Intelligent gas-insulated switchgear

Abstract

The invention relates to the technical field of intelligent electrical equipment, in particular to an intelligent inflatable cabinet, which comprises a cabinet body and an air box assembly arranged on the cabinet body, and the air box assembly comprises a sealed box body and a heat dissipation module arranged outside the box body. A super heat dissipation module is arranged in the middle of the heat dissipation module, and when the temperature in the air box assembly exceeds a set starting temperature threshold value, the super heat dissipation module is started; and meanwhile, the pressure change data are stored until the super heat dissipation module stops running. And comparing the stored pressure change data with the pressure change data in a normal state, and if the pressure fluctuations are different, indicating that water vapor or leakage exists in the gas tank assembly. Through the arrangement of the super heat dissipation module, the heat dissipation efficiency of the interior of the gas-insulated switchgear can be accelerated, and whether the interior of the gas tank assembly contains moisture or not can be judged according to the change data of the pressure in the gas tank assembly in a low-temperature intervention state. The device has the advantages that heat dissipation can be achieved, and whether water is contained or not can be detected.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of intelligent electrical equipment, and particularly relates to an intelligent gas-filled cabinet. BACKGROUND

[0002] The gas-filled cabinet is also called gas insulated switchgear (GIS), and is widely used in power systems because the live parts are enclosed in a metal shell, the risk of electric shock and partial discharge is reduced, and the gas-filled cabinet is easy to maintain. The gas-filled cabinet is filled with SF6, nitrogen or mixed gas (such as C4) as an insulating medium. The environment-friendly gas-filled cabinet uses nitrogen or dry air to avoid the problem of greenhouse effect. At present, with the increasing requirement for environmental protection, the power system of some voltage levels is filled with dry air.

[0003] Although the gas-filled cabinet has many advantages and is widely used, it still has some problems. One of the problems is the heat dissipation problem. Because the internal filling gas has a much smaller heat conduction effect than other media, the heat dissipation problem of the gas-filled cabinet must be solved in the use process. Secondly, the filled dry gas is an important prerequisite for insulation, so the increase of the water content in the filled gas and the gas leakage problem will affect the safe operation of the gas-filled cabinet, and even cause a safety operation accident. Therefore, it is an urgent requirement to design a gas-filled cabinet with good heat dissipation and detection function. SUMMARY

[0004] The technical problem to be solved by the application is to design an intelligent gas-filled cabinet which can realize heat dissipation of the internal part of the gas-filled cabinet and detect whether the internal part of the gas-filled cabinet contains moisture.

[0005] The technical solution to the technical problem to be solved by the application is an intelligent gas-filled cabinet, which comprises a cabinet body and a gas tank assembly arranged on the cabinet body, the gas tank assembly comprises a sealed tank body and a heat dissipation module arranged outside the tank body, an ultra-heat dissipation module is arranged in the middle of the heat dissipation module, the ultra-heat dissipation module is started when the temperature inside the gas tank assembly exceeds a set starting temperature threshold, and the pressure change data is saved until the ultra-heat dissipation module stops running. The saved pressure change data is compared with the pressure change data in the normal state, and if the pressure fluctuation is different, it indicates that there is water vapor in the internal part of the gas tank assembly, and an alarm signal is sent.

[0006] Better, the pressure fluctuation amplitude Δ1 = Pc (T1) -Pc (T2) , Δ2 = Ps (T1) -Ps (T2) , wherein Δ1 is the pressure fluctuation amplitude in the normal state, Pc represents the pressure value at each temperature point, and Pc (T1)Pc represents the pressure in the start temperature threshold state (T2) Ps represents the pressure at each temperature point, T2 is the contrast temperature, i.e. the temperature at which the pressure fluctuation amplitude is maximum in the normal state; Δ2 is the pressure fluctuation amplitude in the actual running state.

[0007] Better, the pressure fluctuation data in the two data collection before and after calculation: according to the collected pressure change data, Δ2 = Ps (T1) -Ps (T2) , Δ2 is the pressure fluctuation amplitude in the current running state and Ps represents the pressure at each temperature point, T1 is the start temperature threshold, and T2 is the contrast temperature; according to the saved pressure change data, the pressure fluctuation data Δ1 = Pc (T1) -Pc (T2) , Δ1 is the previous pressure fluctuation amplitude and Pc represents the pressure at each temperature point; when Δ2-Δ1>Δx, it is determined that it contains water vapor, where Δx is the pressure fluctuation limit.

[0008] Better, set the interval days, and perform clustering processing on the saved pressure data after the interval days;

[0009] All saved pressure change data is combined into an intervention state data set; the sample data in the intervention state data set is standardized to obtain a pressure feature data set; the pressure feature data set is clustered; the cluster classes of the two clustering processing results before and after the interval days are compared: if the cluster classes obtained after clustering processing are the same, it means there is no anomaly; if they are different, it means that there is an abnormal pressure fluctuation data, and a water-containing alarm signal is sent.

[0010] Better, first, calculate the overlap matrix, and count the sample overlap ratio of each cluster class after the previous clustering processing and each cluster class after the subsequent clustering processing; then apply the Hungarian algorithm to establish the mapping relationship of each cluster class in the two clustering processing results; if one-to-one, it means that the cluster classes are the same, otherwise they are different.

[0011] Better, extract the abnormal cluster class through simulation test and use the abnormal cluster class to construct a water vapor detection model; after the gas cabinet is running, input the real-time pressure change data collected when the super heat dissipation module is running into the water vapor detection model; the water vapor detection model calculates whether the gas tank assembly contains water in the current state.

[0012] Better, the cooling box is arranged at the position opposite to the super heat dissipation module in the interior of the gas tank assembly, a through hole is arranged at the lower end of the cooling box, and a horn-shaped or cone-shaped gas collecting port with a small upper end and a large lower end is arranged at the through hole.

[0013] Better, a hydrophobic and air-permeable film is arranged at the position of the capillary hole.

[0014] Better, the length direction of the capillary hole is towards the side of the cooling box in contact with the super heat dissipation module.

[0015] Better, the super heat dissipation module is arranged in nine and arranged in a nine-grid mode, and the super heat dissipation module in the middle is arranged with the cooling box at the corresponding position on the inner side of the gas tank assembly.

[0016] The beneficial effects of the present application are that the super heat dissipation module can accelerate the heat dissipation efficiency of the interior of the gas-filled cabinet, and the data of the pressure change in the interior of the gas tank assembly is analyzed and processed to calculate whether the interior of the gas tank assembly contains moisture under the condition of low-temperature intervention. The beneficial effects of realizing heat dissipation and detecting moisture are achieved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 is a schematic diagram of the gas tank assembly in an embodiment of the present application.

[0018] Figure 2 is a schematic diagram of the gas tank assembly in an embodiment of the present application.

[0019] Figure 3 is a schematic diagram of the super heat dissipation module and the cooling box in an embodiment of the present application.

[0020] Figure 4 is Figure 3 is an enlarged view of the circular area.

[0021] In the figure: 100, gas tank assembly; 110, heat dissipation module; 200, cooling box; 210, gas collecting port; 111, super heat dissipation module. DETAILED DESCRIPTION

[0022] In order to make the technical solutions and beneficial effects of the present application clearer, the embodiments of the present application are further explained in detail below.

[0023] An intelligent gas-filled cabinet includes a cabinet body and a gas tank assembly 100 arranged on the cabinet body, wherein the gas tank assembly 100 includes a sealed tank body, a heat dissipation module 110 arranged outside the tank body, and an electrical equipment such as a circuit breaker arranged inside the tank body. The interior of the tank body is filled with dry gas, wherein the dry gas is dry air.

[0024] The heat dissipation module 110 is a heat dissipation sealing plate, which is a sealing plate of the air tank assembly 100 and is provided with fins for heat dissipation. The heat dissipation module 110 is provided with an ultra heat dissipation module 111, which is in close contact with the panel near the air tank assembly 100 to cool the inside of the air tank assembly 100. The ultra heat dissipation module 111 is an electronic refrigeration sheet. The electronic refrigeration sheet (also known as a semiconductor refrigeration sheet) is a device that uses the thermoelectric effect to achieve refrigeration. When a direct current passes through a PN junction composed of two different semiconductor materials (P-type and N-type), the energy level transition of charge carriers will absorb heat at the cold end and release heat at the hot end, forming a temperature difference and transferring heat from one end to the other, ultimately achieving the effect of reducing the temperature at one end. The electronic refrigeration sheet can quickly achieve cooling.

[0025] As known, when the air tank assembly contains water vapor, the pressure inside the air tank assembly 100 will change, but the change range is small, and a long time of accumulation is needed to see the change of the air pressure. In actual application, the pressure data cannot be used as a parameter to detect or predict possible water vapor penetration. Water vapor will condense into ice at low temperature, such as at zero temperature, which will have a greater impact on the pressure inside the air tank assembly 100. Therefore, whether the air tank assembly 100 contains water vapor or how much water vapor it contains is determined by detecting the pressure change caused by intervention.

[0026] Under normal conditions, the heat dissipation module is naturally cooled or cooled by a fan to improve the heat dissipation efficiency. After the gas-filled cabinet is operated, a starting temperature threshold T1 is set. When it is detected that the temperature inside the air tank assembly is greater than the starting temperature threshold T1, the ultra heat dissipation module 111 is started. When the ultra heat dissipation module is started, the change trend of the pressure inside the air tank assembly 100 during the process of cooling the air tank assembly 100 from starting to ending is recorded and saved in the database.

[0027] Under normal conditions, when cooling is performed, the pressure change is Δ1, where the normal condition is a state without water vapor. Under the condition of containing water vapor, the pressure change is Δ2. Due to the effect of the ultra heat dissipation module 110, the water vapor condenses or crystallizes, and the partial pressure of the water vapor inside the air tank assembly 100 decreases, so Δ2> Δ1. Therefore, if the pressure fluctuation is greater than the normal pressure fluctuation, it indicates that water vapor is contained. The temperature when Δ1 is the maximum value under the normal condition can be selected as the comparison temperature T2. Then the pressure fluctuation amplitude when the comparison temperature is reached under the cooling state of the ultra heat dissipation module 111 is extracted, and the two pressure fluctuation values corresponding to the comparison temperature are compared. The reference pressure of the pressure fluctuation amplitude is the pressure when the ultra heat dissipation module 111 reaches the set starting temperature threshold T1. That is, Δ1 = Pc (T1) -Pc (T2) , Δ2 = Ps(T1) -Ps (T2) Wherein Pc is the pressure in normal state, Ps is the pressure in actual collection state. More accurately, when Δ2-Δ1>Δx, it is determined to contain water vapor. Wherein Δx is the pressure fluctuation limit value which can be set as a protection parameter. Wherein the normal state is the initial state or the data collected before leaving the factory and built-in.

[0028] Or, compare the pressure change data collected in the intervention state of the two times of starting the super-heat dissipation module 111. According to the pressure change data collected this time, Δ2=Ps (T1) -Ps (T2) is calculated. According to the saved pressure change data, the pressure fluctuation data Δ1=Pc (T1) -Pc (T2) of the previous data collection is obtained. When Δ2-Δ1>Δx, it is determined to contain water vapor, wherein Δx is the pressure fluctuation limit value.

[0029] In order to improve the accuracy of detection, the pressure change data can also be processed by big data processing to obtain more accurate detection results.

[0030] Wherein the saved data is the pressure series at equal time intervals. In the early stage of operation, only data is saved. After running for 6 months, big data processing is started on the saved pressure change trend data. That is, whether the clustering processing result contains an abnormal pressure fluctuation cluster is detected. If a new abnormal cluster is obtained compared with the cluster in the previous clustering result, it means that the water vapor penetration phenomenon may occur. The specific method includes the following steps.

[0031] After the gas-filled cabinet reaches the site and starts to run, the water content detection and early warning system in the gas tank assembly is started. The system operation includes the following steps:

[0032] Step 1, collect the running data. After the gas-filled cabinet is running, save the pressure data, wherein the saved data includes pressure value and time label. In order to improve the accuracy of prediction, the time interval of sampling in this embodiment is set to 1s to 60s.

[0033] A temperature threshold is set, which is the start temperature threshold T1. When the temperature inside the gas tank assembly is higher than or equal to the start temperature threshold T1, the super heat dissipation module 111 is started and kept running for a set time ty. The time ty is the length of time for keeping the refrigeration running state. Ty can be set according to actual needs, or can be dynamically determined according to the temperature change. The actual refrigeration running state is the interval running state of the super heat dissipation module 111, which is divided into several time periods within the total running time ty. After running for a time period, it stops running for a time period, and then runs for a time period, and so on until the running time ty ends, so as to reduce energy consumption.

[0034] When the super heat dissipation module starts running, the pressure change data under the intervention state is collected to construct the gas tank assembly pressure change data sequence, that is, the pressure change trend data in the ty time period is collected. In this embodiment, the running time of the super heat dissipation module 111 is 15 minutes.

[0035] Further, the end time point of data collection of the gas tank assembly pressure change data sequence under the intervention state is after the super heat dissipation module stops running for a time tb. For example, tb = ty.

[0036] Further, the end time point of data collection of the gas tank assembly pressure change data sequence under the intervention state is after the super heat dissipation module stops running for a time tb. For example, tb = ty.

[0037] Based on the above-mentioned confirmation method of the data collection time point of the gas tank assembly pressure change data sequence under the intervention state, four different lengths of the gas tank assembly pressure change data sequence under the intervention state can be determined. In actual operation, the length of the specific data sequence can be determined according to actual needs.

[0038] The above-mentioned collected gas tank assembly pressure change data sequence under the intervention state is composed of an intervention state data set.

[0039] Step 2, data processing is performed on all data in the intervention state data set. The length of the data sequence is shortened by using a sliding window data processing method. Then, the variance and standard deviation of the data in the sliding window time period are calculated to simplify the data.

[0040] In this embodiment, a 10-minute sliding window processing method is used, in which the step length of sliding is set to 3-5 minutes. Then, the average value and variance value of the data in the sliding window are calculated to form a new feature data sequence. The new feature data sequence is composed of a pressure feature data set.

[0041] Step 3, the cluster classes are obtained by clustering the pressure feature data set by a clustering algorithm. Compare the number of cluster classes before and after clustering. If the number of cluster classes increases, it means that an anomaly has occurred.

[0042] Specifically, clustering processing can be performed once every 10-30 days. Then compare the cluster class results before and after clustering. If the cluster classes obtained after clustering processing are the same, it means that there is no anomaly. If they are different, it means that abnormal data has occurred, indicating that it may contain excessive water vapor.

[0043] Further, in order to better identify abnormal data, the data set is reconstructed. The data within three months after the gassing cabinet is running and the data within 30 days before the current date are taken to form sample data, and a pressure feature data set is constructed, and then clustering processing is performed.

[0044] If the number of cluster classes before and after clustering is different, start the abnormal cluster class determination program. First, calculate the overlap matrix to count the sample overlap ratio of each cluster class after the previous clustering processing and each cluster class after the subsequent clustering processing. Then apply the Hungarian algorithm to establish the mapping relationship of each cluster class in the clustering processing results before and after. Further determine the newly generated abnormal cluster class.

[0045] Before the gassing cabinet is shipped, the abnormal cluster class is extracted in a simulated manner. That is, by using an old and aged sealing gasket packaging test gas tank assembly, or by introducing a specific amount of water vapor to extract the abnormal cluster class. Then use the abnormal cluster class to build a water vapor detection model.

[0046] First, all samples in the abnormal cluster class are summarized, and the water content corresponding to the sample data in the test state is recorded. Then use the abnormal sample data set composed of the samples of the abnormal cluster class to train the water vapor detection model. In actual operation, the trained model is transplanted to the server of the monitoring system or inside the relay protection terminal, by reading the collected pressure parameters, which are pressure change data under the intervention state of the super heat dissipation module 111. Then input the collected pressure change data into the water vapor detection model to calculate the water vapor content.

[0047] If the water vapor content increases during the process of limiting cooling or rapid cooling of the gas tank assembly by the super heat dissipation module, the water vapor will condense at ultra-low temperature, and then the pressure will fluctuate. If there is no water vapor, the pressure fluctuation is small. The higher the water vapor content, the greater the pressure fluctuation. There is no way to detect the water content by changing the properties of the water vapor in the gas tank assembly.

[0048] In the actual test process, the pressure value is also tried to use big data analysis, but when the water vapor content is less, the data processing result needs much more sample data than the data amount in the state of using the super heat dissipation module intervention, that is, the time of water containing operation needs to be long enough to detect the water in the gas tank assembly. After using the super heat dissipation module intervention, the required data amount needs about 5 days of data to achieve detection. Therefore, timely alarm and early warning can be made.

[0049] In order to better intervene the water vapor in the gas tank assembly, a cooling box 200 is arranged in the gas tank assembly 100 at the position opposite to the super heat dissipation module. The cooling box 200 is a cube and is rounded. Alternatively, the cooling box 200 is a cylinder and the edge is rounded. The upper end of the cooling box 200 is tightly attached to the position corresponding to the super heat dissipation module 111 to realize cold conduction. The lower end of the cooling box 200 is provided with a through hole, and the through hole is provided with a horn-shaped or conical gas collecting port 210 with a small upper end and a large lower end. The tip of the gas collecting port 210 abuts against or is integrally formed with the top end of the cooling box 200. The lower end of the gas collecting port 210 is sealingly connected with the edge of the through hole. At this time, a closed space is formed in the cooling box 200. Then, a capillary hole is arranged at the upper end of the gas collecting port to realize air inlet. Further, a hydrophobic and breathable film is covered at the capillary hole, which can pass water vapor but cannot pass water droplets and water mist. Better, the length direction of the capillary hole is towards the side in contact with the super heat dissipation module 111.

[0050] In the process of cooling by the super heat dissipation module 111, the cooling box 200 itself is cooled first. The cooling box 200 itself is made of a material with excellent heat conduction effect. In the state of heat dissipation intervention, it can not only serve as a heat dissipation component to penetrate into the interior of the gas tank assembly 100 to improve the heat dissipation efficiency, but also can collect water vapor to improve the effect of pressure intervention. When the overall temperature of the cooling box 200 decreases, the gas in the cooling box 200 is rapidly cooled, and then the pressure in the cooling box 200 decreases. The gas in the interior of the gas tank assembly 100 enters the interior of the cooling box 200 under the action of pressure. In the process of flowing into the gas, if it contains water vapor, the water vapor will also enter the interior of the cooling box 200. When the temperature of the super heat dissipation module 111 is lower than zero, because the capillary hole is directed to the super heat dissipation module 111, after the water vapor enters the cooling box 200, it first impacts on the surface adjacent to the super heat dissipation module 111 of the cooling box 200, that is, the top surface of the cooling box 200. After the water vapor contacts the low-temperature inner surface of the cooling box 200, condensation or crystallization occurs. At this time, the pressure increment δ caused by the water vapor becomes 0 or close to 0. At this time, the pressure data will fluctuate greatly. Then it can be determined whether there is water vapor by detecting whether there is fluctuation, and the amount of water vapor can be determined by detecting the amplitude of fluctuation. Combined with clustering processing in big data processing and training of prediction model, it can accurately detect whether the gas tank assembly 100 contains water vapor or the amount of water vapor.

[0051] In addition, the overall temperature of the cooling box 200 is low, which forms a cold vortex and can inhale part of the gas, thereby driving the flow of the gas and increasing the overall heat dissipation efficiency of the gas tank assembly 100.

[0052] At the same time, the super heat dissipation module 111 is intermittently operated, that is, it is operated for a period of time, stopped for a period of time, and then operated for a period of time, and so on. Therefore, the pressure in the cooling box 200 will change constantly and show a breathing state. That is, when the super heat dissipation module 111 is running, the air pressure in the cooling box 200 decreases, starts to inhale and changes the inhaled water vapor into liquid or solid state. When the super heat dissipation module 111 stops refrigeration, the temperature of the cooling box 200 rises, starts to exhale, and because the liquid water or solid water vaporizes slowly, it can achieve water locking. Through the cycle of intermittent operation, the water vapor can be collected and locked to the maximum extent. In the process of cooling by the super heat dissipation module 111, the pressure is most intervened, the data fluctuation is the largest, and the accuracy of the calculation results of the clustering processing and the prediction model is improved.

[0053] In the embodiment, the height of the cooling box 200 is set to 1 cm, and the cooling box 200 can also be embedded in the heat dissipation module. At this time, the heat dissipation module 110 is provided with a groove, and the cooling box 200 is embedded in the groove to keep the cooling box 200 flat with the interior of the gas tank assembly 100.

[0054] In order to achieve the effect of cooling better, the super heat dissipation module 111 adopts the nine-square grid mode, that is, nine electronic refrigerating sheets are arranged in the nine-square grid mode and attached to the outside of the air tank assembly. The middle refrigerating sheet is attached to the inside of the air tank assembly corresponding to the cooling box. A cold trap is generated by the peripheral electronic refrigerating sheet, so that the temperature of the middle electronic refrigerating sheet is the lowest, thereby keeping the low temperature state below zero, and better water locking effect is achieved.

[0055] Further, the fan is arranged outside the heat dissipation module 110 to cool the heat dissipation module 110 and provide the super heat dissipation module 111 with a refrigeration space.

[0056] In summary, the above is only the preferred embodiment of the present application, and is not used to limit the scope of the present application. Through the above description, relevant personnel can make various changes and modifications without deviating from the technical idea of the present application. The technical scope of the present application is not limited to the content in the specification. Any equivalent changes and modifications of the shape, structure, features and spirit of the present application shall be included in the scope of the claims of the present application.

Claims

1. An intelligent inflatable cabinet, comprising a cabinet body and an air box assembly (100) arranged on the cabinet body, wherein the air box assembly (100) comprises a sealed cabinet body and a heat dissipation module (110) arranged outside the cabinet body, characterized in that: A super heat dissipation module (111) is provided in the middle of the heat dissipation module (110). When the temperature inside the air box assembly (100) exceeds a set start-up temperature threshold, the super heat dissipation module (111) is started; and at the same time, pressure change data is saved until the super heat dissipation module (111) stops running. The stored pressure change data is compared with the pressure change data under normal conditions. If the pressure fluctuations are different, it indicates that water vapor exists inside the air box assembly (100), and an alarm signal is issued.

2. The intelligent inflatable cabinet according to claim 1, characterized in that: Calculate the pressure fluctuation amplitude Δ1=Pc (T1) -Pc (T2) , Δ2=Ps (T1) -Ps (T2) , Among them, Δ1 is the pressure fluctuation amplitude under normal conditions, and Pc is the pressure value at each temperature point. (T1) Indicates the pressure at the start temperature threshold state, Pc (T2) The pressure value when the pressure fluctuation amplitude is the largest under normal conditions, T2 is the comparison temperature, i.e., the temperature when the pressure fluctuation amplitude is the largest under normal conditions; Δ2 is the pressure fluctuation amplitude under actual operating conditions, and Ps is the pressure value at each temperature point. When Δ2-Δ1>Δx, it is determined to contain water vapor, where Δx is the pressure fluctuation limit.

3. The intelligent inflatable cabinet according to claim 1, characterized in that: Calculate the pressure fluctuation data before and after the two data collections: Calculate Δ2=Ps based on the collected pressure change data (T1) -Ps (T2) , Δ2 is the pressure fluctuation amplitude under the current operating state and Ps represents the pressure value at each temperature point, T1 is the starting temperature threshold, and T2 is the comparison temperature; Obtain the pressure fluctuation data Δ1=Pc during the previous data acquisition based on the saved pressure change data (T1) -Pc (T2) , where Δ1 is the amplitude of the previous pressure fluctuation and Pc represents the pressure value at each temperature point; When Δ2-Δ1>Δx, it is determined to contain water vapor, where Δx is the pressure fluctuation limit.

4. The intelligent inflatable cabinet according to claim 1, characterized in that: Set the interval number of days and perform clustering on the saved pressure data after the interval number of days; All the saved pressure change data are combined into an intervention state data set; the sample data in the intervention state data set are standardized to obtain a pressure feature data set; clustering is performed based on the pressure feature data set; Compare the clusters of the two clustering processing results before and after the interval of several days: if the clusters obtained after clustering processing are the same, it means there is no abnormality; If they are different, it indicates that abnormal pressure fluctuations have occurred, and a water alarm signal will be issued.

5. The intelligent inflatable cabinet according to claim 4, characterized in that: First, the overlap matrix is ​​calculated to count the sample overlap ratios of each cluster after the previous clustering process and the next clustering process. Then, the Hungarian algorithm is applied to establish the mapping relationship between the clusters in the results of the two clustering processes. If there is a one-to-one correspondence, the clusters are the same, otherwise they are different.

6. The intelligent inflatable cabinet according to claim 1, characterized in that: Through simulation experiments, abnormal clusters are extracted and used to build a water vapor detection model; After the inflatable cabinet is in operation, the real-time pressure change data collected during the operation of the super heat dissipation module (111) is input into the water vapor detection model; The water vapor detection model calculates whether there is water inside the air box assembly (100) in the current state.

7. The intelligent inflatable cabinet according to claim 1, characterized in that: A cooling box (200) is provided inside the air box assembly (100) at a position opposite to the super heat dissipation module (111). A through hole is provided at the lower end of the cooling box (200), and a trumpet-shaped or cone-shaped air collecting port (210) with a small upper end and a large lower end is installed at the through hole; the tip of the air collecting port (210) abuts against the top of the cooling box (200) or is integrally formed; the lower end of the air collecting port (210) is sealed and connected to the edge of the through hole; and capillaries are provided at the upper end of the air collecting port (210) for ventilating the interior of the cooling box (200).

8. The intelligent inflatable cabinet according to claim 1, characterized in that: The capillary pores are covered with a hydrophobic breathable membrane.

9. The intelligent inflatable cabinet according to claim 8, characterized in that: The length direction of the capillary pores is toward the side of the cooling box (200) that contacts the super heat dissipation module (111).

10. The intelligent inflatable cabinet according to claim 8, characterized in that: Nine super heat dissipation modules (111) are provided and arranged in a nine-square grid pattern, wherein a cooling box (200) is provided at a corresponding position on the inner side of the air box assembly of the middle super heat dissipation module (111).

Images