and identifies images after feature extraction, promptly detecting anomalies; Comprehensive evaluation module: Based on the image analysis module, combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, the module comprehensively evaluates the operation of electrical components and promptly detects abnormalities. Risk prediction module: Based on the image analysis module, combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, and referring to historical data, predict potential problems and risks in the operation of electrical components; Alarm module: includes an alarm that promptly issues an alert when abnormal situations or potential risks are detected; Control Center: A programmable PLC control module; network-connected to data collection module, image acquisition module, image preprocessing module, feature extraction module, image analysis module, comprehensive evaluation module, risk prediction module, and alarm module.

[0062] Furthermore, the image analysis module analyzes and identifies the extracted features from the image to promptly detect anomalies; this includes the following steps: 1. Model Training and Deployment: Based on a large number of normal and abnormal image samples of electrical components, a Convolutional Neural Network (CNN) model is trained. Hyperparameters of the model, such as the learning rate and number of iterations, are adjusted to optimize the network structure and achieve optimal performance. After training, the validated CNN model is deployed to the computing device of the image analysis module to provide algorithmic support for subsequent image analysis. 2. Image Input: The feature extraction module transmits the processed image data to the image analysis module. The image data contains feature information such as color, texture and shape, and enters the module in a specific data format (such as tensor form) to await analysis. 3. Feature Map Extraction: The input image enters the convolutional layer of the CNN model. Different convolutional kernels slide across the image to perform convolution operations, generating multiple feature maps. These maps capture feature information of different levels and types in the image, such as edges and texture details, enabling further extraction and abstraction of image features. 4. Feature Dimensionality Reduction and Filtering: The feature maps extracted by the convolutional layers are then processed by the pooling layers. The pooling layers downsample the feature maps using methods such as max pooling or average pooling, reducing data dimensionality and computational cost, while retaining the most critical feature information in the image and filtering out representative features. 5. Fully connected layer processing: The pooled feature data is flattened and input into the fully connected layer. The fully connected layer fuses all the features and maps the feature data to a specific output dimension through a series of neuron calculations, outputting the probability values ​​of different states (normal, abnormal) of the corresponding electrical components. 6. Abnormal Situation Judgment: The probability value output by the fully connected layer is compared with a preset threshold. If the probability value of a certain type of abnormal situation exceeds the preset threshold, the electrical component is determined to have that type of abnormality, such as component damage, loose connection, or surface overheating. 7. Result Output: The results of the identified anomalies will be output in the form of structured data. At the same time, the anomaly information will be fed back to the comprehensive evaluation module and the alarm module for subsequent comprehensive evaluation and alarm triggering. If no anomalies are found in the image analysis, normal results will also be output to maintain the continuous monitoring process of the monitoring system.

[0063] Furthermore, the comprehensive evaluation module, based on the image analysis module and combined with the monitoring results from temperature, humidity, voltage, current, and smoke sensors, comprehensively evaluates the operating status of electrical components and promptly detects abnormalities; this includes the following steps: 1. Data Collection and Integration: Receives anomaly identification results (such as component damage, loose connection, etc.) output by the image analysis module; acquires real-time temperature, humidity, voltage, and current data transmitted by the data collection module; integrates monitoring data from the smoke sensor; and unifies the format and synchronizes the time of multi-source heterogeneous data. 2. Data preprocessing: Check the integrity of sensor data, mark or complete missing values; perform data smoothing to eliminate interference from random fluctuations; verify the rationality of the data and remove outliers that are significantly outside the normal range; calculate the real-time rate of change of each parameter (such as the rate of temperature rise). 3. Threshold judgment: Compare real-time sensor data with preset safety thresholds; generate a single abnormality label for each monitored parameter (e.g., temperature > 85℃, current > 110% of rated value); record the name of the parameter exceeding the threshold, the degree of exceedance, and the duration. 4. Multidimensional feature correlation analysis: Establish correlation models between parameters (such as the correlation between temperature and current); identify abnormal linkages between parameters (such as normal voltage but sudden increase in current); construct a fault feature vector space and map the current state to known fault modes; 5. Image analysis result fusion: Cross-validate the monitoring results of the image analysis module with temperature sensor data; analyze the potential impact of mechanical structural anomalies (such as loosening) on ​​electrical parameters; and determine the fault development trend by combining image time-series changes. 6. Risk Level Assessment: Calculate the weights of each indicator based on the Analytic Hierarchy Process (AHP); construct a fuzzy comprehensive evaluation model to calculate the overall risk score; classify risks into different levels (e.g., normal, caution, warning, and emergency); generate a risk heatmap to visually display key risk points; and conduct a comprehensive assessment according to the following formula: S=w T D T +w I D I +w V D V +w G D G +aI TH +bI EL +cI GI ;w T +w I +w V +w G =1; where w T ,w I ,w V ,w G These are the weighting coefficients for temperature, current, voltage, and image analysis results (determined through the analytic hierarchy process). D T , I D I D V and D G These are standardized risk values ​​for temperature, current, voltage, and image analysis (typically ranging from [0,1], with higher values ​​indicating higher risk). a, b, and c are weighting coefficients for temperature-current correlation, voltage-current correlation, and image-electrical parameter correlation (these need to be calibrated based on historical data). TH This refers to anomalies in the cross-correlation coefficient between temperature and current (such as the degree to which the sliding window cross-correlation coefficient deviates from the normal range). EL These are abnormal values ​​indicating dynamic linkage between voltage and current (such as the magnitude of a sudden change in current despite normal voltage). GI It is the cross-validation of outliers between image analysis results and electrical parameters (such as the correlation between current fluctuations caused by mechanical loosening).

[0064] 7. Fault Type Identification: Fault type classification is performed based on decision tree algorithm; typical fault modes such as short circuit, overload, and poor contact are identified; the probability confidence of each fault type is calculated; 8. Assessment Results Output: Generate a structured assessment report, including abnormal parameters, risk levels, and possible causes. Output a visual dashboard to display the changing trends of key indicators.

[0065] Furthermore, the risk prediction module, based on the image analysis module and combined with monitoring results from temperature, humidity, voltage, current, and smoke sensors, and referencing historical data, predicts potential problems and risks in the operation of electrical components; this includes the following steps: 1. Data Collection and Integration: Acquire multi-source data, collecting real-time data from temperature, humidity, voltage, current, and smoke sensors, recording timestamps, parameter values, and other information. Receive results from the image analysis module. Extract historical monitoring data from the database. Link historical fault records, marking the fault occurrence time, type, and related parameter anomalies. Convert the formats of multi-source heterogeneous data to ensure consistency in data type and units. Align data based on timestamps, using methods such as linear interpolation to handle time asynchrony issues, ensuring that various types of data from the same moment are available for analysis. 2. Data Preprocessing: This includes missing value handling, outlier removal, data smoothing, and feature engineering. Feature engineering calculates derived features, such as the rate of temperature change and the ratio of current to voltage, enriching the data's feature dimensions. Data is normalized or standardized to map it to the 0-1 range, improving model training efficiency and accuracy.

[0066] 3. Historical data analysis: 3.1 Fault Mode Mining: Cluster analysis is performed on historical fault data, and algorithms such as K-means and DBSCAN are used to identify common fault modes. For example, historical overload fault data is clustered to summarize the typical changes in parameters such as current and voltage when an overload occurs. The evolution trend of parameters before the fault occurs is analyzed to determine fault warning characteristics. For example, before a short-circuit fault occurs, there may be a trend of sudden voltage drop, current surge, and rapid temperature rise. 3.2 Association Rule Mining: Using the Apriori algorithm or FP-growth algorithm, the association relationships between different parameters are mined. For example, it was found that when the humidity exceeds 80% and the voltage fluctuation exceeds ±5%, the probability of equipment failure due to poor contact increases by 30%.

[0067] 3.3 Time Series Feature Extraction: Fourier transform and wavelet transform are performed on time series data (such as temperature and current changes over time) to extract frequency domain features and analyze the periodicity and volatility of the data. A sliding window method is used to extract statistical features (such as mean, variance, maximum, and minimum values) within a fixed time window. 4. Predictive Model Construction: Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) models are selected for joint prediction using multi-source data. The preprocessed data is divided into training, validation, and test sets. The model is trained using the training set, and model performance is optimized on the validation set by adjusting model parameters (such as the depth of the decision tree and the number of hidden neurons in the LSTM) to avoid overfitting or underfitting. Cross-validation (such as K-fold cross-validation) is used to improve the reliability of model evaluation. The model is evaluated using metrics such as accuracy, recall, F1 score, mean squared error (MSE), and mean absolute error (MAE). Risk prediction is performed according to the following formula: Trend k (t fa )={Σ N i=1 {|D k hist (t fa -i△t)-D k hist [t fa -(i+1)△t]|w i}} / [Σ N i=1 w i ]; w i =e -λi；式中，Trendk (t fa ) is at the time t when the fault occurs. fa Previously, the evolution trend of parameter k (such as temperature, current, voltage, etc.) was studied. This trend was determined by analyzing the changes in parameter values ​​over a period of time before the fault occurred, reflecting the dynamic behavior of the parameter as the fault approached. fa This refers to the time the fault occurred. (D) his (t) represents the historical monitoring value of parameter k at time t. Δt is the time step. N is the size of the time window used to calculate the trend. i λ is the weighting factor used to emphasize the importance of recent data; k is the electrical parameter that needs to be monitored and analyzed, such as temperature (T), current (I), voltage (V), etc. i is the time step; Δt is the time step size or time interval in time series analysis or trend calculation. e is the base of the natural logarithm, a constant, with an approximate value of 2.71828.

[0068] 5. Real-time Risk Prediction: Real-time collected and preprocessed sensor data and image analysis results are input into a trained prediction model. The Support Vector Machine (SVM) classification model outputs the probability of different types of failures occurring within a future timeframe. The Long Short-Term Memory (LSTM) network model predicts future values ​​of key parameters. The prediction results are combined with preset risk thresholds to assess potential risks. Risks are categorized into different levels, such as low, medium, high, and critical, and assigned corresponding color codes or warning signals.

[0069] 6. Forecast Output: Generates a structured report containing forecast results, risk level, possible causes, and recommended measures. A dashboard displays real-time values ​​and forecast trend curves of key parameters, and visualization tools such as heatmaps and Sankey diagrams visually represent risk distribution and propagation paths. When medium- to high-risk conditions are predicted, relevant maintenance personnel are promptly notified via SMS, email, and audible / visual alarms to ensure rapid problem resolution.

[0070] This invention provides an electrical automation control method, comprising the following steps: S1. Open cabinet door 2 to place different electrical components onto different support mechanisms 3 for fixation; close cabinet door 2. S2. Under normal circumstances, the heat absorption box 8 and S-shaped bend 4 are used to absorb heat and cool the inside of the cabinet 1; the data collection module of the monitoring mechanism collects real-time and historical data of electrical components; S3, the image acquisition module uses multiple high-definition cameras to acquire high-definition images of the interior of cabinet 1 and electrical components; S4. The image preprocessing module preprocesses the acquired images, including filtering and denoising, grayscale conversion, normalization, and image enhancement. S5. The feature extraction module extracts features from the preprocessed image. The extracted features include color, texture, and shape. S6. The image analysis module analyzes and identifies the images after feature extraction, and promptly identifies abnormal situations. S7. The comprehensive evaluation module, based on the image analysis module and combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, comprehensively evaluates the operation of electrical components and promptly detects abnormalities. S8. The risk prediction module, based on the image analysis module and combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, and referring to historical data, predicts potential problems and risks in the operation of electrical components. S9. When an abnormal situation or potential risk is detected, the alarm module will issue an alarm in a timely manner. S10. When the temperature inside cabinet 1 reaches its maximum value, enhanced cooling is achieved through intake fan 5 and exhaust fan 6. At the same time, oil pump 72 is activated to increase the flow rate of heat transfer oil and enhance the cooling effect.

[0071] The working principle of the electrical automation control cabinet of this invention is as follows: Opening the cabinet door 2 allows different electrical components to be placed and fixed onto different support mechanisms 3; closing the cabinet door 2; under normal circumstances, the heat absorption box 8 and S-shaped bend 4 absorb heat and cool the interior of the cabinet 1; the data collection module of the monitoring mechanism collects real-time and historical data of the electrical components; the image acquisition module acquires high-definition images of the interior of the cabinet 1 and the electrical components through multiple high-definition cameras; the image preprocessing module preprocesses the acquired images, including noise reduction, grayscale conversion, normalization, and image enhancement; the feature extraction module extracts features from the preprocessed images, including color, texture, and shape; image segmentation... The image analysis module analyzes and identifies the extracted features from the images to promptly detect anomalies. The comprehensive evaluation module, based on the image analysis module and the monitoring results from temperature, humidity, voltage, current, and smoke sensors, comprehensively evaluates the operation of electrical components and promptly identifies anomalies. The risk prediction module, based on the image analysis module, the monitoring results from temperature, humidity, voltage, current, and smoke sensors, and referring to historical data, predicts potential problems and risks in the operation of electrical components. The alarm module includes an alarm that promptly issues an alert when an anomaly or potential risk is detected. When the temperature inside cabinet 1 reaches its maximum value, enhanced cooling is achieved through intake fan 5 and exhaust fan 6. Simultaneously, oil pump 72 is activated to increase the flow rate of the heat transfer oil, enhancing the cooling effect.

[0072] This invention utilizes the sliding design of the support mechanism 3, allowing operators to quickly pull out the support plate 31, enabling "plug-and-play" installation of electrical components, improving installation efficiency and avoiding the space limitations of traditional fixed installations. Through a zoned storage design, electrical components can be arranged according to their function and power consumption level, reducing cable clutter and allowing for quick location of faulty components during later maintenance, thus improving work efficiency. It achieves synergistic passive and active heat dissipation to enhance cooling: dual passive heat dissipation is achieved through phase change wax and the S-shaped bend 4; the composite phase change wax in the heat absorption box 8 absorbs heat at 45℃ through a solid-liquid phase change, and, combined with copper wire mesh and copper sheet thermal bridges, rapidly conducts heat to the heat-conducting oil in the S-shaped bend 4, suppressing internal temperature rise. The S-shaped bend 4, made of copper with fins, allows the heat-conducting oil to circulate under the drive of the oil pump 72, further dissipating heat to the oil storage tank 71, which, in practice, reduces the temperature in key areas of the cabinet. When the temperature exceeds the threshold, intake fan 5, exhaust fan 6, and oil pump 72 activate in tandem, forming a dual active cooling system of "air convection + liquid heat conduction" to improve the cooling rate. Intelligent assessment and risk warning are achieved. The comprehensive assessment model combines sensor data and image analysis results, using fuzzy logic algorithms to classify anomalies and avoid misjudgments based on single parameters. A time-series model trained on historical data can predict potential risks such as aging trends of electrical components and poor contact in advance.

[0073] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An electrical automation control method, characterized in that, Includes the following steps: S1. Secure the electrical components to their respective support mechanisms; close the cabinet door. S2. Under normal circumstances, the heat absorption box and S-shaped bend pipe are used to absorb heat and cool the inside of the cabinet; the data collection module of the monitoring mechanism collects real-time and historical data of electrical components; S3, the image acquisition module acquires high-definition images of the cabinet interior and electrical components; S4. The image preprocessing module preprocesses the acquired images; S5. The feature extraction module extracts features from the preprocessed image; S6. The image analysis module analyzes and identifies the images after feature extraction to detect abnormal situations. S7. The comprehensive evaluation module integrates multi-source data to comprehensively evaluate the operation of electrical components and promptly detect abnormalities. S8. The risk prediction module integrates multi-source data and references historical data to predict potential problems and risks in the operation of electrical components. S9. When an abnormal situation or potential risk is detected, the alarm module will issue an alarm in a timely manner. S10. When the temperature inside the cabinet reaches its maximum value, the intake and exhaust fans are used to enhance cooling, and the oil pump is started to increase the flow rate of the heat transfer oil to enhance the cooling effect.

2. The electrical automation control method according to claim 1, characterized in that: Step S6 includes the following steps: S61. Model Training and Deployment: Select a Convolutional Neural Network (CNN) model and train it based on a large number of normal and abnormal image samples of electrical components; adjust the model's hyperparameters to achieve optimal performance; after training, deploy the validated CNN model to the computing device of the image analysis module. S62, Image Input: Transmit the processed image data to the image analysis module; S63. Feature Map Extraction: The input image enters the convolutional layer of the convolutional neural network (CNN) model. Different convolutional kernels slide on the image to perform convolution operations and generate multiple feature maps. S64. Feature Dimensionality Reduction and Filtering: The feature map extracted by the convolutional layer is then processed by the pooling layer. S65. Fully connected layer processing: The pooled feature data is flattened and input into the fully connected layer. The fully connected layer fuses all the features and outputs the probability values ​​of different states of the corresponding electrical components. S66. Abnormal situation judgment: Based on the probability value output by the fully connected layer, compare it with the preset threshold and determine the abnormal situation of the electrical component. S67. Output Results: Output the results of the identified abnormal situations.

3. The electrical automation control method according to claim 1, characterized in that: Step S7 includes the following steps: S71. Data Collection and Integration: Receives anomaly identification results from the image analysis module; acquires real-time data from the data collection module; accesses monitoring data from the smoke sensor; and unifies the format and synchronizes the time of multi-source heterogeneous data. S72. Data preprocessing: Check the integrity of sensor data, mark or fill in missing values; perform data smoothing, verify the rationality of data, and calculate the real-time change rate of each parameter. S73, Threshold Judgment: Compare real-time sensor data with preset safety thresholds; record the name of the parameter exceeding the threshold, the degree of exceeding the limit, and the duration. S74. Multidimensional Feature Correlation Analysis: Establish a correlation model between parameters; identify abnormal linkages between parameters; construct a fault feature vector space and map the current state to known fault modes; S75. Image Analysis Result Fusion: Cross-validate the monitoring results of the image analysis module with temperature sensor data; analyze the potential impact of mechanical structure anomalies on electrical parameters; and determine the fault development trend by combining image time-series changes. S76. Risk Level Assessment: Calculate the weights of each indicator based on the analytic hierarchy process; construct a fuzzy comprehensive evaluation model to calculate the comprehensive risk score; and classify risks into different levels. S77. Fault Type Identification: Fault type classification based on decision tree algorithm; identification of typical fault modes; calculation of the probability confidence of each fault type; S78. Evaluation Result Output: Generate a structured evaluation report and output it visually.

4. The electrical automation control method according to claim 1, characterized in that: Step S8 includes the following steps: S81. Data Collection and Integration: Acquire multi-source data; perform format conversion on multi-source heterogeneous data, and align data based on timestamps; S82. Data preprocessing: including missing value handling, outlier removal, data smoothing, and feature engineering; S83, Historical Data Analysis; S84. Prediction Model Construction: Select Support Vector Machine (SVM) and Long Short-Term Memory (LSTM) models to perform joint prediction based on multi-source data; S85. Real-time Risk Prediction: Input real-time collected sensor data and image analysis results into the trained prediction model; the Support Vector Machine (SVM) classification model outputs the probability of different types of failures occurring in the future; the Long Short-Term Memory (LSTM) network model predicts the future values ​​of key parameters; and the potential risks are assessed by combining the prediction results with preset risk thresholds. S86. Prediction Result Output: Generate and output the prediction report.

5. The electrical automation control method according to claim 4, characterized in that: In step S83, cluster analysis is performed on historical fault data to identify common fault modes; the correlation between different parameters is explored; Fourier transform is performed on time series data to extract frequency domain features and analyze the periodicity and volatility of the data.

6. The electrical automation control method according to claim 1, characterized in that: The support mechanism includes a support plate, a trapezoidal rod, a handle, a support rod, and a slider; Two guide rails are fixedly installed in parallel on the partition of each electrical component storage area, and the guide rails have trapezoidal grooves. Two U-shaped slide rails are fixedly installed symmetrically below the partition of each electrical component storage area. Two trapezoidal rods are fixedly installed in parallel below the support plate. The trapezoidal rods slide with the corresponding guide rails. A handle is fixedly installed on the outside of the trapezoidal rod. A support rod is fixedly installed at the lower end of the handle. A slider is fixedly installed at the rear end of the support rod. The slider is slidably mounted on the U-shaped slide rail.

7. The electrical automation control method according to claim 1, characterized in that: The fuel supply system includes a fuel tank, a fuel pump, a fuel supply pipe, and a fuel return pipe; An oil storage tank is fixedly installed on the outside of the cabinet, and an oil pump is fixedly installed on the oil storage tank. The input end of the oil pump extends into the oil storage tank, and the output end of the oil pump is connected to the oil supply pipe. A return oil pipe is installed on the oil storage tank. Several S-shaped bends are connected end to end, and their two ends are connected to the oil supply pipe and the return oil pipe, respectively.

8. The electrical automation control method according to claim 7, characterized in that: A copper wire mesh is fixedly installed inside the heat absorption box. The copper wire mesh is located at the bottom of the heat absorption box and is connected to the S-shaped bend tube through a copper sheet. The amount of phase change wax in the heat absorption box is three-quarters of the volume of the heat absorption box.

9. The electrical automation control method according to claim 1, characterized in that: Monitoring agencies include: Data collection module: collects real-time and historical data of electrical components; Image acquisition module: Includes multiple high-definition cameras and LED lights to acquire high-definition images of the cabinet interior and electrical components; Image preprocessing module: preprocesses the acquired images, including noise reduction, grayscale conversion, normalization, and image enhancement; Feature extraction module: Extracts features from the preprocessed image; Image analysis module: Analyzes and identifies images after feature extraction, promptly detecting anomalies; Comprehensive evaluation module: Based on the image analysis module, combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, the module comprehensively evaluates the operation of electrical components and identifies abnormalities. Risk prediction module: Based on the image analysis module, combined with the monitoring results of temperature, humidity, voltage, current and smoke sensors, and referring to historical data, predict potential problems and risks in the operation of electrical components; Alarm module: includes an alarm that sounds when an abnormal situation or potential risk is detected; Control Center: Network connected to the data collection module, image acquisition module, image preprocessing module, feature extraction module, image analysis module, comprehensive evaluation module, risk prediction module, and alarm module.

10. An electrical automation control cabinet, comprising: Cabinet body, cabinet door, support mechanism, S-shaped bend, intake fan, exhaust fan, oil supply assembly, heat absorption box, and monitoring mechanism; characterized in that: The cabinet is equipped with a rotatable cabinet door; multiple horizontal and vertical partitions are fixedly installed vertically inside the cabinet; a support mechanism is slidably installed on each horizontal partition, and the support mechanism is used to place electrical components; the multiple horizontal and vertical partitions divide the internal space of the cabinet into multiple electrical component storage areas; an air intake fan is fixedly installed on the lower part of one side of the cabinet, and an exhaust fan is fixedly installed on the lower part of the other side of the cabinet. Several S-shaped bends are installed inside the cabinet; an oil supply assembly is fixedly installed on the outside of the cabinet, and the oil supply assembly is connected to each S-shaped bend; several heat absorption boxes are fixedly installed inside the cabinet, and the heat absorption boxes are filled with phase change wax; the heat absorption boxes are used to absorb heat and cool down the inside of the cabinet. The cabinet is equipped with a monitoring system that monitors the operation of the electrical components inside the cabinet and promptly detects any abnormalities.

Images