Energy storage cabinet thermal management system based on dynamic wind pressure adjustment

Abstract

The invention discloses an energy storage cabinet thermal management system based on dynamic wind pressure adjustment, and relates to the technical field of thermal management. The energy storage cabinet thermal management system based on dynamic wind pressure adjustment comprises a data acquisition module used for continuously acquiring operation condition data and in-cabinet thermal imaging data of a set energy storage cabinet at each time point; the working condition thermal load analysis module is used for analyzing thermal risk assessment features based on the operation data; the thermal field identification and analysis module is used for analyzing thermal field disorder characteristics by a pre-trained in-cabinet thermal identification model; the thermal regulation characteristic generation module analyzes thermal regulation characteristic values based on thermal risks and thermal field disorder characteristics, and dynamic wind pressure regulation processing is performed on the set energy storage cabinet through the thermal regulation characteristic values in the dynamic wind pressure regulation module, so that the system can sense the change trend of heat transfer of the cabinet on the overall level; and the heat dissipation requirement is dynamically matched, so that the condition of uneven heat dissipation or cooling lag is avoided, and the thermal safety of the cabinet is remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of thermal management, in particular to an energy storage cabinet thermal management system based on dynamic air pressure regulation. BACKGROUND

[0002] As an important unit of electrochemical energy storage system, the energy storage cabinet will continuously generate heat in the internal components during operation. If the heat cannot be effectively discharged, it will affect the stable operation of the cabinet and the service life of the equipment. At present, the thermal management of the energy storage cabinet is mainly based on air cooling, that is, air flow is generated by a fan to take away the heat inside the cabinet. Common fan control methods include start-stop control based on temperature threshold, etc. With the continuous improvement of the capacity of the energy storage system and the increasing complexity of the operating conditions, the thermal management system gradually develops towards intelligence and dynamics, gradually evolving into dynamic air pressure regulation.

[0003] The prior art such as the patent application with the publication number CN119376464B discloses a thermal management control method and system for energy storage power supply. The method includes: obtaining real-time temperature data of the energy storage power supply and performing heat source distribution analysis to obtain corresponding temperature distribution information; obtaining heat dissipation operating parameters of the energy storage power supply, combining the temperature distribution information to perform adjustment analysis, and obtaining corresponding initial heat dissipation control strategy; obtaining real-time monitoring data of the energy storage power supply, combining the temperature distribution information and the initial heat dissipation control strategy to perform charge-discharge dynamic analysis, and obtaining corresponding charge-discharge adjustment information; identifying hot spots based on the real-time temperature data and the heat management strategy to obtain corresponding potential hot spot information and perform early warning processing to obtain corresponding hot spot early warning result; and performing comprehensive control analysis on the heat management strategy, the initial heat dissipation control strategy and the hot spot early warning result to obtain a corresponding global heat management control scheme. The present application can comprehensively and intelligently control the thermal management of the energy storage power supply.

[0004] Based on the above scheme discovery, the limitations of the prior art at least include the following problems, the prior art lacks global recognition and adaptive adjustment of the internal thermal field distribution and dynamic heat dissipation state of the energy storage cabinet, the internal components of the energy storage cabinet continuously release heat during actual operation, the internal temperature field shows uneven time and space distribution, and shows complex dynamic evolution characteristics with load changes and cooling condition differences, the prior art is difficult to reveal the blocking of the heat transfer path inside the cavity, and it is also difficult to capture the signs of local temperature abnormalities expanding gradually in the early stage, resulting in insufficient heat dissipation balance, at the same time, the operating environment of the energy storage cabinet is changing, for example, the increase of external temperature will increase the overall heat dissipation burden, the air humidity fluctuation may affect the heat exchange efficiency, and the dust deposition may cause local poor ventilation, the interaction of these internal and external factors makes the heat management state have high dynamicity and uncertainty, if the adjustment mode still stays in the fixed or lagging control level, it is difficult to match the actual heat dissipation demand in time, which may cause the coexistence of high energy consumption and insufficient cooling, thereby weakening the long-term operation reliability of the energy storage cabinet. SUMMARY

[0005] In view of the deficiencies of the prior art, the present application provides an energy storage cabinet thermal management system based on dynamic air pressure regulation, which solves the problem that the prior art is difficult to realize global recognition and dynamic adaptive adjustment of the thermal field of the energy storage cabinet, and is prone to heat dissipation imbalance.

[0006] To achieve the above purpose, the present application realizes the following technical scheme: an energy storage cabinet thermal management system based on dynamic air pressure regulation, comprising: a data acquisition module for continuously acquiring operating condition data and cabinet internal thermal imaging data of each time point of a set energy storage cabinet; a working condition thermal load analysis module for analyzing the cabinet comprehensive thermal load characteristic value of the corresponding time point based on the operating condition data of each time point of the set energy storage cabinet, and analyzing the cabinet thermal risk assessment characteristic value of the set energy storage cabinet; a thermal field recognition analysis module for analyzing the thermal field disorder characteristic value of the set energy storage cabinet based on the pre-trained cabinet internal thermal recognition model and the cabinet internal thermal imaging data of each time point; a thermal regulation characteristic generation module for analyzing the thermal regulation characteristic value of the set energy storage cabinet based on the cabinet thermal risk assessment characteristic value and the thermal field disorder characteristic value of the set energy storage cabinet; and a dynamic air pressure regulation module for performing dynamic air pressure regulation processing on the set energy storage cabinet based on the thermal regulation characteristic value.

[0007] Further, the operation condition data includes an air flow circulation efficiency factor, a cavity heat accumulation value, a cavity dust deposition distribution index, a cavity heat and humidity accumulation factor, an electromagnetic radiation noise intensity value, and a cabinet vibration response amplitude value. The specific steps of analyzing the specific cabinet comprehensive thermal load characteristic value of the set energy storage cabinet at each time point are as follows: based on the operation condition data of the set energy storage cabinet at each time point, analyzing the thermal risk characteristic set at the corresponding time point, including a cabinet heat dissipation degradation characteristic value and a structure and electrical stability characteristic value; obtaining the external thermal load characteristic value of the set energy storage cabinet at each time point, and combining the thermal risk characteristic set to analyze the cabinet comprehensive thermal load characteristic value at the corresponding time point.

[0008] Further, the specific steps of analyzing the thermal risk characteristic set of the set energy storage cabinet at each time point are as follows: based on the air flow circulation efficiency factor, the cavity heat accumulation value, the cavity dust deposition distribution index, and the cavity heat and humidity accumulation factor of the set energy storage cabinet at each time point, analyzing the cabinet heat dissipation degradation characteristic at the corresponding time point; based on the electromagnetic radiation noise intensity value and the cabinet vibration response amplitude value of the set energy storage cabinet at each time point, analyzing the structure and electrical stability characteristic value at the corresponding time point.

[0009] Further, the specific steps of obtaining the external thermal load characteristic value of the set energy storage cabinet at each time point are as follows: obtaining the external environment data of the set energy storage cabinet at each time point, and performing standardization processing; based on the external environment data of the set energy storage cabinet at each time point after standardization processing, analyzing the external thermal load characteristic value at the corresponding time point.

[0010] Further, the specific steps of analyzing the cabinet thermal risk evaluation characteristic value of the set energy storage cabinet are as follows: based on the cabinet comprehensive thermal load characteristic value of the set energy storage cabinet at each time point, analyzing the load characteristic set of the set energy storage cabinet; based on the load characteristic set of the set energy storage cabinet, analyzing the cabinet thermal risk evaluation characteristic value of the set energy storage cabinet.

[0011] Further, the cabinet internal thermal imaging data specifically includes a temperature value and a two-dimensional coordinate of each pixel point in the thermal imaging, and the cabinet internal thermal identification model includes a feature extraction sub-network and a time sequence correlation sub-network.

[0012] Further, the specific steps of analyzing the thermal field disorder characteristic value of the set energy storage cabinet are as follows: inputting the cabinet internal thermal imaging data of the set energy storage cabinet at each time point into the pre-trained cabinet internal thermal identification model, analyzing the cabinet internal thermal abnormal characteristic set of the set energy storage cabinet, including an air flow cooling efficiency characteristic value, a hot spot diffusion characteristic value, a temperature difference imbalance fluctuation characteristic value, and a thermal spot aggregation risk characteristic value; based on the cabinet internal thermal abnormal characteristic set of the set energy storage cabinet, analyzing the thermal field disorder characteristic value of the set energy storage cabinet.

[0013] Further, the specific steps of analyzing the set of abnormal thermal characteristics in the cabinet of the energy storage cabinet are as follows: in the feature extraction subnetwork of the cabinet thermal identification model, receiving the cabinet thermal imaging data of each time point of the set energy storage cabinet, and extracting the cabinet thermal field characteristic vector of the corresponding time point; in the time sequence association subnetwork of the cabinet thermal identification model, based on the cabinet thermal field characteristic vector of each time point of the set energy storage cabinet, analyzing the set of abnormal thermal characteristics in the cabinet of the set energy storage cabinet.

[0014] Further, the feature extraction subnetwork includes an input preprocessing layer and a feature construction output layer, and the specific steps of extracting the cabinet thermal field characteristic vector of each time point of the set energy storage cabinet are as follows: in the input preprocessing layer of the feature extraction subnetwork, receiving the cabinet thermal imaging data of each time point of the set energy storage cabinet and performing preprocessing; in the feature construction output layer of the feature extraction subnetwork, based on the preprocessed cabinet thermal imaging data of each time point of the set energy storage cabinet, analyzing the cabinet thermal field characteristic vector of each time point of the set energy storage cabinet.

[0015] Further, the specific steps of performing dynamic air pressure adjustment on the set energy storage cabinet based on the thermal regulation characteristic value are as follows: comparing and analyzing the thermal regulation characteristic value of the set energy storage cabinet with the preset thermal regulation characteristic interval; and based on the comparison and analysis result, performing dynamic air pressure adjustment on the set energy storage cabinet.

[0016] The present application has the following advantages: (1) The energy storage cabinet thermal management system based on dynamic air pressure adjustment realizes global identification and dynamic self-adaptive adjustment of the internal thermal field of the energy storage cabinet by introducing the working condition thermal load analysis module and the thermal field identification analysis module. Specifically, the system not only collects operating condition data, but also combines cabinet thermal imaging data, analyzes the temperature distribution inside the cabinet through the pre-trained thermal identification model, generates thermal field disorder characteristic values that can reflect real-time heat dissipation pressure and abnormal evolution, and then generates thermal regulation characteristic values after fusion with comprehensive thermal risk assessment characteristic values to drive the dynamic air pressure adjustment module for regulation, thereby enabling the system to perceive the change trend of cabinet heat transfer on the overall level, dynamically match the heat dissipation demand, and avoid uneven heat dissipation or cooling lag, thereby significantly improving the thermal safety of the cabinet.

[0017] (2) The energy storage cabinet thermal management system based on dynamic air pressure regulation realizes accurate expression of the thermal load of the energy storage cabinet by establishing a fusion method of internal working condition operation state and external environmental condition, considers the dynamic response of the cabinet interior and the long-term disturbance of the external environment during the operation of the system, and generates a comprehensive thermal load characteristic value according to the same, the characteristic value can reflect the fluctuation of the cabinet internal heat dissipation performance and the continuous influence of the external condition on the heat dissipation process, thereby ensuring that the regulation and control are more adaptive, and then the cabinet still maintains stable thermal management capability under complex and changeable working conditions, and significantly improves the safety margin of operation.

[0018] (3) The energy storage cabinet thermal management system based on dynamic air pressure regulation realizes feature modeling of the cabinet internal thermal field and identification of the dynamic evolution law by constructing a cabinet internal thermal identification model composed of a feature extraction sub-network and a time sequence association sub-network, in the feature extraction stage, the model can extract a cabinet internal thermal field feature vector expression of multiple types of features in the spatial level from the thermal imaging data of each time point, in the time sequence association stage, the model processes the cabinet internal thermal field feature vectors of continuous time points based on a long short-term memory structure, which can retain long-term stable trends and identify short-term abnormal fluctuations, and generates a cabinet internal thermal abnormal feature set, thereby enabling the system to simultaneously establish global cognition of the heat dissipation process in the spatial and time dimensions, and then improving the accuracy of feature extraction and providing more accurate decision basis for dynamic air pressure regulation.

[0019] Of course, implementing any product of the present application does not necessarily need to achieve all the advantages described above at the same time. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 A block diagram of the energy storage cabinet thermal management system based on dynamic air pressure regulation of the present application.

[0021] Figure 2 A specific step flowchart for analyzing and setting the cabinet comprehensive thermal load characteristic value of each time point of the energy storage cabinet in the energy storage cabinet thermal management system based on dynamic air pressure regulation of the present application.

[0022] Figure 3 A time sequence feature data schematic diagram for setting the energy storage cabinet in the energy storage cabinet thermal management system based on dynamic air pressure regulation of the present application.

[0023] Figure 4 A specific step flowchart for analyzing and setting the cabinet internal thermal abnormal feature set of the energy storage cabinet in the energy storage cabinet thermal management system based on dynamic air pressure regulation of the present application. DETAILED DESCRIPTION

[0024] Please refer to Figure 1The embodiment of the present application provides a technical scheme: a kind of energy storage cabinet thermal management system based on dynamic wind pressure regulation, comprising: data acquisition module, for in the set sliding window (such as 3 seconds, and the time length of first sliding window does not satisfy 3 seconds, fan maintains preset safety benchmark wind pressure, and the data collected are stored until the time length of first sliding window satisfies 3 seconds, and carry out analysis and adjustment) inside, continuously obtain the operating condition data of each time point of the set energy storage cabinet, cabinet thermal imaging data;Condition thermal load analysis module, for based on the operating condition data of each time point of the set energy storage cabinet, analyze its corresponding time point cabinet comprehensive thermal load characteristic value, and analyze the cabinet thermal risk assessment characteristic value of the set energy storage cabinet;Thermal field identification analysis module, for based on pre-trained cabinet thermal identification model, and combine the cabinet thermal imaging data of each time point (one time point is a frame), analyze the thermal field disorder characteristic value of the set energy storage cabinet;Thermal regulation characteristic generation module, for based on the cabinet thermal risk assessment characteristic value of the set energy storage cabinet, thermal field disorder characteristic value, analyze the thermal regulation characteristic value (characterize the overall thermal management regulation demand degree of energy storage cabinet under current operating condition) of the set energy storage cabinet;Dynamic wind pressure regulation module, for based on thermal regulation characteristic value, the set energy storage cabinet is handled by dynamic wind pressure regulation.

[0025] Wherein, the specific formula for calculating the thermal regulation characteristic value of the set energy storage cabinet is as follows: ; Wherein, is the thermal regulation characteristic value of the set energy storage cabinet, is the cabinet thermal risk assessment characteristic value of the set energy storage cabinet, is the cabinet thermal risk regulation coefficient stored in the database, is the thermal field disorder characteristic value of the set energy storage cabinet, is the thermal field disorder regulation coefficient stored in the database, , is the difference regulation coefficient stored in the database, and in the present embodiment, the cabinet thermal risk regulation coefficient , thermal field disorder regulation coefficient , difference regulation coefficient stored in the database are 0.438, 0.562 and 0.386 respectively.

[0026] The specific steps of dynamic wind pressure regulation processing of the set energy storage cabinet based on thermal regulation characteristic value are as follows: comparing and analyzing the thermal regulation characteristic value of the set energy storage cabinet with the preset thermal regulation characteristic interval;Based on the comparison and analysis result, the set energy storage cabinet is dynamically wind pressure regulated, which is specifically: If the thermal regulation characteristic value of the set energy storage cabinet is lower than the lower limit of the preset thermal regulation characteristic interval, the internal heat dissipation state of the cabinet is stable, the fan is maintained in a low wind pressure operation mode to reduce energy consumption and maintain basic air circulation; if the thermal regulation characteristic value of the set energy storage cabinet is within the preset thermal regulation characteristic interval, the wind pressure of the fan is increased to the reference wind pressure level corresponding to the interval compared with the low wind pressure mode, and the reference wind pressure level is set by experimental calibration or operation experience, such as 110% to 130% of the low wind pressure mode, to maintain the normal heat dissipation capacity of the cabinet; if the thermal regulation characteristic value of the set energy storage cabinet is higher than the upper limit of the preset thermal regulation characteristic interval, it indicates that the cabinet heat dissipation state is at a high load level, according to the amplitude of the thermal regulation characteristic value exceeding the upper limit, the wind pressure level is increased in proportion to the linear (for example, 5% increase in wind pressure for every 10% exceeding the upper limit), to enhance the air circulation in the cabinet, restore the temperature distribution balance, and reduce the risk of overheating.

[0027] Specifically, as shown in Figure 2 The running condition data includes air circulation efficiency factor, cavity heat accumulation value, cavity dust deposition distribution index, cavity humidity accumulation factor, electromagnetic radiation noise intensity value, and cabinet vibration response amplitude. The specific steps of analyzing the cabinet comprehensive thermal load characteristic value at each time point of the set energy storage cabinet are as follows: based on the running condition data of the set energy storage cabinet at each time point, the thermal risk characteristic set at the corresponding time point is analyzed, including the cabinet heat dissipation degradation characteristic value and the structure and electrical stability characteristic value; the external thermal load characteristic value at each time point of the set energy storage cabinet is obtained, and the cabinet comprehensive thermal load characteristic value at the corresponding time point is analyzed in combination with the thermal risk characteristic set.

[0028] The air circulation efficiency factor is the effective degree of the internal air flow of the energy storage cabinet in the circulation process, which can be obtained by setting wind speed sensors and wind pressure sensors at the inlet and outlet of the cabinet, collecting the inlet and outlet air speeds and the inlet and outlet air pressures at each time point, and performing ratio processing on the collected values, such as outlet air speed / inlet air speed, and then performing weighted processing on the results to obtain the air circulation efficiency factor.

[0029] The cavity heat accumulation value is the degree of heat stratification caused by the upward floating of hot air in the energy storage cabinet. When the gas at the top of the cabinet is significantly lighter than that at the bottom, and the pressure difference is large, it indicates that there is heat accumulation at the top, which affects the uniformity of heat dissipation. The cavity heat accumulation value can be obtained by setting pressure difference sensors and flow sensors at the top and bottom of the cabinet, collecting the top and bottom pressure values and the top and bottom air flow values at each time point, and performing deviation processing on the collected values, such as |top pressure value-bottom pressure value| / top pressure value, and then performing weighted processing on the deviation processing results to obtain the cavity heat accumulation value.

[0030] The cavity dust deposition distribution index is the accumulation degree of dust in different areas inside the cabinet. Dust blocking the air duct or filter screen will increase the airflow resistance, reduce the efficiency of the fan, and thus affect the heat dissipation capacity. The cavity dust deposition distribution index can be obtained by arranging multiple optical dust sensors at the air inlet and filter screen area of the cabinet, collecting the dust concentration values at different positions, extracting the maximum dust concentration value and the minimum dust concentration value, and performing ratio processing, i.e. (maximum dust concentration value - minimum dust concentration value) / maximum dust concentration value.

[0031] The cavity heat and humidity accumulation factor is the abnormal coupling degree of local humidity and airflow state inside the energy storage cabinet. When the humidity is high for a long time and the airflow speed is weakened, local condensation is easy to occur, reducing the heat dissipation efficiency. The cavity heat and humidity accumulation factor can be obtained by arranging humidity sensors and flow sensors at three different height positions, i.e. top, middle and bottom, of the cabinet, collecting the top humidity, middle humidity, bottom humidity and airflow speed at the corresponding positions at each time point, analyzing the humidity difference between the top and middle, i.e. |top humidity - middle humidity| / top humidity, and multiplying it by the airflow speed ratio between the middle and top, at the same time, analyzing the humidity difference between the middle and bottom, and multiplying it by the airflow speed ratio between the bottom and middle, and then performing weighted processing on the two multiplication results to obtain the cavity heat and humidity accumulation factor.

[0032] The electromagnetic radiation noise intensity value is the electromagnetic interference intensity generated by busbars, cables and power devices during high-power operation of the energy storage cabinet. Excessive electromagnetic radiation noise can easily lead to an increase in local electrical stress and cause poor heat dissipation. The electromagnetic radiation noise intensity value can be obtained by arranging EMI / RFI electromagnetic radiation sensors at the busbar area and cable convergence area of the cabinet, collecting the electromagnetic signal amplitude at each time point, filtering the effective signal in the target frequency band (e.g. 150 kHz-30 MHz), and then converting the signal amplitude to a preset reference amplitude (e.g. 1 μV / m) to obtain the electromagnetic radiation noise intensity value.

[0033] The cabinet vibration response amplitude is the mechanical vibration intensity caused by the fan, airflow or external disturbance during the operation of the cabinet. Excessive vibration can cause the fan to deviate, the airflow distribution to be abnormal, and thus affect the heat management effect. The cabinet vibration response amplitude can be obtained by arranging MEMS three-axis acceleration sensors at the cabinet frame, fan mounting point and battery module support point, collecting three-direction acceleration values at each time point, and then performing weighted processing on the square root of the sum of the three-direction acceleration values.

[0034] The specific formula for calculating the cabinet comprehensive heat load characteristic value of the energy storage cabinet at a certain time point is as follows: ; wherein, is the cabinet comprehensive heat load characteristic value of the energy storage cabinet at a certain time point, to set the cabinet heat dissipation degradation characteristic value of the energy storage cabinet at a certain time point, to store the heat dissipation degradation adjustment coefficient in the database, to set the structure electrical stability characteristic value of the energy storage cabinet at a certain time point, to store the electrical stability adjustment coefficient in the database, to set the external thermal load characteristic value of the energy storage cabinet at a certain time point, to store the external adjustment coefficient in the database, .

[0035] It needs to be explained that the heat dissipation degradation adjustment coefficient , the electrical stability adjustment coefficient , the external adjustment coefficient The acquisition steps are as follows: reading the cabinet heat dissipation degradation characteristic value, the structure electrical stability characteristic value, and the external thermal load characteristic value of the energy storage cabinet at each time point, extracting the cabinet heat dissipation degradation characteristic mean value, the structure electrical stability characteristic mean value, and the external thermal load characteristic mean value of the energy storage cabinet, and performing summation processing to obtain the comprehensive thermal load sum value, and performing ratio processing on the cabinet heat dissipation degradation characteristic mean value, the structure electrical stability characteristic mean value, and the external thermal load characteristic mean value of the energy storage cabinet with the comprehensive thermal load sum value respectively, and taking the corresponding results as the heat dissipation degradation adjustment coefficient , the electrical stability adjustment coefficient , and the external adjustment coefficient .

[0036] The specific implementation example of calculating the cabinet comprehensive thermal load characteristic value of the energy storage cabinet at a certain time point is as follows, and the existing data includes: the cabinet heat dissipation degradation characteristic value, the structure electrical stability characteristic value, and the external thermal load characteristic value of the energy storage cabinet at 5 time points (randomly selected), as shown in Table 1 and Figure 3 .

[0037] The heat dissipation degradation adjustment coefficient stored in the database is about: 0.416; The electrical stability adjustment coefficient stored in the database is about: 0.342; The external adjustment coefficient stored in the database is about: 0.242; Substitute the data in Table 1 and the above coefficients into the specific formula for calculating the cabinet comprehensive thermal load characteristic value of the energy storage cabinet at a certain time point to obtain: The cabinet comprehensive thermal load characteristic value of the first time point of the energy storage cabinet is set to be approximately 0.465, that is, 0.416*0.403+0.342*(1 / (1+0.568))+0.242*0.327; The cabinet comprehensive thermal load characteristic value of the second time point of the energy storage cabinet is set to be approximately 0.505, that is, 0.416*0.463+0.342*(1 / (1+0.524))+0.242*0.364; The cabinet comprehensive thermal load characteristic value of the third time point of the energy storage cabinet is set to be approximately 0.469, that is, 0.416*0.421+0.342*(1 / (1+0.536))+0.242*0.294; The cabinet comprehensive thermal load characteristic value of the fourth time point of the energy storage cabinet is set to be approximately 0.472, that is, 0.416*0.397+0.342*(1 / (1+0.486))+0.242*0.316; The cabinet comprehensive thermal load characteristic value of the fifth time point of the energy storage cabinet is set to be approximately 0.466, that is, 0.416*0.412+0.342*(1 / (1+0.514))+0.242*0.284.

[0038] The specific steps of analyzing the thermal risk characteristic set of each time point of the energy storage cabinet are as follows: based on the air flow circulation efficiency factor, the cavity heat accumulation value, the cavity dust deposition distribution index and the cavity hygrothermal accumulation factor of each time point of the energy storage cabinet, the cabinet heat dissipation degradation characteristic of the corresponding time point is analyzed, which is specifically: the air flow circulation efficiency factor, the cavity heat accumulation value, the cavity dust deposition distribution index and the cavity hygrothermal accumulation factor of each time point are standardized, the weighted processing is carried out based on the standardized processing result (and the weighted processing result is normalized to map its value to 0-1), and in the weighted processing process, the standardized air flow circulation efficiency factor is taken as the inverse, that is, 1 / (1+standardized air flow circulation efficiency factor), so as to obtain the cabinet heat dissipation degradation characteristic of the corresponding time point; based on the electromagnetic radiation noise intensity value and the cabinet vibration response amplitude of each time point of the energy storage cabinet, the structure and electrical stability characteristic value of the corresponding time point is analyzed, which is specifically: the electromagnetic radiation noise intensity value and the cabinet vibration response amplitude of each time point are standardized, the weighted processing is carried out based on the standardized processing result, and the weighted processing result is taken as the inverse, that is, 1 / (1+the weighted processing result), so as to obtain the structure and electrical stability characteristic value of the corresponding time point. When the structure and electrical stability characteristic value tends to be high, it indicates that the electrical stress and the structure state are both at a stable level, the cabinet cooling air flow can be uniformly distributed, which is conducive to maintaining a good cooling environment.

[0039] The specific steps of obtaining the external thermal load characteristic value of the set energy storage cabinet at each time point are as follows: obtaining the external environment data of the set energy storage cabinet at each time point (including the shell heat conduction disturbance factor, the environment humidity value, the cabinet outer surface solar radiation absorption rate value, and the environment dust concentration value), and performing standardization processing (i.e., standardization processing of the shell heat conduction disturbance factor, the environment humidity value, the cabinet outer surface solar radiation absorption rate value, and the environment dust concentration value of the set energy storage cabinet at each time point); based on the external environment data of the set energy storage cabinet at each time point after standardization processing, analyzing the external thermal load characteristic value at the corresponding time point, which is specifically: performing weighted processing on the shell heat conduction disturbance factor, the environment humidity value, the cabinet outer surface solar radiation absorption rate value, and the environment dust concentration value of the set energy storage cabinet at each time point after standardization processing (and performing normalization processing on the weighted processing result, mapping its value to 0-1), to obtain the external thermal load characteristic value at the corresponding time point, which is used to represent the comprehensive thermal burden of the external environment condition on the heat dissipation capacity of the energy storage cabinet, and the larger the value, the higher the external thermal pressure of the cabinet and the greater the risk of heat dissipation limitation.

[0040] The shell heat conduction disturbance factor is the heat conduction fluctuation of the energy storage cabinet shell under the rapid change of the external environment. If the shell heat disturbance is too large, it will cause abnormal temperature gradient in the cabinet, affecting the heat dissipation balance. It can be obtained by arranging temperature sensors at corresponding positions on the outer surface and inner surface of the cabinet shell, collecting the outer surface temperature and inner surface temperature at each time point, and performing ratio processing, i.e., outer surface temperature / inner surface temperature, and taking the result as the shell heat conduction disturbance factor.

[0041] The environment humidity value is obtained by a humidity sensor.

[0042] The cabinet outer surface solar radiation absorption rate value is the ability of the cabinet shell to absorb heat under solar radiation. The higher the value, the faster the temperature rise of the cabinet outer surface, and the greater the internal heat dissipation pressure. It can be obtained by arranging light sensors and temperature sensors on the surface of the cabinet shell, collecting solar radiation intensity value and cabinet outer surface temperature value, and environment temperature value at each time point; combining the reflectivity of the cabinet shell coating (obtained from the standard material optical parameter database corresponding to the material) to analyze the solar radiation reference value, i.e., solar radiation intensity value x (1-reflectivity), and comprehensively analyzing the solar radiation intensity value and the cabinet outer surface temperature value, and the environment temperature value, i.e., (solar radiation intensity value / solar radiation reference value) x |cabinet outer surface temperature value-environment temperature value|, and taking the result as the cabinet outer surface solar radiation absorption rate value.

[0043] The environmental dust concentration value is the concentration of particulate matters in the air of the environment where the energy storage cabinet is located. The higher the external dust concentration, the easier it is to cause the air inlet and filter screen to be blocked, thereby affecting the long-term heat dissipation performance of the cabinet. The optical dust sensor can be arranged at the air inlet of the cabinet to collect the dust concentration value in the external air at each time point, and the concentration value is taken as the environmental dust concentration value.

[0044] In this embodiment, through the processing of this step, the multi-source data received by the energy storage cabinet during operation can be uniformly mapped and fused, thereby avoiding the judgment deviation caused by the fragmentation of single-point information, and the heat load characteristic value can more comprehensively reflect the overall heat dissipation pressure borne by the cabinet. Secondly, in the analysis process, by introducing the standardization and weighting of data, not only the comparability between different physical quantities is realized, but also the dynamic adjustment according to the actual importance of the influence on heat dissipation is realized, so that the result is more in line with the actual operation state. Finally, the cabinet comprehensive heat load characteristic value formed can be used as the core input to provide accurate quantitative basis for subsequent risk assessment and wind pressure adjustment, so as to ensure that the system can realize stable and efficient heat dissipation management during operation, thereby significantly improving the reliability of the cabinet as a whole.

[0045] Specifically, the specific steps of analyzing the cabinet heat risk assessment characteristic value of the set energy storage cabinet are as follows: based on the cabinet comprehensive heat load characteristic value of the set energy storage cabinet at each time point, analyzing the load characteristic set of the set energy storage cabinet, which is specifically: based on the cabinet comprehensive heat load characteristic value of the set energy storage cabinet at each time point, extracting the cabinet comprehensive heat load characteristic mean value, the cabinet comprehensive heat load characteristic maximum value and the cabinet comprehensive heat load characteristic minimum value of the set energy storage cabinet, i.e. the load characteristic set; based on the load characteristic set of the set energy storage cabinet, analyzing the cabinet heat risk assessment characteristic value of the set energy storage cabinet, which is specifically: based on the cabinet comprehensive heat load characteristic mean value, the cabinet comprehensive heat load characteristic maximum value and the cabinet comprehensive heat load characteristic minimum value of the set energy storage cabinet, respectively extracting the fluctuation ratio, i.e. (cabinet comprehensive heat load characteristic maximum value-cabinet comprehensive heat load characteristic minimum value) / cabinet comprehensive heat load characteristic mean value, the peak value ratio (cabinet comprehensive heat load characteristic maximum value / cabinet comprehensive heat load characteristic mean value), the valley value ratio (cabinet comprehensive heat load characteristic minimum value / cabinet comprehensive heat load characteristic mean value), and performing weighted processing to obtain the cabinet heat risk assessment characteristic value of the set energy storage cabinet.

[0046] In this embodiment, through this step, more representative risk assessment indexes can be extracted on the basis of the comprehensive thermal load characteristics of the cabinet. Single mean or instantaneous data can only reflect a certain time point or overall trend, and it is difficult to reveal potential abnormal risks. By introducing the comparison relationship of the maximum value, the minimum value and the mean value, extreme fluctuations and local unstable states that may occur during operation can be effectively captured. In addition, the fluctuation ratio can reflect the stability of the overall load. If the ratio is high, it means that the cabinet operating conditions have large fluctuations. The peak value ratio reveals the pressure bearing capacity of the system under high load. The valley value ratio helps to identify potential heat dissipation hazards under low load. Through weighted processing, different indexes are integrated into a unified risk assessment characteristic value according to the importance of actual influence, so that the results are comprehensive. Finally, this method enhances the early warning ability of the system to abnormal working conditions, so as to ensure that the cabinet can maintain stable thermal management performance under complex operating environment.

[0047] Specifically, the cabinet thermal imaging data specifically refers to the temperature value and two-dimensional coordinates of each pixel point in the thermal imaging, and the cabinet thermal identification model comprises a feature extraction subnetwork and a time sequence correlation subnetwork.

[0048] The specific steps of analyzing the thermal field disorder characteristic value of the set energy storage cabinet are as follows: input the cabinet thermal imaging data of each time point of the set energy storage cabinet into the pre-trained cabinet thermal identification model, analyze the cabinet thermal abnormal characteristic set of the set energy storage cabinet, including the airflow cooling efficiency characteristic value, the hot spot diffusion characteristic value, the temperature difference imbalance fluctuation characteristic value, and the thermal spot aggregation risk characteristic value; based on the cabinet thermal abnormal characteristic set of the set energy storage cabinet, analyze the thermal field disorder characteristic value of the set energy storage cabinet, which is specifically: the airflow cooling efficiency characteristic value, the hot spot diffusion characteristic value, the temperature difference imbalance fluctuation characteristic value, and the thermal spot aggregation risk characteristic value of the set energy storage cabinet are weighted, and the airflow cooling efficiency characteristic value is taken as the inverse, i.e. 1 / (1+airflow cooling efficiency characteristic value), to obtain the thermal field disorder characteristic value of the set energy storage cabinet. When the thermal field disorder characteristic value is high, it means that the internal thermal field of the cabinet is in a disorder state, indicating that the thermal management performance of the cabinet is decreased and there is an overheating risk.

[0049] As shown in Figure 4 The specific steps of analyzing the cabinet thermal abnormal characteristic set of the set energy storage cabinet are as follows: in the feature extraction subnetwork of the cabinet thermal identification model, the cabinet thermal imaging data of each time point of the set energy storage cabinet is received, and the cabinet thermal field feature vector of the corresponding time point is extracted; in the time sequence correlation subnetwork of the cabinet thermal identification model, based on the cabinet thermal field feature vector of each time point of the set energy storage cabinet, the cabinet thermal abnormal characteristic set of the set energy storage cabinet is analyzed, which is specifically: the time sequence correlation subnetwork comprises an input layer, an LSTM layer and an output layer. In the input layer of the time correlation subnetwork, the in-cabinet thermal field feature vector of each time point of the set energy storage cabinet is received, and each feature in the thermal field feature vector is normalized; In the LSTM layer of the time correlation subnetwork, based on the normalized in-cabinet thermal field feature vector of each time point of the set energy storage cabinet, a time sequence thermal evolution feature vector of the set energy storage cabinet is extracted, that is, the normalized in-cabinet thermal field feature vector sequence is sequentially input into the LSTM layer, the LSTM layer updates the state of the in-cabinet thermal field feature vector of each time point through the interaction mechanism of the input gate, the forgetting gate and the output gate, and accumulates the historical information in the cell state. In this process, the LSTM network not only can retain the stable change rule in the long time scale, but also can suppress the interference of short-time noise, so as to extract more representative time sequence mode. Based on this processing process, the network can automatically learn the continuity, mutation and periodicity of the airflow cooling path attenuation feature, the hot spot shape complex feature, the opposite temperature difference imbalance feature and the hot spot distribution feature between different time points, and map them into a multi-dimensional time sequence thermal evolution feature vector, such as: For the airflow cooling path attenuation feature in the in-cabinet thermal field feature vector of each time point, a sliding average processing is performed to extract the airflow cooling efficiency feature. When the feature value is low, it indicates that the temperature drop of the cooling path is insufficient, and the internal airflow of the cabinet may be short-circuited or the fan may be abnormally operated, resulting in a decrease in heat exchange efficiency. For the hot spot shape complex feature in the in-cabinet thermal field feature vector of each time point, a variance processing is performed to extract the hot spot diffusion feature. When the feature value is high, it indicates that the boundary complexity of the hot spot region continues to rise, showing a diffusion or splitting trend, and there may be a risk of local overheating diffusion or abnormal heat dissipation. For the opposite temperature difference imbalance feature in the in-cabinet thermal field feature vector of each time point, a range processing (extracting the difference between the maximum value of the opposite temperature difference imbalance feature and the minimum value of the opposite temperature difference imbalance feature) is performed to obtain the temperature difference imbalance fluctuation feature. When the feature value is large, it indicates that the temperature field of the cabinet is uneven, which may be caused by uneven airflow distribution, unbalanced fan operation or local heat dissipation failure. For the hot spot distribution feature in the in-cabinet thermal field feature vector of each time point, the hot spot distribution feature mean and the hot spot distribution feature variance are extracted respectively, and a weighted processing is performed to extract the hot spot aggregation risk feature. When the feature is high, it indicates that the high temperature region in the cabinet has strong and concentrated distribution, and there is a risk of local overheating and uneven heat dissipation. The airflow cooling efficiency feature, the hot spot diffusion feature, the temperature difference imbalance fluctuation feature and the hot spot aggregation risk feature are spliced into a time sequence thermal evolution feature vector. In the output layer of the time sequence correlation subnetwork, the airflow cooling efficiency feature, the hotspot diffusion feature, the temperature difference imbalance fluctuation feature, and the hot spot aggregation risk feature in the time sequence thermal evolution feature vector of the set energy storage cabinet are processed through a Sigmoid function, and the results are mapped between 0 and 1 to obtain specific airflow cooling efficiency feature values, hotspot diffusion feature values, temperature difference imbalance fluctuation feature values, and hot spot aggregation risk feature values, thereby obtaining the cabinet thermal anomaly feature set of the set energy storage cabinet.

[0050] The pre-training step of the cabinet thermal identification model is as follows: An annotated data set is obtained, which is composed of cabinet thermal imaging data of an energy storage cabinet and corresponding heat dissipation state labels. The labels are annotated by experts in the thermal management field according to cabinet operation monitoring records and actual heat dissipation state feedback. Each sample in the annotated data set includes cabinet thermal imaging images of the set energy storage cabinet at consecutive time points, and true value labels of the airflow cooling path attenuation feature, the hotspot shape complexity feature, the opposite temperature difference imbalance feature, and the hot spot distribution feature in the thermal imaging images, ensuring that each sample has complete feature annotation information.

[0051] The annotated data set is divided into a training set, a validation set, and a test set, for example, 80% of the data is used for training, 10% of the data is used for validation, and 10% of the data is used for testing.

[0052] The cabinet thermal identification model is trained. Taking the time sequence correlation subnetwork as an example, the cabinet thermal field feature vector sequence is input into the LSTM layer, and the gating mechanism of the input gate, the forget gate, and the output gate is used to update the state of the sequence features, capture the long-term dependence relationship and short-term fluctuation pattern between the thermal field features at different time points, and learn the time sequence evolution law of airflow cooling, hotspot diffusion, temperature difference imbalance, and hot spot aggregation. The LSTM network optimizes the parameters through the back propagation algorithm (BPTT) to minimize the prediction error (such as mean square error MSE or cross-entropy loss), and gradually improves the fitting ability of the model to the thermal field dynamic process.

[0053] During the training process, the model is iteratively optimized using an optimization algorithm such as the Adam optimizer, and the learning rate, LSTM hidden layer unit number, and other hyperparameters are adjusted to improve the convergence speed and generalization performance of the model. The training process is evaluated by the validation set to prevent overfitting. When the validation performance is stable, the optimal model is selected and the performance is verified on the test set to ensure that the model can accurately extract the airflow cooling efficiency feature, the hotspot diffusion feature, the temperature difference imbalance fluctuation feature, and the hot spot aggregation risk feature on unseen thermal imaging data.

[0054] Finally, the trained in-cabinet thermal identification model will be saved, and the model parameters will be used in the subsequent practical application stage to ensure online deployment and real-time analysis of the in-cabinet thermal field state during the operation of the energy storage cabinet, and output of the in-cabinet thermal abnormal feature set to provide data support for dynamic air pressure regulation and thermal management optimization.

[0055] The feature extraction sub-network includes an input preprocessing layer and a feature construction output layer. The specific steps of extracting the in-cabinet thermal field feature vector of each time point of the set energy storage cabinet are as follows: in the input preprocessing layer of the feature extraction sub-network, the in-cabinet thermal imaging data of each time point of the set energy storage cabinet is received and preprocessed, which specifically includes: based on the median filtering or Gaussian filtering algorithm, the thermal imaging image is smoothed to reduce isolated high-temperature pixel points caused by sensor noise or external interference, thereby improving the stability of feature extraction; In the feature construction output layer of the feature extraction sub-network, based on the preprocessed in-cabinet thermal imaging data of each time point of the set energy storage cabinet, the in-cabinet thermal field feature vector of each time point of the set energy storage cabinet is analyzed, which specifically includes: calling the target detection model to identify the in-cabinet thermal imaging of each time point of the set energy storage cabinet (i.e., the preprocessed thermal imaging data as input, the target detection model extracts the convolution features and generates the candidate regions to locate the positions of the air inlet and air outlet in the thermal imaging, and outputs the corresponding bounding box coordinates; by decoding the bounding box coordinates, the pixel range of the air inlet area and the air outlet area corresponding to each time point can be obtained), to obtain the air inlet area and the air outlet area of each time point, extract the air inlet temperature mean value (i.e., the mean value of the temperature values of each pixel point in the air inlet area), the air inlet two-dimensional coordinates (i.e., the mean value of the two-dimensional coordinates of each pixel point in the air inlet area), the air outlet temperature mean value (i.e., the mean value of the temperature values of each pixel point in the air outlet area), and the air outlet two-dimensional coordinates (i.e., the mean value of the two-dimensional coordinates of each pixel point in the air outlet area) of the air inlet area and the air outlet area, respectively, and analyze the distance between the air inlet area and the air outlet area (analyze the air inlet two-dimensional coordinates and the air outlet two-dimensional coordinates based on the Euclidean distance formula), and perform ratio processing with the air inlet temperature mean value and the air outlet temperature mean value, i.e., max[0, (air inlet temperature mean value-air outlet temperature mean value) / distance between air inlet area and air outlet area], to extract the airflow cooling path attenuation feature of each time point. When the feature value is low, it indicates that the airflow cannot effectively carry away heat in the cabinet, and there may be airflow short circuit or abnormal operation of the fan. For the temperature value of each pixel point in thermal imaging, the temperature mean value and the temperature standard deviation value are extracted respectively, and a temperature threshold value (i.e. temperature mean value + 2 x temperature standard deviation value) is set based thereon, and the pixel points higher than the temperature threshold value are marked as high-temperature pixel points, and all the high-temperature pixel points are subjected to connected processing (based on the image connected domain analysis method, the spatially adjacent high-temperature pixel points are classified into the same connected region; specifically, the 4-neighborhood or 8-neighborhood judgment rule is adopted, if two high-temperature pixel points are adjacent in the horizontal direction, the vertical direction or the diagonal direction, they are regarded as connected, by scanning the whole image one by one and merging the connected pixel set, a plurality of high-temperature regions independent of each other can be obtained), so as to obtain a plurality of high-temperature regions, and the edge detection processing (such as Canny operator, the edge strength is calculated through the gradient amplitude and direction, and the high-temperature edge pixel points are extracted based on the double-threshold suppression method) is performed on each high-temperature region, so as to extract a plurality of high-temperature edge pixel points of each high-temperature region, the two-dimensional coordinates of all the high-temperature edge pixel points are extracted, the Euclidean distance between adjacent high-temperature edge pixel points is calculated, and the results are accumulated to obtain the perimeter of each high-temperature region, and the total number of high-temperature pixel points of each high-temperature region is counted as the area of the corresponding high-temperature region, and the perimeter of the corresponding high-temperature region is subjected to ratio processing, i.e. perimeter 2 / area, and the result is subjected to weighted processing, so as to extract the hotspot shape complexity feature of each time point, the higher the feature, the more complex the boundary of the hotspot region, which presents the diffusion or splitting trend, and it may be a precursor of local overheating diffusion or abnormal heat dissipation; The thermal imaging is divided based on the horizontal center line (i.e. the geometric center line in the height direction of the thermal imaging) to obtain the upper half region and the lower half region of the thermal imaging, and the temperature values of all the pixel points in the upper half region and the lower half region of the thermal imaging are subjected to mean value respectively, the upper half temperature mean value and the lower half temperature mean value are extracted, the difference value (taking the absolute value) between the two is subjected to ratio conversion with the average level, so as to obtain the opposite temperature difference imbalance feature of each time point, when the feature is large, it indicates that there is obvious difference in the temperature field of the upper and lower regions of the cabinet, which may be caused by uneven air distribution, unbalanced fan operation or local heat dissipation failure, thereby prompting the potential thermal management risk; The center two-dimensional coordinates of the corresponding high-temperature region (i.e., the mean value of the two-dimensional coordinates of each high-temperature pixel point in the high-temperature region) of several high-temperature regions in the thermal imaging is read, the Euclidean distance of any adjacent high-temperature region is calculated, and a weighted processing is performed to extract the spatial clustering feature, and the thermal energy of each high-temperature region (i.e., the mean value of the temperature values of each pixel point in the thermal imaging is processed to obtain the global temperature mean value, and the temperature values of each high-temperature pixel point in the corresponding high-temperature region are processed to obtain the regional temperature mean value, and the difference is processed, and the absolute value is taken to obtain the thermal energy of the corresponding high-temperature region) is extracted, a weighted average processing is performed to obtain the thermal energy mean value, and a weighted processing is performed with the spatial clustering feature to extract the thermal spot distribution feature of each time point. When the thermal spot distribution feature is high, it indicates that the high-temperature region has strong thermal energy and is concentrated with each other, and there is a risk of serious local overheating and uneven heat dissipation in the cabinet.

[0056] In the embodiment, through the design of the step, fine analysis of the internal thermal field of the cabinet and recognition of dynamic evolution law can be realized. In the feature extraction stage, the system not only monitors the single-point temperature, but also extracts features including cooling path efficiency and hot spot boundary form through in-depth analysis of the thermal imaging data, so that the heat dissipation state in the cabinet can be quantitatively expressed in the overall perspective. In the time sequence correlation stage, the long short-term memory structure is used to process the feature sequence of continuous time points, so that the long-term trend and short-term fluctuation can be captured at the same time, the misjudgment caused by instantaneous abnormality or sensor noise can be avoided, more representative time sequence features are extracted in turn, and finally the feature values are nonlinearly mapped, which not only improves the comparability between different features, but also enhances the fitting ability of the model to the complex heat dissipation process. The cabinet internal thermal anomaly feature set generated by the method can reveal the potential overheating risk, thereby significantly improving the operation safety of the energy storage cabinet thermal management.

[0057] Although preferred embodiments of the application have been described, those skilled in the art can make further changes and modifications to the embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including all changes and modifications falling within the scope of the application.

[0058] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A thermal management system for energy storage cabinets based on dynamic wind pressure regulation, characterized in that, include: The data acquisition module is used to continuously acquire the operating condition data and thermal imaging data inside the set energy storage cabinet at each time point; The operating condition heat load analysis module is used to analyze the comprehensive heat load characteristic value of the energy storage cabinet at each time point based on the operating condition data of the set energy storage cabinet at each time point, and to analyze the cabinet thermal risk assessment characteristic value of the set energy storage cabinet. The thermal field identification and analysis module is used to analyze and set the thermal field disturbance characteristic value of the energy storage cabinet based on the pre-trained cabinet internal thermal identification model and combined with the cabinet internal thermal imaging data at each time point. The thermal regulation feature generation module is used to analyze the thermal regulation feature value of the set energy storage cabinet based on the cabinet thermal risk assessment feature value and thermal field disturbance feature value. The dynamic air pressure regulation module is used to dynamically regulate the air pressure of a set energy storage cabinet based on thermal regulation characteristic values.

2. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 1, characterized in that, The operating condition data includes airflow circulation efficiency factor, cavity heat accumulation value, cavity dust deposition distribution index, cavity damp heat accumulation factor, electromagnetic radiation noise intensity value, and cabinet vibration response amplitude. The specific steps for analyzing and setting the comprehensive heat load characteristic value of the energy storage cabinet at each time point are as follows: Based on the operating condition data of the energy storage cabinet at each time point, the thermal risk characteristic set of the corresponding time point is analyzed, including the cabinet heat dissipation degradation characteristic value and the structural electrical stability characteristic value. The external heat load characteristic value of the set energy storage cabinet at each time point is obtained, and combined with the thermal risk characteristic set, the comprehensive heat load characteristic value of the cabinet at the corresponding time point is analyzed.

3. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 2, characterized in that, The specific steps for analyzing and setting the thermal risk characteristic set of the energy storage cabinet at each time point are as follows: Based on the airflow circulation efficiency factor, cavity heat accumulation value, cavity dust deposition distribution index, and cavity damp heat accumulation factor of the energy storage cabinet at each time point, the heat dissipation degradation characteristics of the cabinet at the corresponding time point are analyzed. Based on the electromagnetic radiation noise intensity value and cabinet vibration response amplitude of the energy storage cabinet at each time point, the structural electrical stability characteristic value at the corresponding time point is analyzed.

4. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 2, characterized in that, The specific steps for obtaining the characteristic values ​​of the external heat load at each time point for the set energy storage cabinet are as follows: Acquire external environmental data for each time point of the designated energy storage cabinet and perform standardized processing; Based on the standardized external environmental data of the energy storage cabinet at each time point, the characteristic values ​​of the external heat load at the corresponding time point are analyzed.

5. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 1, characterized in that, The specific steps for analyzing and setting the characteristic values ​​for the thermal risk assessment of energy storage cabinets are as follows: Based on the comprehensive heat load characteristic value of the energy storage cabinet at each time point, the load characteristic set of the energy storage cabinet is analyzed. Based on the load characteristic set of the energy storage cabinet, the characteristic values ​​of the cabinet thermal risk assessment are analyzed.

6. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 1, characterized in that, The thermal imaging data inside the cabinet specifically includes the temperature value and two-dimensional coordinates of each pixel in the thermal imaging. The thermal identification model inside the cabinet includes a feature extraction subnetwork and a temporal correlation subnetwork.

7. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 6, characterized in that, The specific steps for analyzing and setting the thermal field disorder characteristic values ​​of the energy storage cabinet are as follows: The thermal imaging data inside the energy storage cabinet at each time point is input into the pre-trained internal thermal identification model to analyze the internal thermal anomaly feature set of the energy storage cabinet, including airflow cooling efficiency feature value, hot spot diffusion feature value, temperature difference imbalance fluctuation feature value, and hot spot aggregation risk feature value. Based on the set of internal thermal anomaly characteristics of the energy storage cabinet, the thermal field disorder characteristics of the set energy storage cabinet are analyzed.

8. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 7, characterized in that, The specific steps for analyzing and setting the characteristic set of internal thermal anomalies of the energy storage cabinet are as follows: In the feature extraction subnetwork of the cabinet thermal identification model, the cabinet thermal imaging data of the set energy storage cabinet at each time point is received, and the cabinet thermal field feature vector of the corresponding time point is extracted. In the temporal correlation sub-network of the cabinet thermal identification model, the cabinet thermal anomaly feature set of the set energy storage cabinet is analyzed based on the cabinet thermal field feature vector at each time point of the set energy storage cabinet.

9. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 8, characterized in that, The feature extraction subnetwork includes an input preprocessing layer and a feature construction output layer. The specific steps for extracting the internal thermal field feature vector of the energy storage cabinet at each time point are as follows: In the input preprocessing layer of the feature extraction subnetwork, thermal imaging data of the energy storage cabinet at each time point is received and preprocessed. In the feature construction output layer of the feature extraction subnetwork, the feature vector of the internal thermal field of the set energy storage cabinet at each time point is analyzed based on the preprocessed thermal imaging data of the cabinet at each time point.

10. The energy storage cabinet thermal management system based on dynamic wind pressure regulation according to claim 1, characterized in that, The specific steps for dynamically adjusting the air pressure of a set energy storage cabinet based on thermal regulation characteristic values ​​are as follows: The thermal control characteristic values ​​of the set energy storage cabinet are compared and analyzed with the preset thermal control characteristic range; Based on the comparative analysis results, dynamic air pressure adjustment is performed on the set energy storage cabinet.

Images