Electrical control cabinet and operating temperature control method

Abstract

The application discloses an electrical control cabinet and a running temperature control method, relates to the technical field of electrical control cabinets, and discloses a cabinet body, wherein a forced air cooling system, a liquid cooling system, a compression refrigeration system and a running temperature controller are arranged in the cabinet body. The running temperature controller comprises a running temperature monitoring unit, an electrical parameter mapping unit and a running temperature control execution unit. Through the design of the electrical control cabinet and the running temperature control method, the temperature characteristics of each monitoring point in the electrical control cabinet can be predicted in advance. In combination with multi-physical field simulation of a virtual control cabinet, a plurality of different types of active heat dissipation systems are matched, the plurality of types of active heat dissipation systems are freely and cooperatively combined based on virtual cabinet simulation, the capability limitation of a single heat dissipation mode is broken through, and the electrical control cabinet is reasonably controlled in running temperature while the cabinet body temperature is advancedly regulated and controlled.

Description

Technical Field

[0001] This invention relates to the field of electrical control cabinet technology, and more specifically, to an electrical control cabinet and a method for controlling operating temperature. Background Technology

[0002] As a core component in industrial automation, new energy equipment, and other fields, the temperature control performance of electrical control cabinets directly affects the reliability and lifespan of equipment. Traditional temperature control technologies mainly rely on a single heat dissipation method: early methods primarily used natural air cooling, achieving heat convection through cabinet ventilation holes, but this was significantly inefficient in scenarios with power densities ≥50W / cm². With the increasing application of high-power devices, active cooling technologies such as liquid cooling and compression refrigeration have been widely adopted.

[0003] However, in existing technologies, different active cooling systems typically operate independently, failing to form a collaborative mechanism for optimal energy efficiency. For example, while existing technologies propose combinations of air cooling and semiconductor cooling, they lack dynamic switching strategies, resulting in energy waste exceeding 30% in scenarios with fluctuating loads. Secondly, traditional thermal design heavily relies on engineers' experience, adjusting parameters such as fan speed and liquid cooling flow rate through a "trial and error" approach, leading to development cycles that can last for months and making it difficult to handle complex scenarios involving multi-physics coupling (such as fluid-thermal-solid).

[0004] Based on the above, this invention proposes an electrical control cabinet and a method for controlling operating temperature. Summary of the Invention

[0005] In view of the shortcomings of the existing technology, the purpose of this invention is to provide an electrical control cabinet and a method for controlling operating temperature.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] An electrical control cabinet includes a cabinet body, in which an air-cooled system, a liquid-cooled system, a compression refrigeration system, and an operating temperature controller are arranged. The operating temperature controller includes an operating temperature monitoring unit, an electrical parameter mapping unit, and an operating temperature control execution unit.

[0008] The operating temperature monitoring unit is used to continuously generate time-series temperature sets for each temperature monitoring point during the operation of the electrical control cabinet;

[0009] The electrical parameter mapping unit is used to determine the predicted temperature characteristics of each temperature monitoring point and to establish a parameter mapping mechanism between the predicted temperature characteristics of each temperature monitoring point and the virtual electrical control cabinet.

[0010] The temperature control execution unit is used to select a temperature control scheme to control the temperature of the cabinet.

[0011] Furthermore, a method for controlling the operating temperature of an electrical control cabinet includes the following steps:

[0012] Step 1: During the operation of the electrical control cabinet, continuously acquire temperature data from each temperature monitoring point inside the cabinet, and continuously generate time-series temperature sets for each temperature monitoring point;

[0013] Step 2: Generate time-series temperature sets for each temperature monitoring point and determine the predicted temperature characteristics for each temperature monitoring point.

[0014] Step 3: Establish a mapping mechanism between the predicted temperature characteristics of each temperature monitoring point and the parameters of the virtual electrical control cabinet;

[0015] Step 4: Obtain the overall feedback value of each temperature control scheme, and execute the temperature control scheme with the highest overall feedback value to control the temperature of the cabinet.

[0016] Furthermore, the continuous generation method of the time-series temperature set of temperature monitoring points is as follows: whenever a temperature monitoring point acquires a temperature data, it synchronously collects the i-1 temperature data acquired previously, and integrates the i temperature data into a time-series temperature set in the form of a time-series set.

[0017] Furthermore, the method for determining the predicted temperature characteristics of temperature monitoring points is as follows: whenever a temperature monitoring point generates a time series temperature set, the time series temperature set is input into the temperature feature prediction model corresponding to that temperature monitoring point, and the temperature feature prediction model outputs the predicted temperature characteristics of that temperature monitoring point.

[0018] Furthermore, the method for obtaining the comprehensive feedback value of the temperature control scheme is as follows: Select a temperature control scheme, start all active cooling systems included in the virtual electrical control cabinet, and control the virtual electrical control cabinet to run for a duration of T. After the operation ends, calculate the average temperature performance value, and calculate the ratio of the average temperature performance value to the total power consumption of all active cooling systems to obtain the comprehensive feedback value of the temperature control scheme.

[0019] Furthermore, the calculation method for the average temperature performance value is as follows: determine the temperature characteristics of each temperature monitoring point in the virtual electrical control cabinet, import the temperature characteristics of each temperature monitoring point into the temperature characteristic analysis model in sequence, determine the temperature performance value of each temperature monitoring point in sequence, sum and average the temperature performance values ​​of each temperature monitoring point, and calculate the average temperature performance value.

[0020] Furthermore, the construction method of the temperature feature analysis model is as follows: collect multiple temperature features, construct a neural network model, use the temperature features as basic data, train the neural network model, assign a temperature performance value to each temperature feature, the larger the temperature performance value, the better the temperature feature performance, and finally, construct the temperature feature analysis model.

[0021] Furthermore, each temperature monitoring point corresponds to a temperature feature prediction model, and each temperature feature prediction model is built based on an LSTM model, and each temperature feature prediction model is updated regularly.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] This invention, through the design and operation temperature control method of electrical control cabinets, can predict the temperature characteristics of each monitoring point in the electrical control cabinet in advance. Combined with multiphysics simulation of virtual control cabinets and multiple active cooling systems of different types, it enables free and coordinated combination of multiple types of active cooling systems based on virtual cabinet simulation, breaking through the limitations of a single cooling mode. While achieving advance regulation of cabinet temperature, it also enables reasonable operation temperature control of the electrical control cabinet. This constructs a closed-loop intelligent temperature control system of "data-driven prediction - physical simulation verification - energy efficiency optimization decision-making", realizing a technological leap from static design relying on experience rules to dynamic optimization driven by digital twins, significantly improving the temperature stability, energy efficiency ratio and reliability of complex electrical control cabinet operation. Attached Figure Description

[0024] Figure 1 A flowchart illustrating a method for temperature control during the operation of an electrical control cabinet;

[0025] Figure 2 This is a diagram of the internal structure of the electrical control cabinet.

[0026] 100. Cabinet; 200. Air-cooled system; 300. Liquid-cooled system; 400. Compression refrigeration system; 500. Operating temperature controller. Detailed Implementation

[0027] Example 1: Refer to Figure 1 A method for temperature control during the operation of an electrical control cabinet, comprising the following steps:

[0028] Step 1: During the operation of the electrical control cabinet, continuously acquire temperature data from each temperature monitoring point within the cabinet (the temperature monitoring points are arranged according to the size of the cabinet and the heating characteristics of the components, such as near the heating components, air duct inlets and outlets, and cabinet corners; "continuous acquisition" means continuous collection using high-frequency words (usually ≥1 time / second), and continuously generate time-series temperature sets for each temperature monitoring point;

[0029] The continuous generation method of time-series temperature sets of temperature monitoring points: Whenever a temperature monitoring point acquires a temperature data, it synchronously collects the i-1 temperature data acquired previously, and integrates the i temperature data into a time-series temperature set in the form of a time-series set.

[0030] Step 2: Generate time-series temperature sets for each temperature monitoring point and determine the predicted temperature characteristics for each temperature monitoring point.

[0031] The method for determining the predicted temperature characteristics of a temperature monitoring point is as follows: Whenever a temperature monitoring point generates a time series temperature set, the time series temperature set is input into the temperature feature prediction model corresponding to the temperature monitoring point, and the temperature feature prediction model outputs the predicted temperature characteristics of the temperature monitoring point.

[0032] Each temperature monitoring point corresponds to a temperature feature prediction model (e.g., if temperature monitoring points A, B, C, D, and E are set inside cabinet 100, then there are a total of five temperature feature prediction models). Each temperature feature prediction model is built based on an LSTM model, and each temperature feature prediction model is updated regularly. In this embodiment, taking temperature monitoring point A as an example, the construction method of the temperature feature prediction model corresponding to temperature monitoring point A is disclosed: collect all time-series temperature sets generated by temperature monitoring point A (to ensure the integrity and accuracy of the data), build an LSTM model, and divide the time-series temperature sets into training set, validation set, and test set according to a ratio of 70%:15%:15%. To ensure the continuity of the time series during data segmentation and prevent future information leakage into the training set, basic features are extracted from each time series temperature data, such as average temperature, temperature extremes (highest and lowest temperatures), and temperature change rate. An LSTM model is selected as the prediction model (model structure design: Input layer: Determine the dimension of the input layer based on the results of feature engineering. For example, if temperature data from the past 24 hours is used as input, and three features are extracted for each time point: average temperature, temperature extremes, and temperature change rate, then the input layer dimension is 24*3=72. Hidden layer: Design the LSTM hidden layer, including the number of LSTM units and the number of stacked layers. The optimal hidden layer structure can be determined experimentally. Output layer: The output layer should contain predicted temperature features, including the average temperature, temperature extremes, and temperature change rate within the prediction time period T. The dimension of the output layer should match the number of predicted features). A suitable loss function (e.g., Mean Squared Error, MSE) and optimizer (e.g., Adam) are selected, and the model is compiled. The number of training epochs and batch size are set. Hyperparameters such as size are used to train the model using a training set. The model parameters are updated using the backpropagation algorithm. During training, the model performance is monitored using a validation set to prevent overfitting. Based on the performance feedback from the validation set, the model structure or hyperparameters are adjusted to optimize the model performance. The trained model is evaluated using a test set, and error metrics (such as root mean square error RMSE, mean absolute error MAE, etc.) between the predicted results and the actual values ​​are calculated. The predictive performance of the model is analyzed to ensure that the model can maintain good prediction accuracy on unseen data, thus constructing a temperature feature prediction model for temperature monitoring point A.

[0033] Step 3: Establish a mapping mechanism between the predicted temperature characteristics of each temperature monitoring point and the parameters of the virtual electrical control cabinet (e.g., the coordinates of temperature monitoring point A are (X=90mm, Y=60mm, Z=50mm)).

[0034] #Example: Parameter structure of temperature monitoring point A

[0035] pointA_params={

[0036] "coordinate":(90,60,50), # 3D coordinates in mm

[0037] "pred_features": {

[0038] "avg_temp": [25.0, 26.5, 28.0, ...], # Second-by-second average temperature sequence over time T

[0039] "temp_extremes": [(24.5, 29.8), (25.1, 30.2), ...], # Highest / lowest temperature every 10 seconds

[0040] "change_rate": [0.3, 0.5, -0.2, ...] # Rate of temperature change per second (°C / s)

[0041] },

[0042] "time_stamp": "2025-02-11 14:30:00" # Predicted start time}).

[0043] Constructing a Virtual Electrical Control Cabinet: Step S1: Construct a 3D model of the control cabinet (using SolidWorks or ANSYS Design Modeler; cabinet dimensions: typical dimensions (width × depth × height) 600mm × 800mm × 2000mm, considering ventilation holes, cable entry points, etc.); internal component layout: heating elements: IGBT module (power consumption 1000W), power supply module (power consumption 500W), PLC controller (power consumption 50W); cooling components: air cooling system: top axial fan (airflow 20CFM), baffle plate; liquid cooling system: cold plate (material Al6061, flow channel cross-sectional area 5mm × 10mm), circulating pump (flow rate 2L / min); compression refrigeration: evaporator (fin spacing 2mm), condenser (fan speed 2000rpm); sealing structure: dustproof rubber strip (thermal conductivity 0.15W / m・K)).

[0044] Step S2: Define material properties (e.g., if the component is a cabinet, its material is cold-rolled steel plate, define the thermal conductivity as 45W / m・K, the specific heat capacity as 480J / kg・K, and the density as 7850kg / m³; if the component is an IGBT module, its material is Al2O3 ceramic, define the thermal conductivity as 20W / m・K, the specific heat capacity as 750J / kg・K, and the density as 3900kg / m³).

[0045] Step S3: Perform simulations of the air-cooled system, liquid-cooled system, and compression refrigeration system.

[0046] Simulation of air-cooled system:

[0047] Luent settings: Turbulence model: SST k-ω (suitable for flow near the wall);

[0048] Inlet conditions: Fan outlet speed 3m / s (corresponding to 120CFM airflow);

[0049] Export conditions: Pressure outlet (0 Pa);

[0050] Heat source: IGBT module heat flux density 20W / cm² (calculated using Joule heat formula Q=I²R);

[0051] Convection coefficient: natural convection on cabinet surface (5W / m²・K);

[0052] Liquid cooling system simulation:

[0053] COMSOL settings:

[0054] Physical field: Laminar flow + heat transfer in solids;

[0055] Coolant inlet: temperature 25℃, flow rate 2L / min;

[0056] Contact between cold plate and component: thermal contact resistance 0.5K・cm² / W;

[0057] Optimization objective: Maximize heat exchange (Q) while minimizing pump power (P), using dimensionless processing;

[0058] Simulation of a compression refrigeration system:

[0059] ANSYS Simplorer co-simulation:

[0060] Drive circuit modeling:

[0061] IGBT inverter: using the IRF3710 model, switching frequency 20kHz;

[0062] Control algorithm: Digital incremental PID, temperature control accuracy ±0.5℃;

[0063] Compressor modeling:

[0064] ANSYS Maxwell: Coupling of electromagnetic force calculation and piston displacement;

[0065] Displacement: 10cc / rev, speed 3000rpm;

[0066] Evaporator / condenser modeling:

[0067] Evaporator: Air-side convection coefficient 50W / m²・K, refrigerant-side phase change temperature 5℃;

[0068] Condenser: Fan speed 2000rpm, ambient temperature 35℃;

[0069] Multiphysics coupling:

[0070] Refrigerant pressure-temperature relationship: using the NIST REFPROP database;

[0071] Evaporator and cabinet coupling: through heat flux density transfer;

[0072] Step S4: Complete the construction of the virtual electrical control cabinet through verification and experimental comparison.

[0073] Step 4: Obtain the comprehensive feedback value of each temperature control scheme (the temperature control scheme is a random combination of air-cooled system 200, liquid-cooled system 300, and compression refrigeration system 400. If there are other active refrigeration systems, there will be more temperature control schemes. The temperature is controlled according to the combination results. Each temperature control scheme has a unique combination. For example, the first temperature control scheme only has air-cooled system 200, the second only has liquid-cooled system 300, the third only has compression refrigeration system 400, and the fourth includes both air-cooled system 200 and compression refrigeration system 400). The fifth temperature control scheme includes liquid cooling system 300 and compression refrigeration system 400. Different temperature control schemes correspond to different typical parameter configurations. For example, in the first temperature control scheme, the fan speed of air cooling system 200 is 120 CFM. In the second temperature control scheme, the flow rate of liquid cooling system 300 is 2 L / min and the coolant temperature is 25℃. In the fourth temperature control scheme, the fan of air cooling system 200 is set to 80% speed and the compressor of compression refrigeration system 400 is set to 60% power. The temperature control scheme with the largest comprehensive feedback value is used to control the temperature of cabinet 100.

[0074] The method for obtaining the comprehensive feedback value of the temperature control scheme is as follows: Select a temperature control scheme, start all active cooling systems included in the virtual electrical control cabinet according to the corresponding typical parameter configuration, and control the virtual electrical control cabinet to run for a duration of T. After the operation ends, determine the temperature characteristics of each temperature monitoring point in the virtual electrical control cabinet (the location of the temperature monitoring point in the virtual electrical control cabinet, i.e., the actual coordinates of the temperature monitoring point within the cabinet 100), and simultaneously determine the total power consumption of all active cooling systems in the virtual electrical control cabinet (i.e., the power consumption of the air-cooled system 200, liquid-cooled system 200). The sum of the power consumption of system 300 and compression refrigeration system 400 (or the sum of the power consumption of liquid cooling system 300 and compression refrigeration system 400 if air-cooled system 200 is not started) is calculated. The temperature characteristics of each temperature monitoring point are sequentially imported into the temperature characteristic analysis model, and then the temperature performance value of each temperature monitoring point is determined sequentially. The average temperature performance value is calculated by summing and averaging the temperature performance values ​​of each temperature monitoring point. The average temperature performance value is then calculated as a ratio to the total power consumption of all active cooling systems (the ratio calculation is dimensionless). The comprehensive feedback value of the temperature control scheme is then calculated.

[0075] The construction method of the temperature feature analysis model is as follows: Multiple temperature features are collected, a neural network model is built, and the multiple temperature features are divided into training, validation, and test sets in a 60%:20%:20% ratio. The neural network model is trained using the temperature features as the basic data. Each temperature feature is assigned a temperature performance value, with the value ranged from 1 to 30. The temperature performance values ​​are then calibrated for physical meaning. `def map_to_performance(temp_feature, feature_type):`

[0076] "Mapping to the 1-30 range based on feature type"

[0077] if feature_type == "avg_temp":

[0078] # The lower the temperature, the higher the performance value (assuming an ideal temperature of 25℃).

[0079] performance = 30 - min(30, max(1, int((temp_feature - 25) *2)))

[0080] elif feature_type == "change_rate":

[0081] # The lower the rate of change, the higher the performance value.

[0082] performance = 30 - min(30, max(1, int(change_rate * 10)))

[0083] The function `return max(1, min(30, performance))` returns `max(1, min(30, performance))`. The magnitude of the temperature performance value has a clear meaning; the larger the value, the better the temperature feature performance. The neural network model is repeatedly trained using the training set, and the performance of the training phase is verified using the validation set. The model parameters are adjusted in a timely manner based on the validation results. Hyperparameter tuning, overfitting prevention, and training monitoring are adopted. The final model is evaluated using a test set that was not involved in the training to ensure that the results do not depend on data snooping during the training process. Finally, a temperature feature analysis model is constructed.

[0084] The above method uses an LSTM neural network to predict the temperature characteristics of each monitoring point in the electrical control cabinet in advance. Combined with multiphysics simulation of the virtual control cabinet and multiple active cooling systems of different types, it enables free and coordinated combination of multiple types of active cooling systems based on virtual cabinet simulation. This breaks through the limitations of a single cooling mode and achieves advance temperature control of the cabinet while controlling the operating temperature of the electrical control cabinet in a reasonable manner. This constructs a closed-loop intelligent temperature control system of "data-driven prediction - physical simulation verification - energy efficiency optimization decision-making". It realizes a technological leap from static design relying on empirical rules to dynamic optimization driven by digital twins, and significantly improves the temperature stability, energy efficiency ratio and reliability of complex electrical control cabinet operation.

[0085] Example 2: Refer to Figure 2 An electrical control cabinet includes a cabinet 100, wherein an air-cooled system 200, a liquid-cooled system 300, a compression refrigeration system 400, and an operating temperature controller 500 are arranged inside the cabinet 100. The operating temperature controller 500 includes an operating temperature monitoring unit, an electrical parameter mapping unit, and an operating temperature control execution unit.

[0086] The operating temperature monitoring unit is used to continuously generate time-series temperature sets for each temperature monitoring point during the operation of the electrical control cabinet.

[0087] The electrical parameter mapping unit is used to determine the predicted temperature characteristics of each temperature monitoring point and to establish a parameter mapping mechanism between the predicted temperature characteristics of each temperature monitoring point and the virtual electrical control cabinet.

[0088] The temperature control execution unit is used to select a temperature control scheme to control the temperature of cabinet 100.

[0089] Among them, the air-cooled system 200, the liquid-cooled system 300, and the compression refrigeration system 400 are different active heat dissipation systems, which differ from each other in terms of heat dissipation capacity, power consumption characteristics, and temperature control accuracy. This invention only illustrates the above three active heat dissipation systems. In actual applications, other active heat dissipation systems, such as refrigeration semiconductor systems, can also be installed in the electrical control cabinet.

[0090] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0091] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0092] It should be understood that in the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0093] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0094] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0095] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0096] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0097] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method of operating temperature control of an electrical control cabinet, characterized in that, Including cabinet (100), the air cooling system (200), liquid cooling system (300), compression refrigeration system (400) are arranged in the cabinet (100), operation temperature controller (500), the operation temperature controller (500) includes operation temperature monitoring unit, electrical parameter mapping unit, operation temperature control execution unit; Operation temperature monitoring unit is used to generate the time series temperature set of each temperature monitoring point during the operation of electrical control cabinet continuously; The electrical parameter mapping unit is used to determine the predicted temperature characteristics of each temperature monitoring point, and the parameter mapping mechanism of the predicted temperature characteristics of each temperature monitoring point and the virtual electrical control cabinet is established; The operation temperature control execution unit is used to select the temperature control collocation scheme to control the temperature of the cabinet (100); The operation temperature control method of the electrical control cabinet, the steps are as follows: Step one: during the operation of electrical control cabinet, the temperature data of each temperature monitoring point in the cabinet (100) is continuously obtained, and the time series temperature set of each temperature monitoring point is continuously generated; Step two: the predicted temperature characteristics of each temperature monitoring point are determined each time the time series temperature set of each temperature monitoring point is generated; Step three: the parameter mapping mechanism of the predicted temperature characteristics of each temperature monitoring point and the virtual electrical control cabinet is established; Step four: the temperature control collocation comprehensive feedback value of each temperature control collocation scheme is obtained, and the temperature control collocation scheme with the maximum temperature control collocation comprehensive feedback value is executed to control the temperature of the cabinet (100); the temperature control collocation scheme is randomly combined and collocated with the air cooling system (200), the liquid cooling system (300) and the compression refrigeration system (400); The temperature control collocation comprehensive feedback value of the temperature control collocation scheme is obtained in the following way: select a temperature control collocation scheme, start all active heat dissipation systems contained in the temperature control collocation scheme in the virtual electrical control cabinet, control the virtual electrical control cabinet to run for T length of time, calculate the average temperature performance value after the running is completed, and calculate the temperature control collocation comprehensive feedback value of the temperature control collocation scheme by ratio calculation of the average temperature performance value and the total power consumption of all active heat dissipation systems.

2. The method of claim 1, wherein the temperature of the electrical control cabinet is controlled by the controller. The time series temperature set of the temperature monitoring point is continuously generated in the following way: whenever a temperature monitoring point obtains a temperature data, i-1 temperature data continuously obtained previously are collected synchronously, and the i temperature data are integrated into the time series temperature set in the form of time series set.

3. The method of claim 1, wherein the temperature of the electrical control cabinet is controlled by a temperature control device. The predicted temperature characteristics of the temperature monitoring point are determined in the following way: whenever a time series temperature set of a temperature monitoring point is generated, the time series temperature set is input into the temperature characteristic prediction model corresponding to the temperature monitoring point, and the predicted temperature characteristics of the temperature monitoring point are output by the temperature characteristic prediction model.

4. The method of claim 1, wherein the temperature of the electrical control cabinet is controlled by a temperature control device. The average temperature performance value is calculated in the following way: the temperature characteristics of each temperature monitoring point in the virtual electrical control cabinet are determined, the temperature characteristics of each temperature monitoring point are sequentially input into the temperature characteristic analysis model, the temperature performance value of each temperature monitoring point is sequentially determined, the temperature performance values of each temperature monitoring point are summed and averaged, and the average temperature performance value is calculated.

5. The method of claim 4, wherein the temperature of the electrical control cabinet is controlled by the controller. The temperature characteristic analysis model is constructed in the following manner: a plurality of temperature characteristics are collected, a neural network model is constructed, the temperature characteristics are taken as basic data, the neural network model is trained, a temperature performance value is given to each temperature characteristic, and the greater the temperature performance value, the better the temperature characteristic performance. Finally, the temperature characteristic analysis model is constructed.

6. The method of claim 1, wherein the temperature of the electrical control cabinet is controlled by a temperature control device. Each temperature monitoring point corresponds to a temperature characteristic prediction model, and each temperature characteristic prediction model is constructed based on an LSTM model and is regularly updated.

Images