Simulation analysis method of dual-path liquid-cooled energy storage system

Abstract

The application discloses a simulation analysis method of a two-way liquid cooling energy storage system and belongs to the technical field of two-way liquid cooling energy storage systems, which comprises the following steps: constructing a simulation analysis mathematical model of the two-way liquid cooling energy storage system; setting boundary conditions and collecting real-time data of the two-way liquid cooling energy storage system, simulating the real-time data of the two-way liquid cooling energy storage system according to the simulation analysis mathematical model of the two-way liquid cooling energy storage system, and determining simulation analysis results of the two-way liquid cooling energy storage system; and optimizing the design of cooling plates, the layout of flow channels or the flow of cooling liquid according to the simulation analysis results of the two-way liquid cooling energy storage system, so as to optimize thermal management design. The application solves the problem that the existing two-way liquid cooling energy storage system cannot be effectively simulated and analyzed, thermal management design cannot be effectively optimized, system performance cannot be improved, and the use effect is poor. The application can effectively simulate and analyze the two-way liquid cooling energy storage system, effectively optimize thermal management design, improve system performance, and improve the use effect of the two-way liquid cooling energy storage system.

Description

Technical Field

[0001] This invention relates to the field of dual-channel liquid-cooled energy storage system technology, specifically to a simulation analysis method for dual-channel liquid-cooled energy storage systems. Background Technology

[0002] The dual-channel liquid-cooled energy storage system consists of the following key components: 1) Coolant: typically a water-glycol solution with high thermal conductivity; 2) Cooling plates / pipes: in direct contact with the battery modules to absorb heat; 3) Coolant pump: drives the coolant to circulate within the system; 4) Heat exchanger: dissipates heat from the coolant through air or water cooling; 5) Storage tank: stores the coolant and provides circulation power; 6) Control system: includes temperature, flow rate, and pressure sensors, as well as an intelligent control module for real-time adjustment of system operation.

[0003] The coolant pump draws coolant from the storage tank, flows through the cooling plate to absorb heat from the battery module, then enters the heat exchanger to release heat, and then flows back to the storage tank to form a cycle. This design ensures precise control of battery temperature and improves heat dissipation efficiency.

[0004] Existing technologies cannot effectively simulate and analyze dual-channel liquid-cooled energy storage systems, nor can they effectively optimize thermal management design and improve system performance, resulting in poor performance of dual-channel liquid-cooled energy storage systems. Summary of the Invention

[0005] The purpose of this invention is to provide a simulation analysis method for a dual-channel liquid-cooled energy storage system, which can effectively simulate and analyze the dual-channel liquid-cooled energy storage system, effectively optimize thermal management design and improve system performance, enhance the use effect of the dual-channel liquid-cooled energy storage system, and solve the problems mentioned in the background art.

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

[0007] Simulation analysis methods for dual-channel liquid-cooled energy storage systems include:

[0008] A mathematical model for simulation analysis of a dual-channel liquid-cooled energy storage system is constructed, and the mathematical model is deployed in a real dual-channel liquid-cooled energy storage system simulation analysis environment.

[0009] Set boundary conditions and collect real-time data of the dual-channel liquid-cooled energy storage system. Simulate the real-time data of the dual-channel liquid-cooled energy storage system based on the mathematical model of the simulation analysis of the dual-channel liquid-cooled energy storage system, and determine the simulation analysis results of the dual-channel liquid-cooled energy storage system.

[0010] Based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, optimize the design of the cooling plate, the flow channel layout, or the coolant flow rate, thereby optimizing the thermal management design.

[0011] Preferably, a mathematical model for simulation analysis of a dual-channel liquid-cooled energy storage system is constructed, and the following operations are performed:

[0012] Historical data of dual-channel liquid-cooled energy storage system was collected and divided into training set and test set.

[0013] Based on machine learning technology, a training set is used to train the machine learning model, enabling the machine learning model to autonomously learn the simulation analysis behavior of the dual-channel liquid-cooled energy storage system from the training set, automatically analyze the coolant flow and temperature distribution, and thus determine the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system.

[0014] The mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system was tested and verified based on the test set, and the analytical performance of the mathematical model was evaluated to determine the test results of the mathematical model.

[0015] Based on the test results of the mathematical model, the parameters of the simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system are optimized to determine the optimal simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system.

[0016] Preferably, a simulation analysis is performed on the dual-channel liquid-cooled energy storage system, and the following operations are performed:

[0017] The mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is deployed in the actual simulation analysis environment of the dual-channel liquid-cooled energy storage system.

[0018] Set boundary conditions, including coolant velocity, flow rate, temperature, pressure, and flow direction;

[0019] Real-time data from the dual-channel liquid-cooled energy storage system is input into the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. Based on the mathematical model, the real-time data of the dual-channel liquid-cooled energy storage system is simulated to analyze the flow, pressure, and temperature distribution of the coolant, obtain temperature field, pressure field, and flow field data, and determine the simulation analysis results of the dual-channel liquid-cooled energy storage system.

[0020] Preferably, real-time data from the dual-channel liquid-cooled energy storage system is collected, and the following operations are performed:

[0021] A geometric model of the dual-channel liquid-cooled energy storage system was established using 3D modeling software, including cooling plates, coolant, cold liquid pump, heat exchanger, storage tank, and battery module components.

[0022] The coolant pump draws coolant from the storage tank, flows through the cooling plate to absorb heat from the battery module, then enters the heat exchanger to release heat, and then flows back to the storage tank to form a cycle.

[0023] The battery module's capacity, internal resistance, charge / discharge curve, and heat generation rate are monitored and collected in real time using intelligent data acquisition equipment to obtain battery module data.

[0024] Based on intelligent data acquisition equipment, the flow rate, flow rate, temperature, pressure, flow direction, specific heat capacity, thermal conductivity and viscosity of the coolant are monitored and collected in real time to obtain coolant data;

[0025] Based on battery module data and coolant data, real-time data for a dual-channel liquid-cooled energy storage system is generated.

[0026] Preferably, the real-time data of the dual-channel liquid-cooled energy storage system is processed by performing the following operations:

[0027] The real-time data of the dual-channel liquid-cooled energy storage system is cleaned to remove noise, and outliers related to the simulation analysis of the dual-channel liquid-cooled energy storage system are identified and corrected.

[0028] The real-time data of the dual-channel liquid-cooled energy storage system is converted into a unified data format, and the dimensional differences in the real-time data of the dual-channel liquid-cooled energy storage system are removed to form standardized real-time data of the dual-channel liquid-cooled energy storage system.

[0029] Preferably, the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is tested and verified based on the test set, and the following operations are performed:

[0030] The test set is input into the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. The mathematical model is tested according to the test set to verify its simulation performance and evaluate whether it can achieve the expected effect of simulation analysis of the dual-channel liquid-cooled energy storage system.

[0031] When the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system can achieve the expected effect of simulation analysis of the dual-channel liquid-cooled energy storage system, the optimal mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is directly determined.

[0032] When the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system fails to achieve the expected results for simulation analysis of the dual-channel liquid-cooled energy storage system, the parameters of the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system are adjusted and optimized until the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system can achieve the expected results for simulation analysis of the dual-channel liquid-cooled energy storage system, thereby determining the optimal mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system.

[0033] Preferably, based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, the cooling plate design, flow channel layout, or coolant flow rate are optimized, and the following operations are performed:

[0034] Based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, the temperature distribution of the battery module is analyzed, its thermal uniformity and whether there are excessive hot spots are evaluated, and the design of the cooling plate, the flow channel layout or the coolant flow rate are optimized according to the coolant flow, pressure and temperature distribution, thereby optimizing the thermal management design.

[0035] Preferably, to optimize the thermal management design, perform the following operations:

[0036] Based on the velocity field data of the coolant in different regions, the flow channel design is adjusted, the cross-sectional shape of the flow channel is optimized, or regions with higher flow velocities are added to improve the flow efficiency of the coolant.

[0037] Based on the pressure loss during the coolant flow process, focus on the pressure drop of elbows, valves, and pump components, optimize the flow channel layout, reduce elbows and local resistance components, or adjust the coolant flow rate to reduce pressure loss.

[0038] Based on the temperature field data of the coolant in different areas, we focus on the temperature difference between the coolant inlet and outlet and the uniformity of the coolant temperature distribution. By adjusting the coolant flow rate or the design of the cooling plate, we optimize the heat exchange efficiency between the coolant and the battery module.

[0039] Based on the temperature field distribution of the battery module, the temperature difference between individual battery cells is calculated, and the thermal uniformity between battery modules is improved by adjusting the cooling plate design or optimizing the coolant flow rate.

[0040] Based on the local high-temperature areas in the battery module temperature field, combined with heat flow vector analysis, areas where the temperature exceeds a set threshold are identified. For overheated areas, the coolant flow path is optimized, or the local heat dissipation capacity of the cooling plate is increased.

[0041] Preferably, the temperature distribution of the battery module is analyzed to assess its thermal uniformity and the presence of hot spots, including:

[0042] Obtain the temperature value of each liquid cooling circuit in the battery module at each time point;

[0043] The temperature deviation is obtained by calculating the difference between the temperature value at each time point and the average temperature in the liquid cooling circuit.

[0044] The temperature deviations in the liquid cooling circuit are spatiotemporally weighted and accumulated, and the ratio of this sum to the maximum sum of squares of deviations is used as the first ratio in the liquid cooling circuit.

[0045] The average of the first ratios in each liquid cooling circuit is used as the thermal uniformity index of the battery module.

[0046]

[0047] in, represents the thermal uniformity index of the i-th battery module; m represents the total number of liquid cooling circuits in a single battery module. This represents the temperature value at time point k in the liquid cooling circuit of the i-th battery module; This represents the total number of points in time. This represents the average temperature in the i-th battery module; Represents a discrete window function; This represents the maximum allowable sum of squares of design deviations;

[0048] The thermal uniformity index of the battery module is compared with the preset thermal uniformity index threshold. If the thermal uniformity index of the battery module is greater than or equal to the preset thermal uniformity index threshold, it indicates that the thermal uniformity of the battery module is poor; if the thermal uniformity index of the battery module is less than the preset thermal uniformity index threshold, it indicates that the thermal uniformity of the battery module is good.

[0049] The difference between the temperature value at each time point and the overheating temperature threshold in the liquid cooling circuit is calculated to obtain the overheating temperature difference.

[0050] The overheating temperature difference in the liquid cooling circuit is weighted in time and space and accumulated, and the ratio of this sum to the maximum overheating integral is used to determine the overheating index of the battery module.

[0051]

[0052] in, This represents the overheating index of the i-th battery module; Indicates characteristic functions; Indicates the overheating temperature threshold; This represents the maximum allowable superheat integral. Indicates the overheating penalty coefficient;

[0053] The overheating index of the battery module is compared with the preset overheating index threshold. If the overheating index of the battery module is greater than or equal to the preset overheating index threshold, it indicates that there is a serious overheating in the battery module; if the overheating index of the battery module is less than the preset overheating index threshold, it indicates that there is no overheating in the battery module.

[0054] Preferably, historical data of the dual-channel liquid-cooled energy storage system is collected, and the collected historical data of the dual-channel liquid-cooled energy storage system is divided into the following categories:

[0055] Collect historical data of the dual-channel liquid-cooled energy storage system, obtain battery module data and coolant data from the historical data of the dual-channel liquid-cooled energy storage system, and bind the battery module data with the coolant data in the corresponding cooling circuit;

[0056] The bound battery module data is aligned with the coolant data in the corresponding cooling circuit based on the timestamp to obtain aligned data;

[0057] Treat all aligned data as a system-wide global dataset;

[0058] Obtain the thermocouple array data on the surface of the battery module from the battery module data, and interpolate and complete the thermocouple array data to generate battery module temperature sensor array data.

[0059] Obtain the flow direction values ​​at different locations in the dual-circuit coolant data, and sort the flow direction values ​​at different locations according to spatial coordinates to generate coolant flow direction spatial distribution vector data;

[0060] Obtain the pressure values ​​of key monitoring points in the pipeline from the coolant data, arrange the pressure values ​​of the key location pressure value sequence in the position order to generate the pipeline pressure value sequence, and use the pipeline pressure value sequence as the pipeline pressure monitoring point data.

[0061] Obtain the flow velocity profile data of several sampling points included in the pipe cross section of the coolant data, as the sampled value data of the coolant flow velocity field;

[0062] The battery module temperature sensor array data, coolant flow direction spatial distribution vector data, pipeline pressure monitoring point data, and coolant flow velocity field sampling value data are used as the basic layer spatial features of the system's global dataset.

[0063] Based on the battery module temperature sensor array data and coolant flow velocity field sampling data in the basic layer spatial characteristics, the temperature gradient-flow velocity coupling factor is determined.

[0064] Based on the pipeline pressure monitoring point data and coolant flow data in the spatial characteristics of the basic layer, the pressure fluctuation-flow correlation index is determined.

[0065] The temperature gradient-velocity coupling factor and pressure fluctuation-flow correlation index are used to generate the inter-field coupling features of the derived layer based on the time series, and the inter-field coupling features of the derived layer are time-aligned with the features of the base layer.

[0066] Interpolate and expand the dimensions of the sampled data of the coolant flow velocity field and the data of the battery module temperature sensor array to generate the flow field distribution matrix and the temperature field distribution matrix; expand the flow field distribution matrix and the temperature field distribution matrix into one-dimensional vectors and calculate the Pearson correlation coefficient between the flow field distribution matrix and the temperature field distribution matrix;

[0067] The pressure gradient is determined based on the pressure monitoring data of the pipeline, and the correlation between the pressure gradient and the designed contact area of ​​the cooling plate is calculated.

[0068] The Pearson correlation coefficient and the correlation degree are used to generate global features of the correlation layer based on time series, and the global features of the correlation layer are aligned with the spatial features of the base layer and the field coupling features of the derived layer; the aligned spatial features of the base layer, the field coupling features of the derived layer and the global features of the correlation layer are used as the total feature set of the system's global dataset;

[0069] The total feature set is divided based on a preset ratio.

[0070] Compared with the prior art, the beneficial effects of the present invention are:

[0071] This invention constructs a mathematical model for the simulation analysis of a dual-channel liquid-cooled energy storage system and deploys this model in a real-world simulation environment. Boundary conditions are set, and real-time data from the dual-channel liquid-cooled energy storage system is collected. Based on the mathematical model, the real-time data is simulated to analyze the coolant flow, pressure, and temperature distribution, obtaining temperature, pressure, and flow field data. The simulation analysis results are then determined, and the cooling plate design, flow channel layout, or coolant flow rate are optimized based on these results, thereby improving thermal management design. This invention effectively simulates and analyzes dual-channel liquid-cooled energy storage systems, optimizes thermal management design, and enhances system performance, ultimately improving the overall effectiveness of the dual-channel liquid-cooled energy storage system. Attached Figure Description

[0072] Figure 1 This is a flowchart of the simulation analysis method for the dual-channel liquid-cooled energy storage system of the present invention. Detailed Implementation

[0073] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0074] To address the current limitations in effectively simulating and analyzing dual-channel liquid-cooled energy storage systems, thus hindering the optimization of thermal management design and performance improvement, and ultimately resulting in poor performance of dual-channel liquid-cooled energy storage systems, please refer to [link / reference needed]. Figure 1 This embodiment provides the following technical solution:

[0075] Simulation analysis methods for dual-channel liquid-cooled energy storage systems include:

[0076] A mathematical model for simulation analysis of a dual-channel liquid-cooled energy storage system was constructed, and the mathematical model was deployed in a real dual-channel liquid-cooled energy storage system simulation analysis environment.

[0077] In this embodiment, a mathematical model for simulation analysis of a dual-channel liquid-cooled energy storage system is constructed, and the following operations are performed:

[0078] Historical data of dual-channel liquid-cooled energy storage system was collected and divided into training set and test set.

[0079] Based on machine learning technology, a training set is used to train the machine learning model, enabling the machine learning model to autonomously learn the simulation analysis behavior of the dual-channel liquid-cooled energy storage system from the training set, automatically analyze the coolant flow and temperature distribution, and thus determine the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system.

[0080] The mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system was tested and verified based on the test set, and the analytical performance of the mathematical model was evaluated to determine the test results of the mathematical model.

[0081] The test set is input into the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. The mathematical model is then tested based on the test set to verify its simulation performance and to evaluate whether it can achieve the expected results for simulating the dual-channel liquid-cooled energy storage system.

[0082] When the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system can achieve the expected effect of simulation analysis of the dual-channel liquid-cooled energy storage system, the optimal mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is directly determined.

[0083] When the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system fails to achieve the expected results for simulation analysis of the dual-channel liquid-cooled energy storage system, the parameters of the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system are adjusted and optimized until the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system can achieve the expected results for simulation analysis of the dual-channel liquid-cooled energy storage system, thereby determining the optimal mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system.

[0084] Specifically, the mathematical model of the dual-channel liquid-cooled energy storage system was tested and verified based on the test set. The test results of the mathematical model are shown in Table 1:

[0085] Table 1: Test Results of Mathematical Model

[0086]

[0087] Therefore, by constructing a mathematical model for the simulation analysis of a dual-channel liquid-cooled energy storage system and deploying this model in a real dual-channel liquid-cooled energy storage system simulation analysis environment, it is easier to conduct subsequent simulation analysis of the dual-channel liquid-cooled energy storage system.

[0088] Set boundary conditions and collect real-time data of the dual-channel liquid-cooled energy storage system. Simulate the real-time data of the dual-channel liquid-cooled energy storage system based on the mathematical model of the simulation analysis of the dual-channel liquid-cooled energy storage system, and determine the simulation analysis results of the dual-channel liquid-cooled energy storage system.

[0089] In this embodiment, real-time data from the dual-channel liquid-cooled energy storage system is collected, and the following operations are performed:

[0090] A geometric model of the dual-channel liquid-cooled energy storage system was established using 3D modeling software, including cooling plates, coolant, cold liquid pump, heat exchanger, storage tank, and battery module components.

[0091] The coolant pump draws coolant from the storage tank, flows through the cooling plate to absorb heat from the battery module, then enters the heat exchanger to release heat, and then flows back to the storage tank to form a cycle.

[0092] The battery module's capacity, internal resistance, charge / discharge curve, and heat generation rate are monitored and collected in real time using intelligent data acquisition equipment to obtain battery module data.

[0093] Based on intelligent data acquisition equipment, the flow rate, flow rate, temperature, pressure, flow direction, specific heat capacity, thermal conductivity and viscosity of the coolant are monitored and collected in real time to obtain coolant data;

[0094] Specifically, the coolant data is shown in Table 2:

[0095] Table 2: Coolant Data

[0096]

[0097] Based on battery module data and coolant data, real-time data of the dual-channel liquid-cooled energy storage system is generated. This data is then used to simulate the real-time data of the dual-channel liquid-cooled energy storage system according to the mathematical model for simulation analysis, and to analyze the coolant flow, pressure, and temperature distribution.

[0098] In this embodiment, the real-time data of the dual-channel liquid-cooled energy storage system is processed, and the following operations are performed:

[0099] Cleaning the real-time data of the dual-channel liquid-cooled energy storage system removes noise and identifies outliers related to the simulation analysis of the dual-channel liquid-cooled energy storage system. Correcting these outliers improves the data quality of the real-time data of the dual-channel liquid-cooled energy storage system.

[0100] The real-time data of the dual-channel liquid-cooled energy storage system is converted into a unified data format, and the dimensional differences in the real-time data of the dual-channel liquid-cooled energy storage system are removed to form standardized real-time data of the dual-channel liquid-cooled energy storage system.

[0101] In this embodiment, a simulation analysis is performed on a dual-channel liquid-cooled energy storage system, and the following operations are performed:

[0102] The mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is deployed in the actual simulation analysis environment of the dual-channel liquid-cooled energy storage system.

[0103] Set boundary conditions, including coolant velocity, flow rate, temperature, pressure, and flow direction;

[0104] Real-time data from the dual-channel liquid-cooled energy storage system is input into the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. Based on the mathematical model, the real-time data of the dual-channel liquid-cooled energy storage system is simulated to analyze the flow, pressure, and temperature distribution of the coolant, obtain temperature field, pressure field, and flow field data, and determine the simulation analysis results of the dual-channel liquid-cooled energy storage system.

[0105] Based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, optimize the design of the cooling plate, the flow channel layout, or the coolant flow rate, thereby optimizing the thermal management design.

[0106] In this embodiment, the cooling plate design, flow channel layout, or coolant flow rate are optimized based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, and the following operations are performed:

[0107] Based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, the temperature distribution of the battery module is analyzed, its thermal uniformity and whether there are excessive hot spots are evaluated, and the design of the cooling plate, the flow channel layout or the coolant flow rate are optimized according to the coolant flow, pressure and temperature distribution, thereby optimizing the thermal management design.

[0108] In this embodiment, the thermal management design is optimized by performing the following operations:

[0109] Based on the velocity field data of the coolant in different regions, the flow channel design is adjusted, the cross-sectional shape of the flow channel is optimized, or regions with higher flow velocities are added to improve the flow efficiency of the coolant.

[0110] Based on the pressure loss during the coolant flow process, focus on the pressure drop of elbows, valves, and pump components, optimize the flow channel layout, reduce elbows and local resistance components, or adjust the coolant flow rate to reduce pressure loss.

[0111] Based on the temperature field data of the coolant in different areas, we focus on the temperature difference between the coolant inlet and outlet and the uniformity of the coolant temperature distribution. By adjusting the coolant flow rate or the design of the cooling plate, we optimize the heat exchange efficiency between the coolant and the battery module.

[0112] Based on the temperature field distribution of the battery module, the temperature difference between individual battery cells is calculated, and the thermal uniformity between battery modules is improved by adjusting the cooling plate design or optimizing the coolant flow rate.

[0113] Based on the local high-temperature areas in the battery module temperature field, combined with heat flow vector analysis, areas where the temperature exceeds a set threshold are identified. For overheated areas, the coolant flow path is optimized, or the local heat dissipation capacity of the cooling plate is increased.

[0114] In summary, based on the mathematical model of the dual-channel liquid-cooled energy storage system simulation analysis, the real-time data of the dual-channel liquid-cooled energy storage system is simulated to analyze the coolant flow, pressure, and temperature distribution, obtain temperature field, pressure field, and flow field data, determine the simulation analysis results of the dual-channel liquid-cooled energy storage system, and optimize the cooling plate design, flow channel layout, or coolant flow rate based on the simulation analysis results, thereby optimizing the thermal management design. This method can effectively simulate and analyze the dual-channel liquid-cooled energy storage system, effectively optimize the thermal management design, improve system performance, and enhance the usage effect of the dual-channel liquid-cooled energy storage system.

[0115] In this embodiment, the temperature distribution of the battery module is analyzed to evaluate its thermal uniformity and the presence of hot spots, including:

[0116] Obtain the temperature value of each liquid cooling circuit in the battery module at each time point;

[0117] The temperature deviation is obtained by calculating the difference between the temperature value at each time point and the average temperature in the liquid cooling circuit.

[0118] The temperature deviations in the liquid cooling circuit are spatiotemporally weighted and accumulated, and the ratio of this sum to the maximum sum of squares of deviations is used as the first ratio in the liquid cooling circuit.

[0119] The average of the first ratios in each liquid cooling circuit is used as the thermal uniformity index of the battery module.

[0120]

[0121] in, represents the thermal uniformity index of the i-th battery module; m represents the total number of liquid cooling circuits in a single battery module. This represents the temperature value at time point k in the liquid cooling circuit of the i-th battery module; This represents the total number of points in time. This represents the average temperature in the i-th battery module; Represents a discrete window function; This represents the maximum allowable sum of squares of design deviations;

[0122] The thermal uniformity index of the battery module is compared with the preset thermal uniformity index threshold. If the thermal uniformity index of the battery module is greater than or equal to the preset thermal uniformity index threshold, it indicates that the thermal uniformity of the battery module is poor; if the thermal uniformity index of the battery module is less than the preset thermal uniformity index threshold, it indicates that the thermal uniformity of the battery module is good.

[0123] The difference between the temperature value at each time point and the overheating temperature threshold in the liquid cooling circuit is calculated to obtain the overheating temperature difference.

[0124] The overheating temperature difference in the liquid cooling circuit is weighted in time and space and accumulated, and the ratio of this sum to the maximum overheating integral is used to determine the overheating index of the battery module.

[0125]

[0126] in, This represents the overheating index of the i-th battery module; Indicates characteristic functions; Indicates the overheating temperature threshold; This represents the maximum allowable superheat integral. Indicates the overheating penalty coefficient;

[0127] The overheating index of the battery module is compared with the preset overheating index threshold. If the overheating index of the battery module is greater than or equal to the preset overheating index threshold, it indicates that there is a serious overheating in the battery module; if the overheating index of the battery module is less than the preset overheating index threshold, it indicates that there is no overheating in the battery module.

[0128] In this embodiment, This refers to discrete window functions, including but not limited to the Hanning window. Temperature fluctuations are large during startup / shutdown, so a window function is used to reduce its weight.

[0129] In this embodiment, the indicator function is 1 when the temperature exceeds the threshold and 0 otherwise; its function is to accurately screen the time of overheating, only count the temperature contribution of the real overheating point, and exclude the interference of normal temperature.

[0130] In this embodiment, The smaller the value, the higher the requirements for system heat dissipation design.

[0131] In this embodiment, Value selection logic: For example, if the over-temperature is ≤1℃ and lasts for ≤10 seconds, then the upper limit of the over-temperature amplitude is 1℃, and the allowed duration is 10 seconds (values ​​can be 1~10, 10 is used as an example here); Window function sum: If it is stable within 10 seconds, the rectangular window sum... Unit: ℃ Time step.

[0132] In this embodiment, This represents the overheating penalty coefficient; and The value ranges from 2 to 5. The larger the size, the higher the safety requirements.

[0133] In this embodiment, the energy storage system typically consists of multiple battery modules, and each module contains multiple liquid cooling circuits. Each module and each circuit needs to be evaluated independently to accurately pinpoint areas with poor thermal performance. By introducing a module number *i* and the number of circuits *m*, a refined evaluation of the entire energy storage system can be achieved, analyzing thermal uniformity issues from both module and circuit perspectives, facilitating subsequent targeted optimization and improvement. During battery charging and discharging, the battery temperature continuously changes over time, and the temperature response of different circuits also varies. A reference value is needed to measure the degree of temperature deviation in each circuit. It records the temperature of each loop at every moment, and can capture the dynamic change process of temperature; The weighted average temperature of the module reflects its overall temperature level. Using it as a benchmark to calculate the temperature deviation of each circuit effectively assesses temperature differences between circuits and determines the evenness of heat dissipation from the liquid cooling system. However, during battery operation, temperature fluctuations are significant during startup and shutdown, and these transient data can interfere with the accurate assessment of thermal uniformity. Therefore, a standard is needed to measure the acceptable level of temperature deviation. As a window function, by weighting the temperature over different time periods, it highlights the temperature data in the steady state and suppresses transient interference, making the evaluation results more reflective of the thermal uniformity of the battery under stable operating conditions. It is the maximum allowable sum of squares of deviations, used to normalize temperature deviations, making the thermal uniformity index comparable for different systems or operating conditions, facilitating the setting of qualified standards for thermal uniformity, and judging whether the heat dissipation design of the liquid cooling system meets the requirements.

[0134] In this embodiment, not all temperatures are meaningful for overheat assessment; it is necessary to screen out the moments when temperatures are truly exceeded. At the same time, it is necessary to clearly define the temperature standard for overheating points, because different types of batteries have different temperature tolerances. Indicative functions can accurately filter out the times when the temperature exceeds the threshold, eliminate interference from normal temperature periods, and focus the calculation on data related to hot spots; The overheating temperature threshold is set based on the battery's thermal runaway characteristics and industry safety standards to ensure that anomalies can be detected through the overheating index before the battery may experience thermal runaway risks. The severity of overheating is related not only to the temperature of the overheating but also to the duration of the overheating. It is necessary to quantify the overheating range and set an upper limit for the total allowable overheating to measure whether the harm of the overheating is within an acceptable range. The calculated overheating range directly reflects the degree of danger of the overheating point; The maximum allowable overheat integral is designed to comprehensively assess the harm of overheating to the battery module from the perspective of "overheating amplitude × duration", providing a quantitative basis for judging whether there is a serious overheating risk.

[0135] The working principle and beneficial effects of the above technical solution are as follows: by obtaining the temperature values ​​of each liquid cooling circuit in the battery module at each time point in detail, the thermal state of the battery module can be accurately quantified and analyzed; and the internal thermal behavior of the battery module can be deeply understood, rather than relying solely on overall temperature measurement or rough estimation.

[0136] In this embodiment, historical data of the dual-channel liquid-cooled energy storage system is collected, and the collected historical data of the dual-channel liquid-cooled energy storage system is divided, including:

[0137] Collect historical data of the dual-channel liquid-cooled energy storage system, obtain battery module data and coolant data from the historical data of the dual-channel liquid-cooled energy storage system, and bind the battery module data with the coolant data in the corresponding cooling circuit;

[0138] The bound battery module data is aligned with the coolant data in the corresponding cooling circuit based on the timestamp to obtain aligned data;

[0139] Treat all aligned data as a system-wide global dataset;

[0140] Obtain the thermocouple array data on the surface of the battery module from the battery module data, and interpolate and complete the thermocouple array data to generate battery module temperature sensor array data.

[0141] Obtain the flow direction values ​​at different locations in the dual-circuit coolant data, and sort the flow direction values ​​at different locations according to spatial coordinates to generate coolant flow direction spatial distribution vector data;

[0142] Obtain the pressure values ​​of key monitoring points in the pipeline from the coolant data, arrange the pressure values ​​of the key location pressure value sequence in the position order to generate the pipeline pressure value sequence, and use the pipeline pressure value sequence as the pipeline pressure monitoring point data.

[0143] Obtain the flow velocity profile data of several sampling points included in the pipe cross section of the coolant data, as the sampled value data of the coolant flow velocity field;

[0144] The battery module temperature sensor array data, coolant flow direction spatial distribution vector data, pipeline pressure monitoring point data, and coolant flow velocity field sampling value data are used as the basic layer spatial features of the system's global dataset.

[0145] Based on the battery module temperature sensor array data and coolant flow velocity field sampling data in the basic layer spatial characteristics, the temperature gradient-flow velocity coupling factor is determined.

[0146] Based on the pipeline pressure monitoring point data and coolant flow data in the spatial characteristics of the basic layer, the pressure fluctuation-flow correlation index is determined.

[0147] The temperature gradient-velocity coupling factor and pressure fluctuation-flow correlation index are used to generate the inter-field coupling features of the derived layer based on the time series, and the inter-field coupling features of the derived layer are time-aligned with the features of the base layer.

[0148] Interpolate and expand the dimensions of the sampled data of the coolant flow velocity field and the data of the battery module temperature sensor array to generate the flow field distribution matrix and the temperature field distribution matrix; expand the flow field distribution matrix and the temperature field distribution matrix into one-dimensional vectors and calculate the Pearson correlation coefficient between the flow field distribution matrix and the temperature field distribution matrix;

[0149] The pressure gradient is determined based on the pressure monitoring data of the pipeline, and the correlation between the pressure gradient and the designed contact area of ​​the cooling plate is calculated.

[0150] The Pearson correlation coefficient and the correlation degree are used to generate global features of the correlation layer based on time series, and the global features of the correlation layer are aligned with the spatial features of the base layer and the field coupling features of the derived layer; the aligned spatial features of the base layer, the field coupling features of the derived layer and the global features of the correlation layer are used as the total feature set of the system's global dataset;

[0151] The total feature set is divided based on a preset ratio.

[0152] In this embodiment, based on the battery module temperature sensor array data and coolant flow velocity field sampling data in the base layer spatial features, the temperature gradient-flow velocity coupling factor is determined, including:

[0153] The spatial gradient of the battery module temperature sensor array data is calculated, and the partial derivatives in the x and y directions are calculated using the central difference method. , Represents the temperature gradient; Representing the temperature field, it is a function of spatial location (which can be expressed as x and y coordinates in two dimensions, i.e., ...). Functions of time; Indicates temperature Partial derivative with respect to spatial coordinate x; Indicates temperature Partial derivative with respect to spatial coordinate y; extracting the velocity vector from the sampled data of the coolant flow velocity field. Take the average velocity of the cross section; calculate the dot product. ; This represents the temperature gradient-flow velocity coupling factor. This represents the component of the velocity vector v along the x-axis. This represents the component of the velocity vector v along the y-axis; if If the temperature gradient is greater than 0, the temperature gradient is in the same direction as the flow velocity, the coolant has a strong heat-carrying capacity, and the heat dissipation efficiency is high; if... When the values ​​are less than 0, the two directions are opposite, heat dissipation is hindered, and temperature accumulation is likely to occur.

[0154] In this embodiment, based on the pipeline pressure monitoring point data and the flow rate data in the coolant data within the spatial characteristics of the base layer, the pressure fluctuation-flow correlation index is determined, including:

[0155] Calculate the standard deviation of the pipeline pressure monitoring data, and use the ratio of the standard deviation to the mean flow rate in the coolant data as the pressure fluctuation-flow rate correlation index.

[0156] In this embodiment, the pressure gradient is the ratio of the pressure difference between the inlet and outlet of the cooling plate to the length of the flow path.

[0157] In this embodiment, the basic layer features are stored in tensor form with dimensions of (time step × spatial dimension × number of features); the derived layer features are time series scalars with dimensions of (N×2), where N is the time step, and are time-aligned with the basic layer features; the associated layer features are time series scalars with dimensions of (N×2), and together with the first two layers of features, they form the total feature set of the system's global dataset = [basic layer features, derived layer features, associated layer features].

[0158] The working principle and beneficial effects of the above technical solution are as follows: By collecting and integrating battery module data, coolant data, and various features extracted from these data, including basic layer spatial features, derived layer inter-field coupling features, and correlation layer global features, the state of the dual-path liquid-cooled energy storage system can be comprehensively described. This multi-dimensional description can cover multiple aspects such as the temperature distribution of the battery module, the flow characteristics of the coolant, the pressure state, and the interrelationships between them, thus providing a complete picture of the system's operating state. By performing operations such as data interpolation, sorting, and calculating coupling factors and correlation indices, key features reflecting the system's operating state can be accurately extracted. The total feature set of the system's global dataset contains rich feature information, providing a good data foundation for training the model.

[0159] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0160] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A simulation analysis method for a dual-channel liquid-cooled energy storage system, characterized in that, include: A mathematical model for simulation analysis of a dual-channel liquid-cooled energy storage system is constructed, and the mathematical model is deployed in a real dual-channel liquid-cooled energy storage system simulation analysis environment. Set boundary conditions and collect real-time data of the dual-channel liquid-cooled energy storage system. Simulate the real-time data of the dual-channel liquid-cooled energy storage system based on the mathematical model of the simulation analysis of the dual-channel liquid-cooled energy storage system, and determine the simulation analysis results of the dual-channel liquid-cooled energy storage system. Optimize the cooling plate design, flow channel layout, or coolant flow rate based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, and optimize the thermal management design. Collect real-time data from the dual-channel liquid-cooled energy storage system and perform the following operations: A geometric model of the dual-channel liquid-cooled energy storage system was established using 3D modeling software, including cooling plates, coolant, cold liquid pump, heat exchanger, storage tank, and battery module components. The coolant pump draws coolant from the storage tank, flows through the cooling plate to absorb heat from the battery module, then enters the heat exchanger to release heat, and then flows back to the storage tank to form a cycle. The battery module's capacity, internal resistance, charge / discharge curve, and heat generation rate are monitored and collected in real time using intelligent data acquisition equipment to obtain battery module data. Based on intelligent data acquisition equipment, the flow rate, flow rate, temperature, pressure, flow direction, specific heat capacity, thermal conductivity and viscosity of the coolant are monitored and collected in real time to obtain coolant data; Based on battery module data and coolant data, real-time data for a dual-channel liquid-cooled energy storage system is generated.

2. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 1, characterized in that, Construct a mathematical model for simulation analysis of a dual-channel liquid-cooled energy storage system and perform the following operations: Historical data of dual-channel liquid-cooled energy storage system was collected and divided into training set and test set. Based on machine learning technology, a training set is used to train the machine learning model, enabling the machine learning model to autonomously learn the simulation analysis behavior of the dual-channel liquid-cooled energy storage system from the training set, automatically analyze the coolant flow and temperature distribution, and thus determine the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. The mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system was tested and verified based on the test set, and the analytical performance of the mathematical model was evaluated to determine the test results of the mathematical model. Based on the test results of the mathematical model, the parameters of the simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system are optimized to determine the optimal simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system.

3. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 2, characterized in that, A simulation analysis was performed on the dual-channel liquid-cooled energy storage system, and the following operations were performed: The mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is deployed in the actual simulation analysis environment of the dual-channel liquid-cooled energy storage system. Set boundary conditions, including coolant velocity, flow rate, temperature, pressure, and flow direction; Real-time data from the dual-channel liquid-cooled energy storage system is input into the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. Based on the mathematical model, the real-time data of the dual-channel liquid-cooled energy storage system is simulated to analyze the flow, pressure, and temperature distribution of the coolant, obtain temperature field, pressure field, and flow field data, and determine the simulation analysis results of the dual-channel liquid-cooled energy storage system.

4. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 3, characterized in that, The real-time data from the dual-channel liquid-cooled energy storage system is processed, and the following operations are performed: The real-time data of the dual-channel liquid-cooled energy storage system is cleaned to remove noise, and outliers related to the simulation analysis of the dual-channel liquid-cooled energy storage system are identified and corrected. The real-time data of the dual-channel liquid-cooled energy storage system is converted into a unified data format, and the dimensional differences in the real-time data of the dual-channel liquid-cooled energy storage system are removed to form standardized real-time data of the dual-channel liquid-cooled energy storage system.

5. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 4, characterized in that, The mathematical model of the dual-channel liquid-cooled energy storage system simulation analysis was tested and verified based on the test set. The following operations were performed: The test set is input into the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system. The mathematical model is tested according to the test set to verify its simulation performance and evaluate whether it can achieve the expected effect of simulation analysis of the dual-channel liquid-cooled energy storage system. When the mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system can achieve the expected effect of simulation analysis of the dual-channel liquid-cooled energy storage system, the optimal mathematical model for simulation analysis of the dual-channel liquid-cooled energy storage system is directly determined. When the simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system cannot achieve the expected effect of the simulation analysis of the dual-channel liquid-cooled energy storage system, the parameters of the simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system are adjusted and optimized until the simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system can achieve the expected effect of the simulation analysis of the dual-channel liquid-cooled energy storage system, and the optimal simulation analysis mathematical model of the dual-channel liquid-cooled energy storage system is determined.

6. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 5, characterized in that, Based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, optimize the cooling plate design, flow channel layout, or coolant flow rate by performing the following operations: Based on the simulation analysis results of the dual-channel liquid-cooled energy storage system, the temperature distribution of the battery module is analyzed, its thermal uniformity and whether there are hot spots are evaluated, and the design of the cooling plate, the flow channel layout or the coolant flow rate are optimized according to the coolant flow, pressure and temperature distribution, thereby optimizing the thermal management design.

7. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 6, characterized in that, To optimize thermal management design, perform the following operations: Based on the velocity field data of the coolant in different regions, the flow channel design is adjusted, the cross-sectional shape of the flow channel is optimized, or regions with higher flow velocities are added to improve the flow efficiency of the coolant. Based on the pressure loss during the coolant flow process, focus on the pressure drop of elbows, valves, and pump components, optimize the flow channel layout, reduce elbows and local resistance components, or adjust the coolant flow rate to reduce pressure loss. Based on the temperature field data of the coolant in different areas, we focus on the temperature difference between the coolant inlet and outlet and the uniformity of the coolant temperature distribution. By adjusting the coolant flow rate or the design of the cooling plate, we optimize the heat exchange efficiency between the coolant and the battery module. Based on the temperature field distribution of the battery module, the temperature difference between individual battery cells is calculated, and the thermal uniformity between battery modules is improved by adjusting the cooling plate design or optimizing the coolant flow rate. Based on the local high-temperature areas in the battery module temperature field, combined with heat flow vector analysis, areas where the temperature exceeds a set threshold are identified. For overheated areas, the coolant flow path is optimized, or the local heat dissipation capacity of the cooling plate is increased.

8. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 7, characterized in that, Analyze the temperature distribution of the battery module to assess its thermal uniformity and the presence of hot spots, including: Obtain the temperature value of each liquid cooling circuit in the battery module at each time point; The temperature deviation is obtained by calculating the difference between the temperature value at each time point and the average temperature in the liquid cooling circuit. The temperature deviations in the liquid cooling circuit are spatiotemporally weighted and accumulated, and the ratio of this sum to the maximum sum of squares of deviations is used as the first ratio in the liquid cooling circuit. The average of the first ratios in each liquid cooling circuit is used as the thermal uniformity index of the battery module. in, represents the thermal uniformity index of the i-th battery module; m represents the total number of liquid cooling circuits in a single battery module. This represents the temperature value at the k-th time point in the j-th liquid cooling circuit of the i-th battery module; This represents the total number of points in time. This represents the average temperature in the i-th battery module; Represents a discrete window function; This represents the maximum allowable sum of squares of design deviations; The thermal uniformity index of the battery module is compared with the preset thermal uniformity index threshold. If the thermal uniformity index of the battery module is greater than or equal to the preset thermal uniformity index threshold, it indicates that the thermal uniformity of the battery module is poor; if the thermal uniformity index of the battery module is less than the preset thermal uniformity index threshold, it indicates that the thermal uniformity of the battery module is good. The difference between the temperature value at each time point and the overheating temperature threshold in the liquid cooling circuit is calculated to obtain the overheating temperature difference. The overheating temperature difference in the liquid cooling circuit is weighted in time and space and accumulated, and the ratio of this sum to the maximum overheating integral is used to determine the overheating index of the battery module. in, This represents the overheating index of the i-th battery module; Indicates characteristic functions; Indicates the overheating temperature threshold; This represents the maximum allowable superheat integral. Indicates the overheating penalty coefficient; The overheating index of the battery module is compared with the preset overheating index threshold. If the overheating index of the battery module is greater than or equal to the preset overheating index threshold, it indicates that there is a serious overheating in the battery module; if the overheating index of the battery module is less than the preset overheating index threshold, it indicates that there is no overheating in the battery module.

9. The simulation analysis method for a dual-channel liquid-cooled energy storage system according to claim 2, characterized in that, Historical data of the dual-channel liquid-cooled energy storage system was collected and then divided into categories, including: Collect historical data of the dual-channel liquid-cooled energy storage system, obtain battery module data and coolant data from the historical data of the dual-channel liquid-cooled energy storage system, and bind the battery module data with the coolant data in the corresponding cooling circuit; The bound battery module data is aligned with the coolant data in the corresponding cooling circuit based on the timestamp to obtain aligned data; Treat all aligned data as a system-wide global dataset; Obtain the thermocouple array data on the surface of the battery module from the battery module data, and interpolate and complete the thermocouple array data to generate battery module temperature sensor array data. Obtain the flow direction values ​​at different locations in the dual-circuit coolant data, and sort the flow direction values ​​at different locations according to spatial coordinates to generate coolant flow direction spatial distribution vector data; Obtain the pressure values ​​of key monitoring points in the pipeline from the coolant data, arrange the pressure values ​​of the key location pressure value sequence in the position order to generate the pipeline pressure value sequence, and use the pipeline pressure value sequence as the pipeline pressure monitoring point data. Obtain the flow velocity profile data of several sampling points included in the pipe cross section of the coolant data, as the sampled value data of the coolant flow velocity field; The battery module temperature sensor array data, coolant flow direction spatial distribution vector data, pipeline pressure monitoring point data, and coolant flow velocity field sampling value data are used as the basic layer spatial features of the system's global dataset. Based on the battery module temperature sensor array data and coolant flow velocity field sampling data in the basic layer spatial characteristics, the temperature gradient-flow velocity coupling factor is determined. Based on the pipeline pressure monitoring point data and coolant flow data in the spatial characteristics of the basic layer, the pressure fluctuation-flow correlation index is determined. The temperature gradient-velocity coupling factor and pressure fluctuation-flow correlation index are used to generate the inter-field coupling features of the derived layer based on the time series, and the inter-field coupling features of the derived layer are time-aligned with the features of the base layer. Interpolate and expand the dimensions of the sampled data of the coolant flow velocity field and the data of the battery module temperature sensor array to generate the flow field distribution matrix and the temperature field distribution matrix; expand the flow field distribution matrix and the temperature field distribution matrix into one-dimensional vectors and calculate the Pearson correlation coefficient between the flow field distribution matrix and the temperature field distribution matrix; The pressure gradient is determined based on the pressure monitoring data of the pipeline, and the correlation between the pressure gradient and the designed contact area of ​​the cooling plate is calculated. The Pearson correlation coefficient and the correlation degree are used to generate global features of the correlation layer based on time series, and the global features of the correlation layer are aligned with the spatial features of the base layer and the field coupling features of the derived layer; the aligned spatial features of the base layer, the field coupling features of the derived layer and the global features of the correlation layer are used as the total feature set of the system's global dataset; The total feature set is divided based on a preset ratio.

Images