A multi-channel battery charging and discharging test system and method

Abstract

This invention relates to the field of battery testing technology, and more particularly to a multi-channel battery charge-discharge testing system and method. The method includes the following steps: acquiring voltage data of each battery cell in a multi-channel battery; performing voltage gradient stratification processing on the voltage data of each battery cell to divide the battery cell into multiple voltage intervals; detecting the number density of battery cells within each voltage interval to generate a voltage gradient distribution map; collecting the current distribution coefficients at each connection point of the multi-channel battery; performing current distribution difference analysis on the current distribution coefficients to calculate the actual current distribution ratio of each cell at different charge-discharge stages, generating current distribution difference data; and collecting overall temperature change data of the multi-channel battery during the charge-discharge process, as well as local temperature data of each battery cell. This invention, through battery testing technology, enables the analysis and prediction of the charge-discharge performance of multi-channel batteries, thereby improving the accuracy of multi-channel battery charge-discharge performance evaluation.

Description

Technical Field

[0001] This invention relates to the field of battery testing technology, and in particular to a multi-channel battery charge and discharge testing system and method. Background Technology

[0002] Multichannel batteries are typically composed of multiple battery cells connected in series, parallel, or a series-parallel combination. Under different connection methods, the current distribution and voltage superposition relationships between the cells are complex. In a series structure, even if each cell has the same nominal voltage, differences in internal resistance, capacity, etc., can lead to some cells being overcharged during charging and others being over-discharged during discharging. Existing testing methods struggle to accurately distinguish whether the voltage deviation is due to the connection method or a performance issue with the battery cell itself, affecting the accurate assessment of the battery's health. In a parallel structure, the current distribution among the cells is uneven, but it is difficult to accurately measure the actual current received by each cell during testing, making it impossible to accurately determine the differences in charge-discharge performance between cells, leading to misjudgments of the overall battery pack performance.

[0003] Furthermore, during long-term charge-discharge cycles, the electrode materials and electrolytes within multi-channel batteries undergo physical and chemical changes. Simultaneously, the internal connection structures of the battery pack may loosen or age due to mechanical vibrations, temperature variations, and other factors. Existing charge-discharge testing technologies struggle to monitor and adapt to these internal structural changes in real-time during long-term cycle testing, making it impossible to accurately assess the performance evolution and stability of the battery pack over long-term use. For example, expansion within battery cells leads to changes in internal pressure, affecting the battery's charge-discharge characteristics. However, existing testing methods often fail to capture the impact of these internal structural changes on performance in a timely manner, resulting in an inability to accurately predict the battery pack's lifespan and safety, and ultimately failing to provide effective assurance for the long-term reliable use of the battery pack. Summary of the Invention

[0004] Therefore, it is necessary to provide a multi-channel battery charge and discharge testing system and method to solve at least one of the above-mentioned technical problems.

[0005] To achieve the above objectives, a multi-channel battery charge-discharge testing method is provided, the method comprising the following steps:

[0006] Step S1: Obtain the voltage data of each battery cell in the multi-channel battery, perform voltage gradient layering processing on the voltage data of each battery cell, divide the battery cell into multiple voltage intervals; detect the number density of battery cells in each voltage interval, and generate a voltage gradient distribution map.

[0007] Step S2: Collect the current distribution coefficients of each connection point of the multi-channel battery; perform current distribution difference analysis on the current distribution coefficients, calculate the actual current distribution ratio of each unit in different charging and discharging stages, and generate current distribution difference data;

[0008] Step S3: Collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery; perform thermal flow field simulation analysis on the overall temperature change data and identify the heat propagation path inside the battery pack; combine the local temperature data of each battery cell to determine the impact of the heat propagation path on the current distribution difference data and generate temperature-current correlation data.

[0009] Step S4: Collect capacity data of each battery cell; perform capacity balance analysis on the capacity data of each battery cell, calculate the capacity change trend of each cell in different charge and discharge cycles, and generate battery capacity balance data.

[0010] Step S5: Based on the voltage gradient distribution map, temperature-current correlation data, and battery capacity balance data, construct a charge-discharge performance evaluation model; based on the charge-discharge performance evaluation model, predict the evolution trend of multi-channel battery performance and obtain a long-term battery performance evolution prediction report.

[0011] Preferably, in step S1, obtaining the voltage data of each battery cell in the multi-channel battery and performing voltage gradient hierarchical processing on the voltage data of each battery cell to divide the battery cell into multiple voltage ranges includes:

[0012] The voltage of each battery cell was measured three times using a voltmeter. The first measurement was taken when the battery cell was charged to 20%, the second measurement was taken when it was charged to 50%, and the third measurement was taken when it was charged to 80%.

[0013] The voltage data collected three times were averaged and rounded to two decimal places to obtain the average voltage value of each battery cell.

[0014] The battery cells are divided into three voltage ranges based on the average voltage value. When the average voltage value is lower than twice the arithmetic mean minus the standard deviation of the average voltage values ​​of all battery cells, it is marked as a low voltage range.

[0015] When the average voltage value is between twice the arithmetic mean minus the standard deviation of the average voltage values ​​of all battery cells and twice the arithmetic mean plus the standard deviation, it is marked as the medium voltage range.

[0016] A high voltage range is defined as a range where the average voltage value is higher than twice the arithmetic mean plus the standard deviation of the average voltage values ​​of all battery cells.

[0017] Preferably, the detection of the number density of battery cells within each voltage range in step S1 includes:

[0018] Within each voltage range, sensors are used to scan each battery cell individually, recording the unique identifier, initial voltage value, and location coordinates of each battery cell.

[0019] Within each voltage range, multiple fixed monitoring points are set up, and the spacing between the monitoring points is adjusted according to the arrangement density of the battery cells;

[0020] At each monitoring point, sensors are used to measure local voltage and current density, and the local voltage and current density data at each monitoring point are recorded.

[0021] Based on the local voltage and current density data of the monitoring points, the number of battery cells within the coverage area is counted at each monitoring point, and the volume within the coverage area is calculated according to the arrangement of the battery cells, thus obtaining the number density.

[0022] The quantity density data of each monitoring point is standardized and converted into relative density values;

[0023] Based on the standardized quantity density data, a voltage gradient distribution map is plotted with the voltage range as the horizontal axis and the relative density value as the vertical axis, with each monitoring point corresponding to a data point.

[0024] Preferably, step S2 includes the following steps:

[0025] Step S21: At each connection point of the multi-channel battery, use a high-precision current sensor to collect current data three times: at the start of charging, when charging reaches 50%, and at the end of charging. Each data collection lasts for 10 seconds.

[0026] Step S22: Round the current value collected each time and keep three decimal places to obtain the processed current value;

[0027] Step S23: Calculate the current distribution coefficient for each connection point based on the processed current value, where the current distribution coefficient is the ratio of the current value of each connection point to the sum of the current values ​​of all connection points;

[0028] Step S24: Calculate the actual current distribution ratio of each battery cell at different charging and discharging stages. By analyzing the current distribution coefficient of each connection point, determine the actual current distribution ratio of each battery cell at the start of charging, when charging to 50%, and at the end of charging, so as to obtain the current distribution ratio.

[0029] Step S25: Perform time series analysis on the current distribution ratio of each battery cell in different charging and discharging stages, record the changes in the current distribution ratio of each battery cell at different time points, and obtain current time series information;

[0030] Step S26: Perform volatility analysis on the current time series information, calculate the fluctuation amplitude of the current distribution ratio of each battery cell in different charging and discharging stages, and generate current distribution difference data.

[0031] Preferably, step S3 involves collecting overall temperature change data of the multi-channel battery during charging and discharging, as well as local temperature data of each battery cell, including:

[0032] The temperature data of each battery cell surface of the multi-channel battery is collected and processed at multiple points. Temperature sensors are used to collect temperature data three times at the start of charging, when charging reaches 50%, and when charging ends. Each collection lasts for 5 seconds, so as to obtain the local temperature data of each battery cell at different charging and discharging stages.

[0033] Time series analysis was performed on the collected local temperature data to record the temperature changes of each battery cell at different time points, and the temperature change curve of each battery cell was obtained.

[0034] The temperature change curve is subjected to fluctuation analysis to calculate the temperature fluctuation amplitude of each battery cell in different charging and discharging stages, and the temperature fluctuation characteristic data of each battery cell are obtained.

[0035] The temperature fluctuation characteristic data are classified to obtain the temperature stability classification results;

[0036] Based on the temperature stability classification results, the overall temperature change data of the multi-channel battery during the charging and discharging process were determined.

[0037] Preferably, step S3 involves performing thermal flow field simulation analysis on the overall temperature change data and identifying the heat propagation paths within the battery pack, including:

[0038] Identify the multi-channel battery structure distribution diagram;

[0039] Acquire the physical structure data of the multi-channel battery, which includes the arrangement, connection method and physical location of the battery cells;

[0040] Based on the physical structure data, spatial distribution analysis of the overall temperature change data is performed to generate a temperature spatial distribution map;

[0041] On the temperature distribution map, the local temperature data of each battery cell is spatially mapped to correspond the local temperature data with the physical location of the battery cell, thus generating a temperature mapping map.

[0042] Based on the multi-channel battery structure distribution map, a thermal flow field simulation analysis is performed on the temperature mapping map to simulate the heat propagation process inside the battery pack and obtain the heat propagation path.

[0043] Preferably, in step S3, determining the impact of heat propagation paths on current distribution differences by combining local temperature data of each battery cell includes:

[0044] Time series analysis was performed on the local temperature data of each battery cell to extract the temperature characteristic values ​​of each battery cell at different charging and discharging stages, including the average temperature, temperature fluctuation amplitude and temperature change rate.

[0045] The heat transfer path is segmented into multiple sub-path segments, each containing several battery cells. The starting and ending battery cell positions of each sub-path segment are recorded.

[0046] For each battery cell in a sub-path segment, calculate the cross-correlation coefficient between its local temperature data and current distribution difference data;

[0047] For each sub-path segment, the battery cells are divided into high correlation group and low correlation group according to the cross-correlation coefficient;

[0048] For battery cells in high correlation groups, the linear relationship strength between heat propagation path and current distribution difference data is detected to obtain temperature-current correlation data.

[0049] Preferably, step S4 includes the following steps:

[0050] Step S41: On each battery cell of the multi-channel battery, use a capacity tester to collect capacity data of each battery cell. Collect capacity data three times at 10%, 50% and 90% of the charge and discharge cycle, with each collection lasting 10 minutes.

[0051] Step S42: Perform time series analysis on the capacity data of each battery cell, record the capacity change of each battery cell in different charge and discharge cycle stages, and obtain the capacity change curve of each battery cell.

[0052] Step S43: Perform linear fitting on the capacity change curve of each battery cell, calculate the capacity change rate of each battery cell in different charge and discharge cycles, and obtain the capacity change trend data of each battery cell.

[0053] Step S44: Perform statistical analysis on the capacity change trend data of all battery cells, calculate the mean and standard deviation of the capacity change rate of each battery cell, and obtain the capacity balance index of the battery pack.

[0054] Step S45: Based on the battery pack's capacity balance index, divide the battery cells into high-capacity balance groups and low-capacity balance groups, and generate battery capacity balance data.

[0055] Preferably, step S5 includes the following steps:

[0056] Step S51: For each voltage interval in the voltage gradient distribution map, calculate the average voltage value and voltage fluctuation range of the battery cells in that interval to obtain voltage interval characteristic data;

[0057] Step S52: Normalize the voltage range feature data, temperature-current correlation data, and battery capacity balance data to generate a multi-channel battery feature dataset.

[0058] Step S53: Divide the multi-channel battery feature dataset into a battery feature training set and a battery feature validation set;

[0059] Step S54: Construct a preset model for evaluating charge and discharge performance based on the battery feature training set, and input the battery feature verification set into the preset model for evaluating charge and discharge performance to obtain model verification data;

[0060] Step S55: Adjust the model parameters of the preset model for charge-discharge performance evaluation based on the model validation data to obtain the charge-discharge performance evaluation model;

[0061] Step S56: Based on the charge-discharge performance evaluation model, predict the evolution trend of multi-channel battery performance and obtain a long-term battery performance evolution prediction report.

[0062] The beneficial effects of this invention are as follows: By acquiring the voltage data of each battery cell in a multi-channel battery and performing voltage gradient hierarchical processing, the battery cells can be divided into multiple voltage ranges. This allows for a more detailed division of the voltage state of the battery cells. Simultaneously, the number density of battery cells within each voltage range is detected, generating a voltage gradient distribution map. This precise voltage management method accurately reflects the voltage distribution within the battery pack, providing fundamental data support for subsequent battery performance evaluation and helping to promptly identify areas of abnormal voltage distribution, thereby effectively monitoring the battery's voltage state. Furthermore, by collecting the current distribution coefficients at each connection point of the multi-channel battery and performing current distribution difference analysis, the actual current distribution ratio of each cell at different charging and discharging stages is calculated, generating current distribution difference data. This achieves a quantitative evaluation of the current distribution, clearly indicating the actual current proportion allocated to each battery cell during charging and discharging. This analysis helps reveal the imbalance in current distribution within the battery pack, providing a basis for balanced charging and discharging strategies and avoiding problems such as overcharging or over-discharging of battery cells due to uneven current distribution. This process involves collecting overall temperature change data and local temperature data for each battery cell during the charging and discharging process of a multi-channel battery. Thermal flow field simulation analysis is performed on the overall temperature change data to identify heat propagation paths within the battery pack. Combined with the local temperature data of each battery cell, the impact of heat propagation paths on current distribution differences is determined, generating temperature-current correlation data. This process delves into the interaction between temperature and current, enabling an understanding of the causes of current distribution differences from a thermal management perspective. This correlation analysis allows for optimization of the battery pack's thermal management system, reducing the impact of temperature differences on current distribution and improving the battery pack's performance stability under different temperature conditions. Capacity data for each battery cell is collected and capacity balance analysis is performed to calculate the capacity change trend of each cell in different charge-discharge cycles, generating battery capacity balance data. This provides a clear understanding of the capacity changes of each battery cell, enabling precise control over the capacity state of each cell after multiple charge-discharge cycles. This capacity balance analysis helps in developing reasonable battery maintenance strategies, timely replacement or repair of severely degraded battery cells, ensuring overall battery pack capacity consistency, and extending the battery pack's lifespan. A charge-discharge performance evaluation model is constructed based on voltage gradient distribution maps, temperature-current correlation data, and battery capacity balance data. This model predicts the evolution trend of multi-channel battery performance, resulting in a long-term battery performance prediction report. This comprehensive evaluation model, integrating multiple data sources, fully reflects the battery's charge-discharge performance, considering factors such as voltage, current, temperature, and capacity, and can also predict long-term changes in battery performance.

[0063] This specification also provides a multi-channel battery charge-discharge test system for performing the multi-channel battery charge-discharge test method described above. The multi-channel battery charge-discharge test system includes:

[0064] The voltage gradient layering module is used to acquire the voltage data of each battery cell in the multi-channel battery, perform voltage gradient layering processing on the voltage data of each battery cell, divide the battery cell into multiple voltage intervals, detect the number density of battery cells in each voltage interval, and generate a voltage gradient distribution map.

[0065] The current distribution difference analysis module is used to collect the current distribution coefficients of each connection point of the multi-channel battery; perform current distribution difference analysis on the current distribution coefficients; calculate the actual current distribution ratio of each unit in different charging and discharging stages; and generate current distribution difference data.

[0066] The temperature-current correlation module is used to collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery; perform thermal flow field simulation analysis on the overall temperature change data and identify the heat propagation path inside the battery pack; combine the local temperature data of each battery cell to determine the impact of the heat propagation path on the current distribution difference data and generate temperature-current correlation data.

[0067] The capacity change trend module is used to collect capacity data of each battery cell; perform capacity balance analysis on the capacity data of each battery cell, calculate the capacity change trend of each cell in different charge and discharge cycles, and generate battery capacity balance data.

[0068] The charge-discharge performance evaluation module is used to construct a charge-discharge performance evaluation model based on voltage gradient distribution map, temperature-current correlation data and battery capacity balance data; based on the charge-discharge performance evaluation model, it predicts the evolution trend of multi-channel battery performance and obtains a long-term battery performance evolution prediction report.

[0069] The multi-channel battery charge-discharge testing system of this invention accurately divides the voltage range of battery cells and generates a distribution map through a voltage gradient layering module, quantifies the current distribution ratio of each cell during the charge-discharge stage through a current distribution difference analysis module, determines the impact of heat propagation path on current distribution by combining overall and local temperature data through a temperature-current correlation module, accurately analyzes the capacity change trend of each cell through a capacity change trend module, and constructs an evaluation model based on the above data to predict performance evolution. This system achieves a comprehensive and accurate evaluation and prediction of the charge-discharge performance of multi-channel batteries, providing a scientific basis for battery use, maintenance and management, and effectively improving the reliability and service life of battery systems. Attached Figure Description

[0070] Figure 1A schematic diagram of the steps in a multi-channel battery charge-discharge test method;

[0071] Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2.

[0072] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

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

[0074] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0075] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0076] To achieve the above objectives, please refer to Figures 1 to 2 A multi-channel battery charge-discharge test method, the method comprising the following steps:

[0077] Step S1: Obtain the voltage data of each battery cell in the multi-channel battery, perform voltage gradient layering processing on the voltage data of each battery cell, divide the battery cell into multiple voltage intervals; detect the number density of battery cells in each voltage interval, and generate a voltage gradient distribution map.

[0078] In this embodiment of the invention, during step S1, a high-precision voltage acquisition module is first used. This module has a multi-channel input interface and can simultaneously connect to each battery cell in the multi-channel battery pack to ensure that the voltage data of each battery cell can be acquired synchronously. The voltage acquisition module has a sampling rate of 1000 times per second and a sampling accuracy of 0.1 millivolts, thereby ensuring that the acquired voltage data has high accuracy and high resolution. The acquired voltage data is transmitted to the data processing unit in the form of a digital signal. The data processing unit uses an advanced digital signal processing algorithm to perform voltage gradient stratification processing on the voltage data of each battery cell. Specifically, the reference voltage value for voltage gradient stratification is set to 3.0 volts, and the voltage gradient division interval is 0.05 volts. That is, starting from 3.0 volts, each increment of 0.05 volts is divided into a voltage interval, up to the highest voltage value of 4.2 volts, resulting in a total of 24 voltage intervals. The data processing unit accurately classifies the voltage data of each battery cell into the corresponding voltage interval according to the above-set voltage gradient division standard. Subsequently, the data processing unit accurately counts the number of battery cells within each voltage range, calculating the battery cell number density for each range. The formula for calculating the number density is the number of battery cells in the voltage range divided by the voltage range occupied by that range (in volts). The calculated battery cell number density data for each voltage range is then used to generate a voltage gradient distribution map in the form of a two-dimensional chart. The horizontal axis represents the voltage range, and the vertical axis represents the corresponding battery cell number density. The chart is created using professional data visualization software to ensure that the generated voltage gradient distribution map clearly, intuitively, and accurately reflects the voltage gradient distribution of each battery cell in the multi-channel battery.

[0079] Step S2: Collect the current distribution coefficients of each connection point of the multi-channel battery; perform current distribution difference analysis on the current distribution coefficients, calculate the actual current distribution ratio of each unit in different charging and discharging stages, and generate current distribution difference data;

[0080] In this embodiment of the invention, during step S2, a high-precision current sensor is used to collect the current distribution coefficients at each connection point of the multi-channel battery. The current sensor is installed at each connection point of the battery pack, with a measurement accuracy of 0.01 amperes and a sampling frequency of 500 times per second. The current sensor converts the collected current signal into a digital signal and transmits it to the data processing system. The data processing system first preprocesses the collected current data, including filtering and noise reduction operations, to ensure data accuracy. The preprocessed current data is stored in a data buffer. Subsequently, the data processing system performs current distribution difference analysis on the current distribution coefficients. The analysis process includes segmenting the collected current data according to different charge and discharge stages. The division of charge and discharge stages is based on the battery voltage change curve; the charging stage occurs when the battery voltage rises from 3.0 volts to 4.2 volts, and the discharging stage occurs when the voltage drops from 4.2 volts to 3.0 volts. Within each charge and discharge stage, the data processing system calculates the average value of the current distribution coefficients at each connection point and uses this as the reference current distribution coefficient for that stage. Next, the data processing system calculates the actual current allocation ratio of each battery cell at different charge / discharge stages. Specifically, the calculation method involves dividing the current value of each battery cell at the current charge / discharge stage by the reference current allocation coefficient for that stage to obtain the actual current allocation ratio of that battery cell. The calculated actual current allocation ratio data is stored in the data storage unit and recorded in tabular form, containing the battery cell number, charge / discharge stage, and corresponding current allocation ratio data. Finally, the data processing system generates current allocation difference data based on the calculated actual current allocation ratio data. The current allocation difference data is presented in graphical and tabular form. In the graph, the horizontal axis represents the battery cell number, and the vertical axis represents the current allocation ratio. Data from different charge / discharge stages are distinguished by different colors or lines. The table details the current allocation ratio value of each battery cell at each charge / discharge stage. The generated current allocation difference data is saved in the data storage unit for subsequent analysis and use.

[0081] Step S3: Collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery; perform thermal flow field simulation analysis on the overall temperature change data and identify the heat propagation path inside the battery pack; combine the local temperature data of each battery cell to determine the impact of the heat propagation path on the current distribution difference data and generate temperature-current correlation data.

[0082] In this embodiment of the invention, during step S3, high-precision temperature sensors are first used to collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery. The overall temperature sensor is installed on the surface of the battery pack casing, with a measurement accuracy of 0.1 degrees Celsius and a sampling frequency of 10 times per second. Local temperature sensors are installed on the surface of each battery cell, with a measurement accuracy of 0.01 degrees Celsius and a sampling frequency of 20 times per second. The overall and local temperature sensors convert the collected temperature data into digital signals and transmit them to the data processing system. The data processing system preprocesses the collected temperature data, including data calibration and filtering operations, to ensure the accuracy and stability of the data. Subsequently, the data processing system performs thermal flow field simulation analysis on the overall temperature change data. The thermal flow field simulation analysis uses the finite element analysis method, dividing the battery pack into multiple tiny cells, each cell measuring 1 mm × 1 mm × 1 mm. During the simulation, the initial temperature of the battery pack is set to 25 degrees Celsius, the boundary condition is natural convection heat dissipation, and the convection heat transfer coefficient is 10 W / (m²·°C). Finite element analysis software was used to calculate the temperature distribution and heat propagation paths within the battery pack. Heat propagation path identification was based on changes in the temperature gradient; a significant heat propagation path was considered to exist when the temperature gradient exceeded 0.5 degrees Celsius / cm. Next, the data processing system combined local temperature data from each battery cell to determine the impact of the heat propagation path on current distribution differences. Specifically, local temperature data was compared with heat propagation paths to identify battery cells affected by heat propagation. For battery cells affected by heat propagation, the correlation between their local temperature and overall temperature changes was calculated. When the correlation coefficient was greater than 0.8, the local temperature change of that battery cell was considered highly correlated with the overall temperature change, and heat propagation had a significant impact on the current distribution of that battery cell. Finally, the data processing system generated temperature-current correlation data based on the above analysis results. This temperature-current correlation data was recorded in tabular form, containing the battery cell number, local temperature data, overall temperature change data, heat propagation path identifiers, and current distribution difference data. The generated temperature-current correlation data was stored in a data storage unit for subsequent analysis and use.

[0083] Step S4: Collect capacity data of each battery cell; perform capacity balance analysis on the capacity data of each battery cell, calculate the capacity change trend of each cell in different charge and discharge cycles, and generate battery capacity balance data.

[0084] In this embodiment of the invention, during step S4, a high-precision capacity detection device is first used to collect capacity data of each battery cell. The capacity detection device measures capacity using a constant current charge-discharge method. Specifically, a constant current is applied to each battery cell for charging until the battery voltage reaches a preset charging cutoff voltage of 4.2 volts. Then, the same constant current is used for discharging until the battery voltage drops to the discharge cutoff voltage of 3.0 volts. During the charge-discharge process, the capacity detection device records the charge and discharge capacity of the battery cell in real time, with a measurement accuracy of 0.1 mA / h and a sampling frequency of once per second. The collected capacity data is transmitted to the data processing system in digital signal form. After receiving the capacity data, the data processing system performs capacity balance analysis on the capacity data of each battery cell. The analysis process includes grouping the collected capacity data according to different charge-discharge cycles. Each charge-discharge cycle is defined as a complete process from the start of charging to the end of discharging. The data processing system calculates the capacity change of each battery cell in each charge-discharge cycle. The formula for calculating the capacity change is the discharge capacity of the current cycle minus the discharge capacity of the previous cycle. Subsequently, the data processing system performs trend analysis on the capacity change of each battery cell, using linear regression to fit the capacity change trend line and obtain the capacity change trend of each battery cell in different charge-discharge cycles. Finally, based on the above analysis results, the data processing system generates battery capacity balance data. The battery capacity balance data is recorded in tabular form, containing data such as battery cell number, number of charge-discharge cycles, discharge capacity of each cycle, and capacity change trend. The generated battery capacity balance data is stored in the data storage unit for subsequent analysis and use.

[0085] Step S5: Based on the voltage gradient distribution map, temperature-current correlation data, and battery capacity balance data, construct a charge-discharge performance evaluation model; based on the charge-discharge performance evaluation model, predict the evolution trend of multi-channel battery performance and obtain a long-term battery performance evolution prediction report.

[0086] In this embodiment of the invention, during step S5, the voltage gradient distribution map, temperature-current correlation data, and battery capacity balance data generated in steps S1 to S4 are first imported into the data integration module. The data integration module standardizes these data to ensure consistent data format and dimensions across different data sources. Standardization includes normalizing the voltage range and battery cell density in the voltage gradient distribution map, dimensionless processing of local and overall temperature changes and current distribution differences in the temperature-current correlation data, and standardization of the charge-discharge cycle count and capacity change trend in the battery capacity balance data. Subsequently, the data integration module transmits the processed data to the performance evaluation model construction module. The performance evaluation model construction module uses multivariate regression analysis, with the voltage range and density in the voltage gradient distribution map as independent variables, the local and overall temperature changes and current distribution differences in the temperature-current correlation data as intermediate variables, and the charge-discharge cycle count and capacity change trend in the battery capacity balance data as dependent variables, to construct a charge-discharge performance evaluation model. During model construction, a 95% confidence level was set for the regression analysis, and a stepwise regression method was used to screen out variables that significantly affect battery performance, thereby improving the model's accuracy and reliability. Next, the performance evaluation model construction module uses the constructed charge-discharge performance evaluation model to predict the performance evolution trend of the multi-channel battery. The prediction process includes inputting the current voltage gradient distribution, temperature-current correlation, and battery capacity balance data. Based on this input data, the model calculates battery performance indicators at different future time points, such as battery capacity, voltage stability, and current distribution uniformity. The prediction time span is set to the next 12 months, with a time step of once per month. Finally, the performance evaluation model construction module compiles the prediction results into a long-term battery performance evolution prediction report. The prediction report is presented in the form of charts and tables. The charts include curves showing the changes in battery capacity, voltage stability, and current distribution uniformity over time. The tables list the battery performance indicator values ​​at each time point in detail. The prediction report is stored in the data storage unit.

[0087] Preferably, in step S1, obtaining the voltage data of each battery cell in the multi-channel battery and performing voltage gradient hierarchical processing on the voltage data of each battery cell to divide the battery cell into multiple voltage ranges includes:

[0088] The voltage of each battery cell was measured three times using a voltmeter. The first measurement was taken when the battery cell was charged to 20%, the second measurement was taken when it was charged to 50%, and the third measurement was taken when it was charged to 80%.

[0089] The voltage data collected three times were averaged and rounded to two decimal places to obtain the average voltage value of each battery cell.

[0090] The battery cells are divided into three voltage ranges based on the average voltage value. When the average voltage value is lower than twice the arithmetic mean minus the standard deviation of the average voltage values ​​of all battery cells, it is marked as a low voltage range.

[0091] When the average voltage value is between twice the arithmetic mean minus the standard deviation of the average voltage values ​​of all battery cells and twice the arithmetic mean plus the standard deviation, it is marked as the medium voltage range.

[0092] A high voltage range is defined as a range where the average voltage value is higher than twice the arithmetic mean plus the standard deviation of the average voltage values ​​of all battery cells.

[0093] In this embodiment of the invention, a high-precision digital voltmeter is first used to collect voltage data for each battery cell. The voltmeter has a measurement accuracy of 0.01 volts and a range of 0 to 5 volts. During the charging process, the voltmeter collects voltage data when the battery cell's state of charge reaches 20%, 50%, and 80%, respectively. Monitoring of the charging state is achieved through a power monitoring chip connected to the battery cell, which can accurately and in real-time report the battery cell's charging percentage. The voltage data collected each time is recorded in a data acquisition system with high-capacity storage, capable of storing a large amount of voltage data. After three voltage acquisitions, the data processing unit averages the three voltage data for each battery cell. Specifically, the three voltage values ​​for each battery cell are added together and then divided by 3 to obtain the average voltage value for that battery cell. During the calculation, the data processing unit retains the result to two decimal places to ensure data accuracy. For example, if the voltage values ​​of a battery cell are 3.51 volts, 3.53 volts, and 3.52 volts in three separate measurements, its average voltage value is calculated as (3.51 + 3.53 + 3.52) / 3 = 3.52 volts. Subsequently, the data processing unit calculates the arithmetic mean and standard deviation of the average voltage values ​​of all battery cells. The arithmetic mean is calculated by summing the average voltage values ​​of all battery cells and dividing by the total number of battery cells. The standard deviation is calculated using the statistical standard deviation formula: first, the square of the difference between the average voltage value of each battery cell and the arithmetic mean is calculated; then, these squared values ​​are summed and divided by the total number of battery cells minus 1; finally, the square root of the result is taken. For example, assuming there are 10 battery cells with average voltage values ​​of 3.52 volts, 3.55 volts, 3.50 volts, etc., the data processing unit first calculates the arithmetic mean of these average voltage values, assuming the arithmetic mean is 3.53 volts, and then calculates the standard deviation, assuming the standard deviation is 0.02 volts. Finally, the data processing unit divides each battery cell into voltage ranges based on the calculated arithmetic mean and standard deviation. A battery cell is marked as being in the low-voltage range when its average voltage is less than twice the arithmetic mean minus the standard deviation; it is marked as being in the medium-voltage range when the average voltage is between twice the arithmetic mean minus the standard deviation and twice the arithmetic mean plus the standard deviation; and it is marked as being in the high-voltage range when the average voltage is greater than twice the arithmetic mean plus the standard deviation. For example, if the arithmetic mean is 3.53 volts and the standard deviation is 0.02 volts, then the upper limit of the low-voltage range is 3.53 - 2 × 0.02 = 3.49 volts, and the lower limit of the high-voltage range is 3.53 + 2 × 0.02 = 3.57 volts.

[0094] Preferably, the detection of the number density of battery cells within each voltage range in step S1 includes:

[0095] Within each voltage range, sensors are used to scan each battery cell individually, recording the unique identifier, initial voltage value, and location coordinates of each battery cell.

[0096] Within each voltage range, multiple fixed monitoring points are set up, and the spacing between the monitoring points is adjusted according to the arrangement density of the battery cells;

[0097] At each monitoring point, sensors are used to measure local voltage and current density, and the local voltage and current density data at each monitoring point are recorded.

[0098] Based on the local voltage and current density data of the monitoring points, the number of battery cells within the coverage area is counted at each monitoring point, and the volume within the coverage area is calculated according to the arrangement of the battery cells, thus obtaining the number density.

[0099] The quantity density data of each monitoring point is standardized and converted into relative density values;

[0100] Based on the standardized quantity density data, a voltage gradient distribution map is plotted with the voltage range as the horizontal axis and the relative density value as the vertical axis, with each monitoring point corresponding to a data point.

[0101] In this embodiment of the invention, within each defined voltage range, a high-precision sensor scans each battery cell individually. This sensor possesses high-resolution identification capabilities, accurately recording the unique identifier, initial voltage value, and location coordinates of each battery cell. The sensor's scanning accuracy is 0.01 mm, ensuring precise recording of location coordinates. The initial voltage value measurement accuracy is 0.01 volts, ensuring the accuracy of the voltage data. During the scanning process, the sensor reads the unique identifier of each battery cell via near-field communication technology (such as NFC), records the initial voltage value through a built-in voltage measurement module, and simultaneously records the location coordinates using a high-precision positioning module. Within each voltage range, multiple fixed monitoring points are set according to the battery cell density. The spacing between the monitoring points is calculated using a pre-set algorithm based on the battery cell density to ensure that the monitoring points uniformly cover the entire voltage range. Specifically, when the battery cell density exceeds 10 cells per square centimeter, the monitoring point spacing is set to 5 cm; when the density is less than 10 cells per square centimeter, the monitoring point spacing is adjusted to 10 cm. The location of monitoring points is determined using a precise positioning system, ensuring that each point accurately covers the predetermined area. At each monitoring point, sensors measure local voltage and current density. The measurement accuracy of local voltage is 0.01 volts, and the measurement accuracy of current density is 0.01 amperes / square centimeter. The sensors, through high-precision voltage and current measurement modules, record the local voltage and current density data of each monitoring point in real time and transmit this data to the data processing system. After receiving the local voltage and current density data from the monitoring points, the data processing system counts the number of battery cells within the coverage area of ​​each monitoring point. The coverage area is calculated based on the arrangement of the battery cells, assuming a matrix arrangement. The coverage area of ​​each monitoring point is a square region with a side length equal to half the distance between monitoring points. For example, when the distance between monitoring points is 5 centimeters, the coverage area is a square region with a side length of 2.5 centimeters. The data processing system calculates the volume within the coverage area based on the battery cell arrangement density and the size of the coverage area. The volume is calculated by multiplying the area of ​​the coverage area by the height of the battery cell. Assuming the battery cell height is 1 cm, the volume within the coverage area is the area of ​​the coverage area multiplied by 1 cm. Subsequently, the data processing system calculates the number density based on the number and volume of the battery cells within the coverage area. The number density is calculated by dividing the number of battery cells by the volume within the coverage area. The calculated number density data is then standardized to convert it into a relative density value. The standardization method involves dividing the number density of each monitoring point by the average number density of all monitoring points to obtain the relative density value. Finally, the data processing system plots a voltage gradient distribution map based on the standardized number density data.During the plotting process, the voltage range was used as the horizontal axis and the relative density value as the vertical axis, with each monitoring point corresponding to a data point. Professional data visualization software was used to create the charts, ensuring that the generated voltage gradient distribution map clearly, intuitively, and accurately reflects the number density distribution of battery cells within each voltage range of the multi-channel battery.

[0102] As an example of the present invention, reference is made to... Figure 2 As shown, in this example, step S2 includes:

[0103] Step S21: At each connection point of the multi-channel battery, use a high-precision current sensor to collect current data three times: at the start of charging, when charging reaches 50%, and at the end of charging. Each data collection lasts for 10 seconds.

[0104] Step S22: Round the current value collected each time and keep three decimal places to obtain the processed current value;

[0105] Step S23: Calculate the current distribution coefficient for each connection point based on the processed current value, where the current distribution coefficient is the ratio of the current value of each connection point to the sum of the current values ​​of all connection points;

[0106] Step S24: Calculate the actual current distribution ratio of each battery cell at different charging and discharging stages. By analyzing the current distribution coefficient of each connection point, determine the actual current distribution ratio of each battery cell at the start of charging, when charging to 50%, and at the end of charging, so as to obtain the current distribution ratio.

[0107] Step S25: Perform time series analysis on the current distribution ratio of each battery cell in different charging and discharging stages, record the changes in the current distribution ratio of each battery cell at different time points, and obtain current time series information;

[0108] Step S26: Perform volatility analysis on the current time series information, calculate the fluctuation amplitude of the current distribution ratio of each battery cell in different charging and discharging stages, and generate current distribution difference data.

[0109] In this embodiment of the invention, high-precision current sensors are installed at each connection point of the multi-channel battery. These current sensors have a measurement accuracy of 0.001 amperes and a sampling frequency of 10 times per second. At the start of charging, when charging reaches 50%, and at the end of charging, each current sensor performs a current acquisition, with each acquisition lasting 10 seconds. The acquired current data is transmitted to the data processing system in the form of a digital signal and recorded in the data acquisition log, which includes the acquisition timestamp, connection point number, and corresponding current value. In step S22, the data processing system rounds each acquired current value to three decimal places. Specifically, the system reads the current values ​​acquired at each connection point during the three different charging stages and rounds each current value to ensure that the processed current value is accurate to three decimal places. For example, if the current value acquired at a connection point at the start of charging is 2.3456 amperes, it will be rounded to 2.346 amperes. In step S23, the data processing system calculates the current allocation coefficient for each connection point based on the processed current value. The formula for calculating the current allocation coefficient is the current value of each connection point divided by the sum of the current values ​​of all connection points. The system first calculates the sum of the current values ​​of all connection points in each charging stage, and then calculates the current allocation coefficient for each connection point separately. For example, if at the start of charging, the sum of the current values ​​of all connection points is 100.000 amperes, and the current value of a certain connection point is 2.346 amperes, then the current allocation coefficient for that connection point is 2.346 / 100.000 = 0.02346. In step S24, the data processing system calculates the actual current allocation ratio of each battery cell in different charging and discharging stages. By analyzing the current allocation coefficient of each connection point, the actual current allocation ratio of each battery cell at the start of charging, when charging to 50%, and at the end of charging is determined. Specifically, the system allocates the current allocation coefficient of each connection point to the corresponding battery cell according to the correspondence between the battery cell and the connection point, thereby obtaining the actual current allocation ratio of each battery cell in different charging stages. In step S25, the data processing system performs time-series analysis on the current allocation ratio of each battery cell at different charge and discharge stages. The system records the changes in the current allocation ratio of each battery cell at different time points, generating current time-series information. The time-series analysis uses a sliding window method with a window size of 5 data points and a step size of 1 data point to smooth the current allocation ratio change curve and ensure data stability. Finally, in step S26, the data processing system performs volatility analysis on the current time-series information. The volatility amplitude of the current allocation ratio of each battery cell at different charge and discharge stages is calculated using the formula: the difference between the maximum and minimum current allocation ratios.The system organizes the calculated fluctuation amplitude data into current distribution difference data and records it in tabular form. The table includes the battery cell number, charging / discharging stage, and corresponding fluctuation amplitude value. The generated current distribution difference data is stored in the data storage unit for subsequent analysis and use.

[0110] Of particular importance, step S26 includes the following steps:

[0111] Step S261: Perform segmented weighting on the current time series information to obtain weighted current time series information;

[0112] Step S262: Perform dynamic threshold filtering on the weighted current time series information to obtain the filtered current time series information;

[0113] Step S263: Perform multi-dimensional feature extraction processing on the filtered current time series information to obtain multi-dimensional feature data of current distribution ratio;

[0114] Step S264: Perform correlation analysis on the multi-dimensional feature data to obtain the correlation analysis results;

[0115] Step S265: Perform a stability assessment on the correlation analysis results to obtain the stability assessment results;

[0116] Step S266: Perform anomaly detection processing on the stability assessment results to obtain anomaly detection results;

[0117] Step S267: Adaptively adjust the anomaly detection results to obtain current distribution difference data.

[0118] In this embodiment of the invention, the current time series information is first processed by segmented weighting. Specifically, the current time series information is divided into multiple equal-length time periods according to the time sequence, with each time period having a length of 10 data points. For the current time series data within each time period, a linear weighting method is used for weighting. The weighting coefficient increases sequentially according to the time sequence, with an initial weighting coefficient of 0.1. For each additional data point, the weighting coefficient increases by 0.1 until the weighting coefficient of the last data point reaches 1.0. The weighted current time series information is calculated by weighted summation, i.e., the weighted value of each data point is equal to the current value of that data point multiplied by its corresponding weighting coefficient. Then, the weighted values ​​of all data points within the same time period are summed to obtain the weighted current value for that time period. In step S262, the weighted current time series information is subjected to dynamic threshold filtering. The dynamic threshold is calculated based on the moving average and moving standard deviation of the weighted current values. The calculation window length for the moving average is 5 data points, and the calculation window length for the moving standard deviation is also 5 data points. The dynamic threshold is set to the moving average plus or minus twice the moving standard deviation. The filtering process involves marking the weighted current value for each time period as outlier and removing it if it exceeds the dynamic threshold, thus obtaining the filtered current time series information. Step S263 involves multi-dimensional feature extraction of the filtered current time series information. Feature extraction includes calculating the mean, variance, maximum value, minimum value, kurtosis (the ratio of the maximum value to the mean), and kurtosis (the ratio of the fourth central moment to the square of the variance) of the current time series for each time period. These feature values ​​are calculated using corresponding mathematical formulas. For example, the mean is the arithmetic mean of all current values ​​within the time period, the variance is the average of the squares of the differences between the current value and the mean, the maximum and minimum values ​​are the maximum and minimum current values ​​within the time period, and the kurtosis and kurtosis are calculated according to the above definitions. The obtained multi-dimensional feature data is recorded in tabular form, containing the time period number and the corresponding feature values. In step S264, correlation analysis is performed on the multi-dimensional feature data. The correlation analysis employs the Pearson correlation coefficient method to calculate the correlation coefficient between each eigenvalue and all other eigenvalues. The correlation coefficient is calculated by dividing the covariance of the two eigenvalue sequences by the product of their standard deviations. The correlation coefficient matrix is ​​then used to analyze the correlation between the eigenvalues. If the correlation coefficient between two eigenvalues ​​is greater than 0.8, a strong correlation is considered to exist between them. The correlation analysis results are recorded in matrix form, with each element corresponding to the correlation coefficient between two eigenvalues. Step S265 proceeds to perform a stability assessment on the correlation analysis results. The stability assessment is based on the stability of the correlation coefficients between each eigenvalue in the correlation analysis results. Specifically, the standard deviation of the correlation coefficients between each eigenvalue and all other eigenvalues ​​is calculated.If the standard deviation of the correlation coefficient of a certain feature value is less than 0.1, the feature value is considered to have high stability; conversely, if the standard deviation of the correlation coefficient is greater than 0.1, the feature value is considered to have low stability. The stability assessment results are recorded in tabular form, including the feature value number and the corresponding stability assessment index (standard deviation of the correlation coefficient). In step S266, anomaly detection processing is performed on the stability assessment results. Anomaly detection uses a Z-score-based method. First, the Z-score of each feature value is calculated. The formula for calculating the Z-score is the feature value minus its mean and then divided by its standard deviation. If the absolute value of the Z-score of a feature value is greater than 3, the feature value is considered an anomaly. The anomaly detection results are recorded in the stability assessment result table in the form of labels. Abnormal feature values ​​are labeled as "abnormal," and normal feature values ​​are labeled as "normal." Finally, in step S267, adaptive adjustment processing is performed on the anomaly detection results. The adaptive adjustment processing corrects the current distribution difference data according to the anomaly detection results. For feature values ​​marked as abnormal, the mean of the same feature values ​​in adjacent time periods is used for replacement. If the same characteristic value also shows anomalies in adjacent time periods, the average value of that characteristic value across all time periods is used for replacement. After adaptive adjustment, the final current distribution difference data is obtained, which is recorded in tabular form. The table contains the battery cell number, time period number, and corresponding current distribution difference value.

[0119] Preferably, step S3 involves collecting overall temperature change data of the multi-channel battery during charging and discharging, as well as local temperature data of each battery cell, including:

[0120] The temperature data of each battery cell surface of the multi-channel battery is collected and processed at multiple points. Temperature sensors are used to collect temperature data three times at the start of charging, when charging reaches 50%, and when charging ends. Each collection lasts for 5 seconds, so as to obtain the local temperature data of each battery cell at different charging and discharging stages.

[0121] Time series analysis was performed on the collected local temperature data to record the temperature changes of each battery cell at different time points, and the temperature change curve of each battery cell was obtained.

[0122] The temperature change curve is subjected to fluctuation analysis to calculate the temperature fluctuation amplitude of each battery cell in different charging and discharging stages, and the temperature fluctuation characteristic data of each battery cell are obtained.

[0123] The temperature fluctuation characteristic data are classified to obtain the temperature stability classification results;

[0124] Based on the temperature stability classification results, the overall temperature change data of the multi-channel battery during the charging and discharging process were determined.

[0125] In this embodiment of the invention, a high-precision temperature sensor is used to collect and process the temperature data of each battery cell surface in a multi-channel battery at multiple points. The temperature sensor has a measurement accuracy of 0.01 degrees Celsius and a sampling frequency of twice per second. Temperature data is collected three times for each battery cell at the start of charging, when charging reaches 50%, and at the end of charging, with each collection lasting 5 seconds. The collected temperature data is transmitted to the data processing system in the form of a digital signal and recorded in the data acquisition log, which includes the acquisition timestamp, battery cell number, and corresponding temperature value. Next, the data processing system performs time-series analysis on the collected local temperature data. Specifically, the system arranges the temperature data of each battery cell at different charging stages in chronological order to form a time series. The time-series analysis uses a sliding window method with a window size of 3 data points and a step size of 1 data point to smooth the temperature change curve and reduce noise interference. This method records the temperature changes of each battery cell at different time points, obtaining the temperature change curve for each battery cell. Subsequently, fluctuation analysis is performed on the temperature change curve. Fluctuation analysis is achieved by calculating the temperature fluctuation amplitude of each battery cell at different charging and discharging stages. The temperature fluctuation amplitude is calculated as the difference between the maximum and minimum temperature values ​​during each charging stage. For example, if a battery cell's temperature change curve at 50% charge shows a maximum temperature of 35.2 degrees Celsius and a minimum of 34.8 degrees Celsius, then the temperature fluctuation amplitude for that stage is 0.4 degrees Celsius. This method yields the temperature fluctuation characteristic data for each battery cell. Next, the temperature fluctuation characteristic data is classified. Cluster analysis is used to categorize the battery cell temperature fluctuation characteristics into three types: low fluctuation, medium fluctuation, and high fluctuation. Specifically, battery cells with temperature fluctuation amplitudes less than 0.2 degrees Celsius are classified as low fluctuation; those between 0.2 and 0.5 degrees Celsius are classified as medium fluctuation; and those greater than 0.5 degrees Celsius are classified as high fluctuation. The classification results are recorded in a table, which includes the battery cell number and its corresponding temperature fluctuation category. Finally, based on the temperature stability classification results, the overall temperature change data of the multi-channel battery during charging and discharging is determined. The method for determining the overall temperature variation data is as follows: calculate the percentage of battery cells in each category, and combine this with the temperature fluctuation characteristics of each category to comprehensively evaluate the overall temperature stability of the multi-channel battery. For example, if the percentage of low-fluctuation battery cells is 60%, medium-fluctuation battery cells are 30%, and high-fluctuation battery cells are 10%, then the overall temperature variation data indicates that the battery pack has good temperature stability during charging and discharging, but some battery cells still exhibit high temperature fluctuations. This data is compiled into a report and presented in the form of charts and tables to provide a basis for subsequent battery performance evaluation and optimization.

[0126] Preferably, step S3 involves performing thermal flow field simulation analysis on the overall temperature change data and identifying the heat propagation paths within the battery pack, including:

[0127] Identify the multi-channel battery structure distribution diagram;

[0128] Acquire the physical structure data of the multi-channel battery, which includes the arrangement, connection method and physical location of the battery cells;

[0129] Based on the physical structure data, spatial distribution analysis of the overall temperature change data is performed to generate a temperature spatial distribution map;

[0130] On the temperature distribution map, the local temperature data of each battery cell is spatially mapped to correspond the local temperature data with the physical location of the battery cell, thus generating a temperature mapping map.

[0131] Based on the multi-channel battery structure distribution map, a thermal flow field simulation analysis is performed on the temperature mapping map to simulate the heat propagation process inside the battery pack and obtain the heat propagation path.

[0132] In this embodiment of the invention, a high-precision 3D scanner is used to perform structural scanning on a multi-channel battery to identify its structural distribution. The 3D scanner has a scanning accuracy of 0.01 millimeters, which can accurately capture the arrangement, connection method, and physical location of the battery cells. The scanned data is stored in point cloud form, with each point containing the 3D coordinate information of the battery surface. Subsequently, the physical structure data of the multi-channel battery is extracted from the scanned data. The physical structure data includes the arrangement of the battery cells (such as matrix arrangement or staggered arrangement), the connection method (such as series or parallel connection), and the physical location of the battery cells (represented by 3D coordinates). This data is extracted and organized using professional image processing software to generate a detailed physical structure data file in XML format, containing the battery cell number, arrangement method, connection method, and the 3D coordinates of each battery cell. Next, spatial distribution analysis is performed on the overall temperature change data based on the physical structure data. The overall temperature change data is obtained from the temperature sensors collected in the previous steps, including the temperature data of each battery cell at different charging and discharging stages. The spatial distribution analysis uses Geographic Information System (GIS) technology to combine the temperature data with the physical location of the battery cells to generate a temperature spatial distribution map. In the GIS system, the temperature data of each battery cell is mapped to its corresponding physical location, forming a three-dimensional temperature distribution map. Temperature is represented by different colors, with lighter and darker colors corresponding to higher temperatures. Spatial mapping processing is performed on the local temperature data of each battery cell on the temperature distribution map. This spatial mapping is achieved through the interpolation function of the GIS software, corresponding the local temperature data to the physical location of the battery cell to generate the temperature map. Kriging interpolation is used, which can predict the temperature value of unknown points based on the temperature data of known points, thus generating a continuous temperature distribution map. The temperature map is presented in both two-dimensional and three-dimensional forms: the two-dimensional map shows the planar temperature distribution of the battery pack, while the three-dimensional map shows the three-dimensional temperature distribution. Finally, a thermal flow field simulation analysis is performed on the temperature map based on the multi-channel battery structure distribution map. The thermal flow field simulation analysis is conducted using finite element analysis (FEA) software to simulate the heat propagation process inside the battery pack. During the simulation, the initial temperature distribution of the battery pack is set to the temperature data in the temperature map, the boundary condition is natural convection heat dissipation, and the convection heat transfer coefficient is 10 watts / (square meter·degree Celsius). The temperature distribution and heat propagation paths inside the battery pack were calculated using finite element analysis software. Heat propagation paths were identified based on changes in the temperature gradient; a significant heat propagation path was considered to exist when the temperature gradient exceeded 0.5 degrees Celsius / cm. The simulation results are presented as an animation, demonstrating the dynamic heat propagation process inside the battery pack, and a heat propagation path diagram is generated, with arrows indicating the direction and intensity of heat propagation.

[0133] Preferably, in step S3, determining the impact of heat propagation paths on current distribution differences by combining local temperature data of each battery cell includes:

[0134] Time series analysis was performed on the local temperature data of each battery cell to extract the temperature characteristic values ​​of each battery cell at different charging and discharging stages, including the average temperature, temperature fluctuation amplitude and temperature change rate.

[0135] The heat transfer path is segmented into multiple sub-path segments, each containing several battery cells. The starting and ending battery cell positions of each sub-path segment are recorded.

[0136] For each battery cell in a sub-path segment, calculate the cross-correlation coefficient between its local temperature data and current distribution difference data;

[0137] For each sub-path segment, the battery cells are divided into high correlation group and low correlation group according to the cross-correlation coefficient;

[0138] For battery cells in high correlation groups, the linear relationship strength between heat propagation path and current distribution difference data is detected to obtain temperature-current correlation data.

[0139] In this embodiment of the invention, time series analysis is performed on the local temperature data of each battery cell. Specifically, time series analysis software is used to process the temperature data of each battery cell at different charge and discharge stages. The time series analysis includes calculating the average temperature, temperature fluctuation amplitude, and temperature change rate of each battery cell. The average temperature is the arithmetic mean of the temperature data within each charge and discharge stage; the temperature fluctuation amplitude is the difference between the maximum and minimum temperature values ​​within that stage; and the temperature change rate is the ratio of the temperature change to time within that stage. These characteristic values ​​are calculated using built-in statistical functions in the software and stored in a characteristic value data table, which includes the battery cell number, the charge and discharge stage, and the corresponding temperature characteristic value. Next, the heat propagation path is segmented. Based on the length of the heat propagation path and the distribution density of the battery cells, the path is divided into multiple sub-path segments. Each sub-path segment contains several battery cells, the specific number of which is determined according to the arrangement density of the battery cells. For example, if the arrangement density of battery cells is 10 battery cells per square centimeter, then each sub-path segment contains 5 battery cells. The starting and ending battery cell positions of each sub-path segment are recorded, with position information represented by the battery cell number and three-dimensional coordinates, and stored in the path segmentation data table. Subsequently, for each battery cell in each sub-path segment, the cross-correlation coefficient between its local temperature data and current distribution difference data is calculated. The cross-correlation coefficient is calculated using statistical analysis software by calculating the correlation between two time series. Specifically, the local temperature data series of each battery cell is compared and analyzed with the corresponding current distribution difference data series to calculate their cross-correlation coefficient. The formula for calculating the cross-correlation coefficient is the product of the covariance of the two series and their standard deviations. The calculated cross-correlation coefficients are stored in the correlation data table, which includes the battery cell number, sub-path segment number, and corresponding cross-correlation coefficient. Next, for each sub-path segment, based on the cross-correlation coefficient, the battery cells are divided into high-correlation and low-correlation groups. A threshold of 0.5 is set for the cross-correlation coefficient; if the cross-correlation coefficient of a battery cell is greater than or equal to 0.5, the battery cell is assigned to the high-correlation group; if the cross-correlation coefficient is less than 0.5, it is assigned to the low-correlation group. The grouping results are stored in the correlation grouping data table, which includes the sub-path segment number, battery cell number, and corresponding correlation group. Finally, for battery cells in highly correlated groups, the linear relationship strength between heat transfer path and current distribution difference data is detected. The linear relationship strength is detected using linear regression analysis. Specifically, for each highly correlated battery cell, linear regression analysis is performed with local temperature data as the independent variable and current distribution difference data as the dependent variable. The calculated linear regression coefficient and coefficient of determination (R²) are used as indicators of the linear relationship strength. The results of the linear relationship strength detection are stored in the temperature-current correlation data table, which includes the battery cell number, sub-path segment number, linear regression coefficient, and coefficient of determination.These data provide an important basis for subsequent battery performance evaluation and optimization.

[0140] Preferably, step S4 includes the following steps:

[0141] Step S41: On each battery cell of the multi-channel battery, use a capacity tester to collect capacity data of each battery cell. Collect capacity data three times at 10%, 50% and 90% of the charge and discharge cycle, with each collection lasting 10 minutes.

[0142] Step S42: Perform time series analysis on the capacity data of each battery cell, record the capacity change of each battery cell in different charge and discharge cycle stages, and obtain the capacity change curve of each battery cell.

[0143] Step S43: Perform linear fitting on the capacity change curve of each battery cell, calculate the capacity change rate of each battery cell in different charge and discharge cycles, and obtain the capacity change trend data of each battery cell.

[0144] Step S44: Perform statistical analysis on the capacity change trend data of all battery cells, calculate the mean and standard deviation of the capacity change rate of each battery cell, and obtain the capacity balance index of the battery pack.

[0145] Step S45: Based on the battery pack's capacity balance index, divide the battery cells into high-capacity balance groups and low-capacity balance groups, and generate battery capacity balance data.

[0146] In this embodiment of the invention, in step S41, a high-precision capacity tester is used to collect capacity data for each battery cell of the multi-channel battery. The capacity tester has a measurement accuracy of 0.01 ampere-hours, enabling precise measurement of the charge and discharge capacity of the battery cell. Capacity data is collected three times for each battery cell at 10%, 50%, and 90% charge points during the charge and discharge cycle, with each collection lasting 10 minutes. The collected capacity data is transmitted to the data processing system in the form of a digital signal and recorded in the capacity data log, which includes the collection timestamp, battery cell number, and corresponding capacity value. In step S42, time series analysis is performed on the capacity data of each battery cell. The time series analysis uses a sliding window method with a window size of 5 data points and a step size of 1 data point to smooth the capacity change curve and reduce noise interference. This method records the capacity change of each battery cell at different charge and discharge cycle stages, obtaining the capacity change curve for each battery cell. The capacity change curve is presented in graphical form, with the horizontal axis representing the number of charge and discharge cycles and the vertical axis representing the capacity value. In step S43, a linear fit is performed on the capacity change curve of each battery cell. The linear fit uses the least squares method, determining the fitting parameters by calculating the minimum error between the capacity change curve and the fitted line. The capacity change rate of each battery cell in different charge-discharge cycles is calculated; the formula for the capacity change rate is the slope of the fitted line. The resulting capacity change trend data for each battery cell is stored in a capacity trend data table, which includes the battery cell number, the number of charge-discharge cycles, and the corresponding capacity change rate. In step S44, statistical analysis is performed on the capacity change trend data of all battery cells. The statistical analysis includes calculating the mean and standard deviation of the capacity change rate for each battery cell. The mean is the arithmetic mean of the capacity change rates of all battery cells, and the standard deviation represents the dispersion of the capacity change rates. The calculated mean and standard deviation are used as the capacity balance index of the battery pack and stored in a capacity balance index data table, which includes the battery pack number, the mean capacity change rate, and the standard deviation. Finally, in step S45, based on the capacity balance index of the battery pack, the battery cells are divided into a high-capacity balance group and a low-capacity balance group. The standard deviation threshold for capacity change rate is set to 0.1. If the difference between the capacity change rate of a battery cell and the mean capacity change rate of the battery pack is less than or equal to 0.1, the battery cell is classified into the high-capacity balance group; if the difference is greater than 0.1, it is classified into the low-capacity balance group. The generated battery capacity balance data is recorded in tabular form, which includes the battery cell number, its capacity balance group, and the corresponding capacity change rate.

[0147] Preferably, step S5 includes the following steps:

[0148] Step S51: For each voltage interval in the voltage gradient distribution map, calculate the average voltage value and voltage fluctuation range of the battery cells in that interval to obtain voltage interval characteristic data;

[0149] Step S52: Normalize the voltage range feature data, temperature-current correlation data, and battery capacity balance data to generate a multi-channel battery feature dataset.

[0150] Step S53: Divide the multi-channel battery feature dataset into a battery feature training set and a battery feature validation set;

[0151] Step S54: Construct a preset model for evaluating charge and discharge performance based on the battery feature training set, and input the battery feature verification set into the preset model for evaluating charge and discharge performance to obtain model verification data;

[0152] Step S55: Adjust the model parameters of the preset model for charge-discharge performance evaluation based on the model validation data to obtain the charge-discharge performance evaluation model;

[0153] Step S56: Based on the charge-discharge performance evaluation model, predict the evolution trend of multi-channel battery performance and obtain a long-term battery performance evolution prediction report.

[0154] In this embodiment of the invention, in step S51, each voltage interval in the voltage gradient distribution map is first analyzed. Specifically, the average voltage value and voltage fluctuation range of the battery cells within each voltage interval are calculated. The average voltage value is the arithmetic mean of the voltages of all battery cells within the interval, and the voltage fluctuation range is the difference between the maximum and minimum voltage values ​​within the interval. These calculations are performed by data processing software, and the resulting voltage interval feature data is stored in a voltage interval feature data table, which includes the voltage interval number, average voltage value, and voltage fluctuation range. Next, in step S52, the voltage interval feature data, temperature-current correlation data, and battery capacity balance data are normalized. The normalization process uses a maximum-minimum normalization method to convert all data to the range of 0 to 1. Specifically, for each data point, the difference between it and the minimum value of the data column is calculated, and then divided by the difference between the maximum and minimum values ​​of the data column. The normalized data generates a multi-channel battery feature dataset, which is stored in tabular form and includes the normalized voltage interval feature data, temperature-current correlation data, and battery capacity balance data. In step S53, the multi-channel battery feature dataset is divided into a battery feature training set and a battery feature validation set. 70% of the data is used as the training set, and 30% as the validation set. This division is performed using random sampling to ensure the representativeness of the data distribution in both sets. In step S54, a pre-defined model for charge-discharge performance evaluation is constructed based on the battery feature training set. The model is built using the Support Vector Machine (SVM) algorithm from machine learning, and the model parameters are trained using the training set data. The battery feature validation set is input into the pre-defined model for charge-discharge performance evaluation to obtain model validation data. This data includes a comparison between the model's predictions and actual results on the validation set data, used to evaluate the model's accuracy and generalization ability. In step S55, the model parameters of the pre-defined model for charge-discharge performance evaluation are adjusted based on the validation data. The adjustment process uses a grid search method to optimize the SVM model parameters (such as the penalty parameter C and the kernel function parameter γ). The model performance under different parameter combinations is evaluated using cross-validation, and the optimal parameter combination is selected to obtain the charge-discharge performance evaluation model. Finally, in step S56, the evolution trend of multi-channel battery performance is predicted based on the charge-discharge performance evaluation model. The prediction process involves inputting current battery characteristic data, and the model calculates battery performance indicators at different future time points based on this input data, such as battery capacity, voltage stability, and current distribution uniformity. The prediction time span is set to the next 12 months, with a time step of once per month. The resulting long-term battery performance evolution prediction report is presented in the form of charts and tables. The charts include curves showing the changes in battery capacity, voltage stability, and current distribution uniformity over time, while the tables list the battery performance indicator values ​​at each time point in detail.

[0155] Of particular importance, step S56 includes the following steps:

[0156] Step S561: Obtain historical charge and discharge data of the multi-channel battery;

[0157] Step S561: Segment the historical charge and discharge data of the multi-channel battery, dividing the data into multiple time periods, each time period corresponding to a charge and discharge cycle;

[0158] Step S562: Input historical charge and discharge data into the charge and discharge performance evaluation model, and calculate the average voltage, average current, average temperature and average capacity of each battery cell within the time period.

[0159] Step S563: Perform trend analysis on the performance data of each battery cell and calculate its performance change rate over different time periods, including voltage change rate, current change rate, temperature change rate and capacity change rate.

[0160] Step S564: Conduct a comprehensive evaluation of the performance change rate of each battery cell to determine its performance stability over different time periods and obtain the performance stability level.

[0161] Step S565: Based on the performance stability level of each battery cell, predict its performance evolution trend in different time periods in the future, and generate a performance evolution prediction curve for each battery cell.

[0162] In this embodiment of the invention, in step S561, historical charge-discharge data of the multi-channel battery is first obtained from the data storage unit. This data includes voltage, current, temperature, and capacity data of each battery cell in different charge-discharge cycles. The data is recorded in time series form, with the timestamp of each data point accurate to the second. Next, in step S562, the historical charge-discharge data of the multi-channel battery is segmented. Specifically, the data is divided according to charge-discharge cycles, with each time period corresponding to a complete charge-discharge cycle. Segmentation is completed by data processing software, which automatically identifies the start and end points of the charge-discharge cycle based on the change characteristics of voltage and current. The data for each time period is stored in a separate data segment file, which contains the time period number, start timestamp, end timestamp, and voltage, current, temperature, and capacity data within that time period. In step S563, the historical charge-discharge data is input into the charge-discharge performance evaluation model. Within each time period, the model calculates the average voltage, average current, average temperature, and average capacity of each battery cell within that time period. The average value is calculated using the arithmetic mean method, which is to add up all data points within each time period and divide by the number of data points. The calculated average values ​​are stored in a performance data table, which includes the battery cell number, time period number, and corresponding average voltage, average current, average temperature, and average capacity. Proceed to step S564 to perform trend analysis on the performance data of each battery cell. Trend analysis uses a linear regression method to calculate the performance change rate of each battery cell over different time periods, including voltage change rate, current change rate, temperature change rate, and capacity change rate. Specifically, a linear regression analysis is performed with the time period as the independent variable and the performance indicators (voltage, current, temperature, and capacity) as the dependent variable. The calculated performance change rate is stored in a performance change rate data table, which includes the battery cell number, time period number, and corresponding voltage change rate, current change rate, temperature change rate, and capacity change rate. In step S565, a comprehensive evaluation of the performance change rate of each battery cell is performed. The comprehensive evaluation uses a weighted average method, assigning weights according to the importance of each performance indicator. For example, the weight of the capacity change rate is 0.4, the weight of the voltage change rate is 0.3, the weight of the current change rate is 0.2, and the weight of the temperature change rate is 0.1. The overall performance change rate of each battery cell over different time periods is calculated, and its performance stability level is determined based on the overall performance change rate. Performance stability levels are divided into three categories: high, medium, and low. Battery cells with an overall performance change rate less than 0.05 are rated as high stability, those between 0.05 and 0.1 are rated as medium stability, and those greater than 0.1 are rated as low stability. Performance stability levels are stored in a performance stability data table, which includes the battery cell number, time period number, and corresponding performance stability level.Finally, in step S566, based on the performance stability level of each battery cell, the performance evolution trend over different future time periods is predicted. The prediction process employs time series forecasting methods, such as the Autoregressive Moving Average (ARMA) model. Specifically, based on the historical performance data and current performance stability level of each battery cell, its performance indicators over different future time periods are predicted. The prediction results are presented as performance evolution prediction curves, which include predicted values ​​for voltage, current, temperature, and capacity over the next 12 months. The prediction curves are stored in a performance prediction data table, which contains the battery cell number, the prediction time period, and the corresponding predicted performance indicator values.

[0163] This specification also provides a multi-channel battery charge-discharge test system for performing the multi-channel battery charge-discharge test method described above. The multi-channel battery charge-discharge test system includes:

[0164] The voltage gradient layering module is used to acquire the voltage data of each battery cell in the multi-channel battery, perform voltage gradient layering processing on the voltage data of each battery cell, divide the battery cell into multiple voltage intervals, detect the number density of battery cells in each voltage interval, and generate a voltage gradient distribution map.

[0165] The current distribution difference analysis module is used to collect the current distribution coefficients of each connection point of the multi-channel battery; perform current distribution difference analysis on the current distribution coefficients; calculate the actual current distribution ratio of each unit in different charging and discharging stages; and generate current distribution difference data.

[0166] The temperature-current correlation module is used to collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery; perform thermal flow field simulation analysis on the overall temperature change data and identify the heat propagation path inside the battery pack; combine the local temperature data of each battery cell to determine the impact of the heat propagation path on the current distribution difference data and generate temperature-current correlation data.

[0167] The capacity change trend module is used to collect capacity data of each battery cell; perform capacity balance analysis on the capacity data of each battery cell, calculate the capacity change trend of each cell in different charge and discharge cycles, and generate battery capacity balance data.

[0168] The charge-discharge performance evaluation module is used to construct a charge-discharge performance evaluation model based on voltage gradient distribution map, temperature-current correlation data and battery capacity balance data; based on the charge-discharge performance evaluation model, it predicts the evolution trend of multi-channel battery performance and obtains a long-term battery performance evolution prediction report.

[0169] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, it is intended that all variations falling within the meaning and scope of the equivalents of the application be incorporated into the invention.

[0170] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A multi-channel battery charge-discharge test method, characterized in that, Includes the following steps: Step S1: Obtain the voltage data of each battery cell in the multi-channel battery, perform voltage gradient layering processing on the voltage data of each battery cell, divide the battery cell into multiple voltage intervals; detect the number density of battery cells in each voltage interval, and generate a voltage gradient distribution map. Step S2: Collect the current distribution coefficients of each connection point of the multi-channel battery; perform current distribution difference analysis on the current distribution coefficients, calculate the actual current distribution ratio of each unit in different charging and discharging stages, and generate current distribution difference data; Step S3: Collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery; perform thermal flow field simulation analysis on the overall temperature change data and identify the heat propagation path inside the battery pack; combine the local temperature data of each battery cell to determine the impact of the heat propagation path on the current distribution difference data and generate temperature-current correlation data. Step S4: Collect capacity data of each battery cell; perform capacity balance analysis on the capacity data of each battery cell, calculate the capacity change trend of each cell in different charge and discharge cycles, and generate battery capacity balance data. Step S5: Based on the voltage gradient distribution map, temperature-current correlation data, and battery capacity balance data, construct a charge-discharge performance evaluation model; based on the charge-discharge performance evaluation model, predict the evolution trend of multi-channel battery performance and obtain a long-term battery performance evolution prediction report.

2. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, In step S1, the voltage data of each battery cell in the multi-channel battery is obtained, and the voltage data of each battery cell is processed by voltage gradient layering to divide the battery cell into multiple voltage ranges, including: The voltage of each battery cell was measured three times using a voltmeter. The first measurement was taken when the battery cell was charged to 20%, the second measurement was taken when it was charged to 50%, and the third measurement was taken when it was charged to 80%. The voltage data collected three times were averaged and rounded to two decimal places to obtain the average voltage value of each battery cell. The battery cells are divided into three voltage ranges based on the average voltage value. When the average voltage value is lower than twice the arithmetic mean minus the standard deviation of the average voltage values ​​of all battery cells, it is marked as a low voltage range. When the average voltage value is between twice the arithmetic mean minus the standard deviation of the average voltage values ​​of all battery cells and twice the arithmetic mean plus the standard deviation, it is marked as the medium voltage range. A high voltage range is defined as a range where the average voltage value is higher than twice the arithmetic mean plus the standard deviation of the average voltage values ​​of all battery cells.

3. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, Step S1, which involves detecting the number density of battery cells within each voltage range, includes: Within each voltage range, sensors are used to scan each battery cell individually, recording the unique identifier, initial voltage value, and location coordinates of each battery cell. Within each voltage range, multiple fixed monitoring points are set up, and the spacing between the monitoring points is adjusted according to the arrangement density of the battery cells; At each monitoring point, sensors are used to measure local voltage and current density, and the local voltage and current density data at each monitoring point are recorded. Based on the local voltage and current density data of the monitoring points, the number of battery cells within the coverage area is counted at each monitoring point, and the volume within the coverage area is calculated according to the arrangement of the battery cells, thus obtaining the number density. The quantity density data of each monitoring point is standardized and converted into relative density values; Based on the standardized quantity density data, a voltage gradient distribution map is plotted with the voltage range as the horizontal axis and the relative density value as the vertical axis, with each monitoring point corresponding to a data point.

4. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, Step S2 includes the following steps: Step S21: At each connection point of the multi-channel battery, use a high-precision current sensor to collect current data three times: at the start of charging, when charging reaches 50%, and at the end of charging. Each data collection lasts for 10 seconds. Step S22: Round the current value collected each time and keep three decimal places to obtain the processed current value; Step S23: Calculate the current distribution coefficient for each connection point based on the processed current value, where the current distribution coefficient is the ratio of the current value of each connection point to the sum of the current values ​​of all connection points; Step S24: Calculate the actual current distribution ratio of each battery cell at different charging and discharging stages. By analyzing the current distribution coefficient of each connection point, determine the actual current distribution ratio of each battery cell at the start of charging, when charging to 50%, and at the end of charging, so as to obtain the current distribution ratio. Step S25: Perform time series analysis on the current distribution ratio of each battery cell in different charging and discharging stages, record the changes in the current distribution ratio of each battery cell at different time points, and obtain current time series information; Step S26: Perform volatility analysis on the current time series information, calculate the fluctuation amplitude of the current distribution ratio of each battery cell in different charging and discharging stages, and generate current distribution difference data.

5. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, Step S3 involves collecting overall temperature change data of the multi-channel battery during charging and discharging, as well as local temperature data of each battery cell, including: The temperature data of each battery cell surface of the multi-channel battery is collected and processed at multiple points. Temperature sensors are used to collect temperature data three times at the start of charging, when charging reaches 50%, and when charging ends. Each collection lasts for 5 seconds, so as to obtain the local temperature data of each battery cell at different charging and discharging stages. Time series analysis was performed on the collected local temperature data to record the temperature changes of each battery cell at different time points, and the temperature change curve of each battery cell was obtained. The temperature change curve is subjected to fluctuation analysis to calculate the temperature fluctuation amplitude of each battery cell in different charging and discharging stages, and the temperature fluctuation characteristic data of each battery cell are obtained. The temperature fluctuation characteristic data are classified to obtain the temperature stability classification results; Based on the temperature stability classification results, the overall temperature change data of the multi-channel battery during the charging and discharging process were determined.

6. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, Step S3 involves performing thermal flow field simulation analysis on the overall temperature change data and identifying the heat propagation paths within the battery pack, including: Identify the multi-channel battery structure distribution diagram; Acquire the physical structure data of the multi-channel battery, which includes the arrangement, connection method and physical location of the battery cells; Based on the physical structure data, spatial distribution analysis of the overall temperature change data is performed to generate a temperature spatial distribution map; On the temperature distribution map, the local temperature data of each battery cell is spatially mapped to correspond the local temperature data with the physical location of the battery cell, thus generating a temperature mapping map. Based on the multi-channel battery structure distribution map, a thermal flow field simulation analysis is performed on the temperature mapping map to simulate the heat propagation process inside the battery pack and obtain the heat propagation path.

7. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, In step S3, by combining the local temperature data of each battery cell, the impact of the heat propagation path on the current distribution difference data is determined, including: Time series analysis was performed on the local temperature data of each battery cell to extract the temperature characteristic values ​​of each battery cell at different charging and discharging stages, including the average temperature, temperature fluctuation amplitude and temperature change rate. The heat transfer path is segmented into multiple sub-path segments, each containing several battery cells. The starting and ending battery cell positions of each sub-path segment are recorded. For each battery cell in a sub-path segment, calculate the cross-correlation coefficient between its local temperature data and current distribution difference data; For each sub-path segment, the battery cells are divided into high correlation group and low correlation group according to the cross-correlation coefficient; For battery cells in high correlation groups, the linear relationship strength between heat propagation path and current distribution difference data is detected to obtain temperature-current correlation data.

8. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, Step S4 includes the following steps: Step S41: On each battery cell of the multi-channel battery, use a capacity tester to collect capacity data of each battery cell. Collect capacity data three times at 10%, 50% and 90% of the charge and discharge cycle, with each collection lasting 10 minutes. Step S42: Perform time series analysis on the capacity data of each battery cell, record the capacity change of each battery cell in different charge and discharge cycle stages, and obtain the capacity change curve of each battery cell. Step S43: Perform linear fitting on the capacity change curve of each battery cell, calculate the capacity change rate of each battery cell in different charge and discharge cycles, and obtain the capacity change trend data of each battery cell. Step S44: Perform statistical analysis on the capacity change trend data of all battery cells, calculate the mean and standard deviation of the capacity change rate of each battery cell, and obtain the capacity balance index of the battery pack. Step S45: Based on the battery pack's capacity balance index, divide the battery cells into high-capacity balance groups and low-capacity balance groups, and generate battery capacity balance data.

9. The multi-channel battery charge-discharge test method according to claim 1, characterized in that, Step S5 includes the following steps: Step S51: For each voltage interval in the voltage gradient distribution map, calculate the average voltage value and voltage fluctuation range of the battery cells in that interval to obtain voltage interval characteristic data; Step S52: Normalize the voltage range feature data, temperature-current correlation data, and battery capacity balance data to generate a multi-channel battery feature dataset. Step S53: Divide the multi-channel battery feature dataset into a battery feature training set and a battery feature validation set; Step S54: Construct a preset model for evaluating charge and discharge performance based on the battery feature training set, and input the battery feature verification set into the preset model for evaluating charge and discharge performance to obtain model verification data; Step S55: Adjust the model parameters of the preset model for charge-discharge performance evaluation based on the model validation data to obtain the charge-discharge performance evaluation model; Step S56: Based on the charge-discharge performance evaluation model, predict the evolution trend of multi-channel battery performance and obtain a long-term battery performance evolution prediction report.

10. A multi-channel battery charge / discharge testing system, characterized in that, For performing the multi-channel battery charge-discharge test method as described in claim 1, the multi-channel battery charge-discharge test system includes: The voltage gradient layering module is used to acquire the voltage data of each battery cell in the multi-channel battery, perform voltage gradient layering processing on the voltage data of each battery cell, divide the battery cell into multiple voltage intervals, detect the number density of battery cells in each voltage interval, and generate a voltage gradient distribution map. The current distribution difference analysis module is used to collect the current distribution coefficients of each connection point of the multi-channel battery; perform current distribution difference analysis on the current distribution coefficients; calculate the actual current distribution ratio of each unit in different charging and discharging stages; and generate current distribution difference data. The temperature-current correlation module is used to collect overall temperature change data and local temperature data of each battery cell during the charging and discharging process of the multi-channel battery; perform thermal flow field simulation analysis on the overall temperature change data and identify the heat propagation path inside the battery pack; combine the local temperature data of each battery cell to determine the impact of the heat propagation path on the current distribution difference data and generate temperature-current correlation data. The capacity change trend module is used to collect capacity data of each battery cell; perform capacity balance analysis on the capacity data of each battery cell, calculate the capacity change trend of each cell in different charge and discharge cycles, and generate battery capacity balance data. The charge-discharge performance evaluation module is used to construct a charge-discharge performance evaluation model based on voltage gradient distribution map, temperature-current correlation data and battery capacity balance data; based on the charge-discharge performance evaluation model, it predicts the evolution trend of multi-channel battery performance and obtains a long-term battery performance evolution prediction report.

Images