A green function experimental calibration system based on battery thermal response

Abstract

The application provides a kind of green function experimental calibration system based on battery thermal response, it is related to lithium battery parameter analysis technical field, including: the green function calibration module based on experimental data determines the green function value of each temperature measuring point of battery monomer under each working condition;With kriging interpolation analysis, final temperature measuring point judgment and high confidence position point selection, to determine the sample point increase and decrease module of final sample point set under each working condition;The green function extension module based on inverse distance weighted interpolation analysis final sample point set, to determine the green function value distribution of battery monomer under each working condition.The application obtains green function based on experimental data, improves the accuracy of green function and the prediction accuracy of subsequent temperature field reconstruction from the source, and fuses kriging interpolation, final temperature measuring point judgment, high confidence position point selection and inverse distance weighted interpolation analysis and other steps, so that the finally formed battery monomer green function value distribution is more reasonable and accurate.

Description

Technical Field

[0001] This invention relates to the field of Green's function calibration technology, specifically to an experimental calibration system for Green's function based on battery thermal response. Background Technology

[0002] The core of the method for reconstructing the temperature field of lithium-ion batteries based on the principle of linear decomposition of thermal field lies in the accurate acquisition of the Green function. However, the acquisition of the Green function currently mainly relies on numerical simulation methods such as finite element method. These methods usually require simplified modeling of battery geometry, material thermal properties, boundary conditions, etc., which makes it difficult to truly reflect the actual thermal response of the battery under complex working conditions, resulting in significant deviations between the calculation calibration results and the experiments.

[0003] In the prior art, the battery pack temperature field reconstruction method and related device with the publication number "CN121118386A" decomposes each individual cell in the battery pack into several sub-models, each sub-model containing a single non-homogeneous boundary; then, based on the temperature change of each grid point in each sub-model under the influence of a heat source, a Green's function is constructed to describe the relationship between the grid point temperature response and the heat source input of each sub-model. In this way, the overall Green's function is decomposed into individual sub-Green's functions and simulated one by one. Compared with the overall modeling and calculation, this significantly reduces the modeling and solving difficulty and ensures the accuracy of the Green's function solution results.

[0004] However, existing technologies still have significant shortcomings. For example, current technologies only set a single temperature measurement point on the surface of the battery cell to measure transient temperature rise. The calculated Green's function is only at a single location on the battery cell. However, the Green's function at different locations on the battery cell will have certain differences. That is, the Green's function calculated at a single location by existing technologies cannot fully represent the overall Green's function distribution of the battery cell. If only a single Green's function is used to represent the overall Green's function of the battery cell, it will have an adverse effect on the subsequent reconstruction of the battery cell temperature field. Although existing technologies also mention combining simulation to construct the field distribution, the simulation will produce modeling errors due to the simplification of materials, interfaces and boundary conditions, which will have an adverse effect on the fitting calculation of the Green's function.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide an experimental calibration system for the Green's function based on battery thermal response, so as to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A Green's function experimental calibration system based on battery thermal response includes:

[0009] The operating condition setting module is used to set the internal heat source operating condition driven by the internal heat source, the lateral conduction operating condition driven by the front and rear side thermal excitation, and the bottom forced convection operating condition driven by the forced convection heat transfer of the liquid cooling system.

[0010] The Green function calibration module is used to analyze the heat source input power of a battery cell under various operating conditions and the transient temperature rise at each temperature measuring point on the battery cell, and to obtain the Green function value of each temperature measuring point under various operating conditions.

[0011] The sample point addition / reduction module is used to analyze other temperature measuring points in the neighborhood of any given temperature measuring point based on the Kriging interpolation method. The interpolation is used to obtain the Green's function fitting value of the temperature measuring point under each operating condition, and it is compared with the true Green's function value of the temperature measuring point under each operating condition to determine the final temperature measuring point under each operating condition. For the overlapping parts of the neighborhood of each temperature measuring point, the coefficient of variation is calculated and judged for each location point, and high confidence location points under each operating condition are selected. The final temperature measuring points and high confidence location points are summarized for each operating condition to determine the final sample point set under each operating condition.

[0012] The Green's function extension module analyzes the final sample point set under various operating conditions based on inverse distance weighted interpolation to determine the distribution of Green's function values ​​of individual cells under various operating conditions. In the inverse distance weighted interpolation, an adaptive confidence weight is introduced to correct the distance weight of each point in the final sample point set. The adaptive confidence weight of high-confidence location points is not higher than that of the final temperature measurement point and is positively correlated with the number of its neighbors and negatively correlated with its corresponding coefficient of variation.

[0013] Furthermore, the internal heat source operating conditions are set as follows: the battery cell operates under constant current charging, and the coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature, and the front and rear sides and other surfaces of the battery cell are treated as thermally insulating boundaries.

[0014] The lateral conduction condition is set as follows: the battery cell is not powered, and a constant heat source input power is applied to the front and rear sides of the battery cell. The coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature, and the remaining surfaces of the battery cell, except for the front and rear sides, are treated as thermally insulating boundaries.

[0015] The bottom forced convection condition is set as follows: the battery cell is not powered, and the coolant temperature in the liquid cooling system at the bottom of the battery cell maintains a constant temperature difference with the ambient temperature, and the front and rear sides and other surfaces of the battery cell are treated as thermally insulating boundaries.

[0016] Furthermore, a thermocouple array is arranged on the front side of the battery cell to form a 5*5 temperature measuring point array.

[0017] For internal heat source conditions, the heat source input power of the battery cell is calculated at each moment during the duration of the condition, based on the charging current and state of charge of the battery cell at each moment during the duration of the condition.

[0018] For lateral conduction conditions, the heat source input power of a single battery cell is a set constant value at each moment during the duration of this condition.

[0019] For the bottom forced convection condition, the heat source input power of the battery cell at each moment during the duration of this condition is determined based on the constant temperature difference between the coolant temperature and the ambient temperature.

[0020] For any temperature measuring point, the calculation logic for the transient temperature rise at any moment during the operating period is as follows: extract the temperature value of the temperature measuring point at this moment, and subtract the temperature value of the temperature measuring point at the beginning of the operating period from it, and use it as the transient temperature rise of the temperature measuring point at this moment.

[0021] For any temperature measuring point, the transient temperature rise at any time under any operating condition and the heat source input power of the battery cell at any time under the same operating condition are jointly solved to obtain the Green's function value of the temperature measuring point at any time under each operating condition.

[0022] Furthermore, for any temperature measuring point, the logic for setting its neighborhood is as follows: on the front side of the battery cell, a circular area is drawn with the temperature measuring point as the center and a preset length as the radius to serve as the neighborhood of the temperature measuring point, and the preset length satisfies the constraint that there are at least four temperature measuring points in the neighborhood.

[0023] Furthermore, for any temperature measuring point, the logic for determining the distribution of the Green's function fitting value of its neighborhood at any time under any operating condition is as follows: a two-dimensional coordinate system is constructed on the front side of the battery cell. Temperature measuring points other than the temperature measuring point are extracted from the neighborhood of the temperature measuring point as sample measuring points. The coordinates of the sample measuring points are determined in the two-dimensional coordinate system, and the Green's function value of the sample measuring points at the same time under the same operating condition is extracted. The coordinates and Green's function value of the sample measuring points are analyzed by Kriging interpolation to determine the distribution of the Green's function fitting value of the neighborhood of the temperature measuring point at this time under this operating condition.

[0024] Furthermore, the logic for determining the final temperature measurement point is as follows: For any temperature measurement point, extract its Green's function value at each time under each operating condition as the true value of the Green's function, and retrieve the distribution of the Green's function fitting values ​​of its neighborhood at each time under each operating condition. Extract the Green's function fitting value of the temperature measurement point at each time under each operating condition from the Green's function fitting value distribution. If the relative error between the true value of the Green's function and the Green's function fitting value of a temperature measurement point is not greater than a preset threshold at a certain time under a certain operating condition, then at that time under that operating condition, the temperature measurement point is regarded as the final temperature measurement point.

[0025] Furthermore, for any operating condition at any time, the logic for determining the final sample point set is as follows: Summarize the temperature measurement points determined as the final temperature measurement points at this time under this operating condition, and take the other location points on the front side of the battery cell, excluding the temperature measurement points, as candidate location points. For each candidate location point, determine the number of its neighbors. If only one neighborhood contains the candidate location point, then the candidate location point is not considered a high-confidence location point. Otherwise, extract the Green's function fitting value of the candidate location point at this time under this operating condition from the Green's function fitting value distribution of each neighborhood containing it, and calculate the coefficient of variation. If the calculated coefficient of variation is not greater than the coefficient of variation threshold, then at this time under this operating condition, the candidate location point is considered a high-confidence location point. Summarize the final temperature measurement points and high-confidence location points at this time under this operating condition to form the final sample point set at this time under this operating condition.

[0026] Furthermore, for the final sample point set at any time under any operating condition, the logic for determining the final Green's function value of the final temperature measurement point within it is as follows: extract the Green's function value of the final temperature measurement point at this time under this operating condition, and directly use it as the final Green's function value at this time under this operating condition.

[0027] For the final set of sample points at any time under any working condition, the logic for determining the final Green's function value of the high-confidence location point is as follows: determine all neighborhoods containing the high-confidence location point, and extract the Green's function fitting value of the high-confidence location point at this time under this working condition from the distribution of Green's function fitting values ​​of each neighborhood at this time under this working condition, and take the average value as its final Green's function value at this time under this working condition.

[0028] The calculation logic for the distribution of the Green's function value on the front side of a battery cell at any given moment under any operating condition is as follows: The final set of sample points at this moment under this operating condition is taken as the target set of sample points. The final temperature measurement point and high-confidence location points within the target set are considered as sample location points. All other location points on the front side of the battery cell, excluding the sample location points, are considered as locations to be fitted. For any location to be fitted, the distance weight of each sample location point relative to itself is calculated. Based on the type of each sample location point (including final temperature measurement points and high-confidence location points), the adaptive confidence weight of each sample location point is determined. For each sample location point, its distance weight relative to the location to be fitted is multiplied by its adaptive confidence weight to obtain the corrected weight relative to the location to be fitted. Based on the corrected weight, the final Green's function values ​​of each sample location point are weighted and averaged to determine the Green's function value of the location to be fitted at this moment under this operating condition, thereby determining the distribution of the Green's function value of the battery cell at this moment under this operating condition.

[0029] Furthermore, for any sample location, its distance weight relative to the location to be fitted is negatively correlated with the distance between the two.

[0030] Furthermore, for the sample location points belonging to the final temperature measurement point, their adaptive confidence weight is constantly set to 1;

[0031] For sample location points that belong to high-confidence location points, their adaptive confidence weights are set between 0 and 1.

[0032] Compared with the prior art, the beneficial effects of the present invention are:

[0033] The Green's function experimental calibration system based on battery thermal response of this invention eliminates modeling errors to a large extent by conducting experiments under different operating conditions and then obtaining the Green's function based on the experimental data, thereby improving the accuracy of the Green's function and the prediction accuracy of subsequent temperature field reconstruction from the source. In addition, this invention also realizes the experimental solution of the Green's function values ​​at multiple locations of battery cells by setting multiple temperature measurement points, and integrates Kriging interpolation, final temperature measurement point judgment and selection of high confidence location points to screen out multiple high-quality sample points for inverse distance weighted interpolation analysis, so that the final distribution of the Green's function values ​​of battery cells is more reasonable and accurate. Attached Figure Description

[0034] Figure 1 This is a schematic flowchart of the overall method of the present invention;

[0035] Figure 2 This is a schematic diagram showing the decomposition of the heat source input power of a single battery cell. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0037] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0038] Example:

[0039] Please see Figures 1-2 This invention provides an experimental calibration system for Green's function based on battery thermal response, comprising:

[0040] The operating condition setting module is used to set the internal heat source operating condition driven by the internal heat source, the lateral conduction operating condition driven by the front and rear side thermal excitation, and the bottom forced convection operating condition driven by the forced convection heat transfer of the liquid cooling system.

[0041] It should be noted that, for battery cells in the battery pack except those located at the edges, their boundary conditions can be uniformly divided into three categories: 1) The front and rear sides are in contact with other battery cells and are considered as heat conduction boundaries; 2) The bottom is in contact with the liquid cooling system and is considered as forced convection boundaries; 3) The remaining surfaces are in contact with thermal insulation materials or contained in a closed space under actual operating conditions and are considered as thermal insulation boundaries.

[0042] The internal heat source condition is used to simulate the temperature change of a battery cell caused solely by its own heat generation. The specific settings are as follows: the battery cell operates under constant current charging, and the coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature. This can be achieved through a temperature control device with heating and cooling functions, which will not be elaborated here. This ensures that the battery cell does not effectively exchange heat with the outside through the coolant. Furthermore, the front and rear sides and other surfaces of the battery cell are treated as thermally insulating boundaries. This can be achieved by coating with low thermal conductivity coatings such as polystyrene, aluminum silicate fiber, or vacuum insulation panels, so that the surface of the battery cell does not exchange heat with the outside. Thus, the temperature change of the battery cell under the internal heat source condition is only affected by its own heat generation.

[0043] The lateral conduction mode is used to simulate the temperature change of a battery cell caused only by thermal excitation from the front and rear sides. The specific settings are as follows: the battery cell is not powered, so that no heat is generated inside the battery cell. A constant thermal excitation with a heat source input power is applied to the front and rear sides of the battery cell. This thermal excitation can be applied through devices such as electric heating films or heat flow plates. The coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature. This can be achieved through a temperature control device with heating and cooling functions. This ensures that the battery cell does not exchange heat effectively with the outside through the coolant. In addition, all surfaces of the battery cell except the front and rear sides are treated as thermally insulating boundaries. This can be achieved by coating with low thermal conductivity coatings such as polystyrene, aluminum silicate fiber, or vacuum insulation panels. This ensures that the battery cell exchanges heat with the outside only through the front and rear sides. Thus, the temperature change of the battery cell under the lateral conduction mode is only affected by the thermal excitation from the front and rear sides.

[0044] The bottom forced convection mode is used to simulate the temperature change of a battery cell caused solely by forced convection heat transfer from the liquid cooling system. The specific settings are as follows: the battery cell is not powered, thus preventing heat generation within the cell itself. A constant temperature difference is maintained between the coolant temperature in the liquid cooling system at the bottom of the cell and the ambient temperature. This can be achieved through a temperature control device with heating and cooling functions to force convection heat transfer to the battery cell. Furthermore, the front and rear sides and other surfaces of the battery cell are treated as thermally insulating boundaries. This can be achieved by coating with low thermal conductivity coatings such as polystyrene, aluminum silicate fiber, or vacuum insulation panels, ensuring that the battery cell surface does not exchange heat with the outside environment. Therefore, the temperature change of the battery cell under the bottom forced convection mode is only affected by the forced convection heat transfer from the liquid cooling system.

[0045] The Green function calibration module is used to analyze the heat source input power of a battery cell under various operating conditions and the transient temperature rise at each temperature measuring point on the battery cell, and to obtain the Green function value of each temperature measuring point under various operating conditions.

[0046] As one implementation method, this technical solution selects CATL ternary lithium battery as the battery cell, with a nominal voltage of 3.7 V, a capacity of 117 Ah, and external dimensions of 34*220*105 mm. A thermocouple array is arranged on the front side of the battery cell to form a 5*5 temperature measuring point array, that is, 25 temperature measuring points are set on the front side of the battery cell for temperature acquisition.

[0047] Furthermore, a liquid cooling circulation system commonly used in power batteries is used as the liquid cooling system at the bottom of the battery cell. It includes a water pump, flow meter, harmonica tube liquid cooling plate, radiator, water tank, and ball valve. The circulation process is as follows: the coolant flows into multiple parallel harmonica tube liquid cooling plates after being split through the inlet pipe, and finally merges into the outlet main pipe. After being cooled by the radiator, it forms a circulation. The inlet and outlet main pipes are made of aluminum tubes with an outer diameter of 20 mm; the harmonica tubes are 150 mm wide and 2.4 mm thick; the bottom of the battery cell and the liquid cooling plate are thermally coupled through the module shell, the casing, and thermally conductive silicone grease with a thermal conductivity of 3 W / (m·K) to simulate the actual module working conditions.

[0048] The specific scheme for applying internal heat source conditions to the battery cell is as follows: the ITECH6012C-300-150 charging and discharging equipment is used to charge the battery cell at a constant current. The battery charge and discharge rate is 1C, that is, the battery cell is charged at a rate equal to its rated capacity. For the battery cell with a nominal capacity of 117 Ah in this technical solution, the charging current corresponding to 1C is 117A. The coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature. The front and rear sides and other surfaces of the battery cell are all treated as thermally insulating boundaries.

[0049] It should be noted that the ITECH6012C-300-150 charging and discharging equipment has a voltage range of 0-300 V, a current range of ±150 A, and a power range of ±12 kW. It can operate stably in constant current / constant voltage mode and output current / voltage data in real time. The voltage / current data is input to the LabVIEW program through the Smacq USB3122 data acquisition card. The voltage acquisition accuracy can reach 0.001 V, and the current acquisition accuracy can reach 0.001 A, which meets the requirements of this solution for applying internal heat source conditions to the battery cells.

[0050] The specific scheme for applying lateral conduction to the battery cell is as follows: the battery cell is not powered on so that no heat is generated inside the battery cell. A heat source with a constant input power of 5W is applied to both the front and rear sides of the battery cell. Specifically, the heat excitation can be applied through devices such as electric heating film and heat flow plate to simulate the heat excitation of other battery cells when the battery cell is working in the battery pack. The 5W setting is used to simulate the heat excitation of the front and rear sides of the battery cell during actual operation. The coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature. All surfaces of the battery cell except the front and rear sides are treated as thermally insulating boundaries.

[0051] The specific scheme for applying bottom forced convection to the battery cell is as follows: the battery cell is not powered on so that no heat is generated inside the battery cell. The coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be constant 10 degrees Celsius lower than the ambient temperature to simulate the bottom forced heat exchange experienced by the battery cell during operation. The front and rear sides and other surfaces of the battery cell are all treated as thermally insulating boundaries.

[0052] It should be noted that the settings for regulating the coolant temperature and for treating the surface of the battery cells as an adiabatic boundary have been discussed earlier and will not be repeated here.

[0053] Specifically, for the internal heat source condition, the heat source input power of a single battery cell under this condition is determined based on the charging current; for the lateral conduction condition, the heat source input power of a single battery cell under this condition is a set constant value; and for the bottom forced convection condition, the heat source input power of a single battery cell under this condition is determined based on the temperature difference between the coolant temperature and the ambient temperature. The specific mathematical expressions are as follows:

[0054]

[0055] In the formula, For battery cells in The heat source input power at that moment, For the duration of the working condition, it is a time variable;

[0056] In the formula, This indicates the internal heat source operating condition. In this condition, the heat input of a single battery cell comes from the internal heat generated during charging. However, since the internal heat generation of a single battery cell cannot be directly controlled, the formula for the heat generation of a single battery cell is used to calculate the equivalent heat source input power. This refers to the charging current of a single battery cell. Under internal heat source conditions, the battery cell is in a constant current charging state, hence the charging current... These are constant values; for example, in this technical solution, the battery charge / discharge rate is 1C, and the charging current is 117A. For battery cells in The terminal voltage at any given time can be acquired using the ITECH6012C-300-150 charging / discharging device, or by connecting an external voltmeter or multimeter, in this technical solution. For battery cells in The state of charge at any given time is based on the initial state of charge of the battery cell at the start of the internal heat source condition, the charging current of the battery cell under the internal heat source condition, and the state of charge from the end of the internal heat source condition to the start of the internal heat source condition. The duration of a moment is estimated, specifically expressed as follows: , This refers to the initial state of charge of a single battery cell at the moment of initiation of internal heat source operation, which can be specifically measured using the open-circuit voltage method. This refers to the nominal capacity of a single battery cell. Indicates the start time of the operating condition. Indicates that the internal heat source operating condition has been cut off. Duration of a moment For battery cells in The temperature at any given time can be measured at various temperature points on each individual battery cell. The average temperature at time t. For battery cells in and The open-circuit voltage under various states of charge and temperatures can be calculated by interpolation using the OCV-SOC mapping table. The OCV-SOC mapping table records the open-circuit voltage values ​​under various states of charge and temperatures. The specific interpolation method is common knowledge in the field of heat sources and will not be elaborated here. Indicates in The partial derivative of the open-circuit voltage with respect to temperature can be obtained by differentiating the OCV curves at different temperatures.

[0057] It should be noted that, This represents the power loss caused by the difference between the terminal voltage and the open-circuit voltage during the charging process of a single battery cell. This loss is converted into heat. The influence of temperature change on open-circuit voltage is considered, and multiplied by current and temperature, it reflects how temperature change affects the heat generation of the battery. Therefore, this technical solution integrates these two parts to construct a formula for the heat generation of a single battery cell to calculate the equivalent heat source input power.

[0058] In the formula, This indicates the lateral conduction condition. Under this condition, the heat source input to the battery cell comes from the external thermal excitation applied to the front and rear sides of the battery cell with a constant heat source input power. For example, in this technical solution, if a heat source input power of 5W is applied to both the front and rear sides of the battery cell, the heat source input power under the lateral conduction condition is constant at 10W.

[0059] In the formula, This indicates a forced convection mode at the bottom. Under this mode, the heat source input for the battery cell comes from the forced convection heat transfer between the bottom of the battery cell and the liquid cooling system. These represent the mass flow rate and specific heat capacity of the coolant in a liquid cooling system. These are the ambient temperature and the coolant temperature, respectively, and the difference between them is constant. For example, in this technical solution, the difference between the two is constant at 10 degrees Celsius.

[0060] As shown in the attached diagram of the following instruction manual Figure 2 As shown, the total heat source input power of a single battery cell can be decomposed into three categories, specifically including the heat source input power under internal heat source conditions. Heat source input power under lateral conduction conditions And the heat source input power under forced convection conditions at the bottom. ;

[0061] Specifically, for any temperature measuring point, the transient temperature rise at any moment during the duration of the operating condition is calculated as follows: extract the temperature value of the temperature measuring point at this moment, and subtract the temperature value of the temperature measuring point at the beginning of the operating condition from it, which is taken as the transient temperature rise of the temperature measuring point at this moment. For example, for the internal heat source operating condition, the transient temperature rise of the temperature measuring point at any moment during the duration of the internal heat source operating condition is the temperature difference between the temperature measuring point at this moment and the temperature at the beginning of the internal heat source operating condition.

[0062] For any operating condition, the Green's function of any temperature measurement point on a single battery cell under that condition is represented in discrete matrix form, as follows:

[0063]

[0064] In the formula, to For the first The temperature measuring point at the first Under these conditions, from the first sampling time to the... Transient temperature rise at each sampling time, to For the first The temperature measuring point at the first Under these conditions, from the first sampling time to the... Green's function value at each sampling time. to For the first The temperature measuring point at the first Under these conditions, from the first sampling time to the... The heat source input power at each sampling time The time interval between adjacent sampling moments, for example, in this technical solution, the sampling frequency is once every five seconds. It lasts for 5 seconds. The total number of samples taken during the duration of the operating condition. As an index for temperature measurement points on individual battery cells, this technical solution sets up a total of 25 temperature measurement points. , As an index of operating conditions, this technical solution sets up three operating conditions. ;

[0065] It should be noted that the discrete matrix generated above can not only directly connect experimental data with physical models, but also be quickly called online in the subsequent temperature field reconstruction stage, which greatly improves the real-time performance and scalability of temperature field prediction and reconstruction. Through the experimental calibration of the above three working conditions, the corresponding Green's function discrete matrix can be obtained respectively. The overall thermal response of the battery cell can be regarded as the linear superposition of the three, thus providing a high-fidelity foundation for the rapid reconstruction of the temperature field of the battery module.

[0066] The sample point addition / reduction module is used to analyze other temperature measuring points in the neighborhood of any given temperature measuring point based on the Kriging interpolation method. The interpolation is used to obtain the Green's function fitting value of the temperature measuring point under each operating condition, and it is compared with the true Green's function value of the temperature measuring point under each operating condition to determine the final temperature measuring point under each operating condition. For the overlapping parts of the neighborhood of each temperature measuring point, the coefficient of variation is calculated and judged for each location point, and high confidence location points under each operating condition are selected. The final temperature measuring points and high confidence location points are summarized for each operating condition to determine the final sample point set under each operating condition.

[0067] The logic for setting the neighborhood of any temperature measuring point is as follows: On the front side of the battery cell, a circular area is drawn with the temperature measuring point as the center and a preset length as the radius to serve as the neighborhood of the temperature measuring point. The preset length satisfies the constraint that there are at least four temperature measuring points in the neighborhood. That is, it ensures that there are at least three temperature measuring points in the neighborhood of a temperature measuring point in addition to the temperature measuring point located at the center of the circle, so as to improve the reliability of the subsequent neighborhood interpolation results. The specific value of the preset length is selected according to the actual scenario. For example, for the 5*5 temperature measuring point matrix of this technical solution, the preset length value cannot be less than the product of the distance between adjacent temperature measuring points and the square root of two to satisfy the above constraint. Specifically, the value can be between 2 and 3 times the distance between adjacent temperature measuring points.

[0068] Specifically, for any temperature measuring point, the logic for determining the distribution of the Green's function fitting value of its neighborhood under any operating condition at any time is as follows: A two-dimensional coordinate system is constructed on the front side of the battery cell. Specifically, the geometric center of the front side of the battery cell can be used as the origin, the upper edge of the front side of the battery cell can be used as the X-axis, and the perpendicular X-axis can be used as the Y-axis. Temperature measuring points other than the temperature measuring point are extracted from the neighborhood of the temperature measuring point as sample measuring points. The coordinates of the sample measuring points are determined in the two-dimensional coordinate system, and the Green's function value of the sample measuring points under the same operating condition at the same time is extracted. The coordinates and Green's function value of the sample measuring points are analyzed by Kriging interpolation to determine the distribution of the Green's function fitting value of the neighborhood of the temperature measuring point under this operating condition at this time.

[0069] As one implementation method, for a temperature measuring point located at the center of the temperature measuring point array, the logic for determining the distribution of the Green's function fitting values ​​of its neighborhood at the first sampling time under the internal heat source condition is as follows: extract the Green's function values ​​of the sample measuring points in its neighborhood at the first sampling time under the internal heat source condition, and extract the coordinates of the sample measuring points in its neighborhood in the two-dimensional coordinate system. Analyze the coordinates of the sample measuring points and the Green's function values ​​of the sample measuring points at the first sampling time under the internal heat source condition using the Kriging interpolation method to obtain the distribution of the Green's function fitting values ​​of the neighborhood of the temperature measuring point at the first sampling time under the internal heat source condition.

[0070] It should be noted that the above time can be the sampling time mentioned above, or any time under the corresponding working condition. If it is the former, the value can be directly taken from the discrete matrix. If it is the latter, it is necessary to first extract the Green's function value of the sample measurement point at each sampling time under the same working condition, and then extrapolate the Green's function value of the sample measurement point at that time under the corresponding working condition through relevant curve fitting methods (such as conventional cubic spline interpolation).

[0071] It should be noted that, since the internal medium of a battery cell is relatively uniform, the Green's function distribution generally exhibits strong smoothness and continuity. Therefore, in the Kriging interpolation process of this technical solution, a Gaussian model that fits it is used as the theoretical model. It is common knowledge for those skilled in the art to analyze multiple sample points in a planar region based on the Kriging interpolation method to determine the distribution of fitted values ​​in the region. The specific implementation scheme will not be elaborated here.

[0072] The logic for determining the final temperature measurement point is as follows: For any temperature measurement point, extract its Green's function value at each time under each operating condition as the true value of the Green's function. Specifically, the extraction is based on the discrete matrix, and the extraction method is the same as that for the sample measurement points, which will not be elaborated here. Then, retrieve the distribution of the Green's function fitting values ​​of its neighborhood at each time under each operating condition. Extract the Green's function fitting values ​​of the temperature measurement point at each time under each operating condition from the distribution of the Green's function fitting values. If the relative error between the true value of the Green's function and the Green's function fitting value of a temperature measurement point is not greater than a preset threshold at a certain time under a certain operating condition, then at that time under that operating condition, the temperature measurement point is regarded as the final temperature measurement point.

[0073] It should be noted that for a battery cell with a uniform dielectric, the Green's function values ​​at each location should exhibit strong smoothness and continuity. For a temperature measurement point, if the Green's function fitted value obtained by fitting the sample measurement points in its neighborhood differs significantly from the actual Green's function value at that temperature measurement point, it indicates that there is a large difference in the Green's function value between that temperature measurement point and the sample measurement points in its neighborhood. That is, the temperature measurement point does not satisfy the distribution law of the Green's function value in its neighborhood, thus indicating that the temperature measurement point is very likely an outlier. In order to improve the accuracy of subsequent Green's function fitting expansion, it is treated as an outlier temperature measurement point and removed to avoid adverse effects on subsequent Green's function fitting expansion. Conversely, if the Green's function fit is normal, it indicates that the temperature measurement point is reasonable and is retained as the final temperature measurement point.

[0074] It should be noted that relative error is used to measure the degree of deviation between the true value and the fitted value. Specifically, it is calculated by first calculating the absolute difference between the true value and the fitted value, and then dividing the absolute difference by the true value to obtain the relative error. The larger the relative error, the greater the deviation between the true value and the fitted value, and thus the higher the probability that the temperature measuring point is an abnormal temperature measuring point. The preset threshold for relative error is generally set between 10% and 20%, and the specific value is set by the staff according to the actual situation. For example, if the preset threshold is set to 10%, it means that if the relative error of a temperature measuring point exceeds 10%, the temperature measuring point is identified as an abnormal temperature measuring point.

[0075] Specifically, for any time under any operating condition, the logic for determining the final sample point set is as follows: summarize the temperature measurement points determined as the final temperature measurement points at this time under this operating condition, and take the other location points on the front side of the battery cell other than the temperature measurement points as candidate location points. For each candidate location point, determine the number of its neighbors. If only one neighborhood contains the candidate location point, it means that the Green's function fitting value of the candidate location point at this time under this operating condition is unique, lacks a reference for comparison, and cannot determine its confidence level. Therefore, the candidate location point is not taken as a high confidence location point.

[0076] Conversely, from the distribution of Green's function fitting values ​​in each neighborhood containing the selected location, the Green's function fitting value of the selected location at this moment under this operating condition is extracted one by one, and the coefficient of variation is calculated. If the calculated coefficient of variation is not greater than the coefficient of variation threshold, it means that the Green's function fitting values ​​of the selected location in each neighborhood are close to each other under this operating condition, which means that the fitting values ​​of the selected location in each neighborhood are relatively reliable. Therefore, under this operating condition, the selected location is regarded as a high-confidence location. However, if the calculated coefficient of variation is greater than the coefficient of variation threshold, it means that the fitting values ​​of the selected location in each neighborhood differ greatly under this operating condition, which means that the fitting values ​​of the selected location in at least one neighborhood are unreasonable. Therefore, under this operating condition, the selected location is not regarded as a high-confidence location. The final temperature measurement points and high-confidence location points under this operating condition are summarized to form the final sample point set under this operating condition.

[0077] It should be noted that the coefficient of variation is the ratio of the standard deviation to the mean. The larger the value, the greater the difference between the data. The specific calculation process is common knowledge to those skilled in the art and will not be elaborated here. The threshold for the coefficient of variation can generally be set between 5% and 20%, and the specific value is set by the staff according to the actual situation. For example, if the threshold for the coefficient of variation is set to 10%, it means that if the coefficient of variation of a candidate location point at a certain moment under a certain working condition exceeds 10%, it is considered that the deviation between the fitted values ​​of the candidate location point at this moment under this working condition is large. Therefore, at this moment under this working condition, the candidate location point will not be regarded as a high confidence location point.

[0078] The Green's function extension module is based on inverse distance weighted interpolation analysis of the final sample point set under each operating condition to determine the distribution of the Green's function value of the battery cell under each operating condition. In the inverse distance weighted interpolation, an adaptive confidence weight is introduced to correct the distance weight of each point in the final sample point set. The adaptive confidence weight of the high confidence point is not higher than that of the final temperature measurement point and is positively correlated with the number of its neighbors and negatively correlated with its corresponding coefficient of variation.

[0079] Among them, for the final sample point set at any time under any working condition, the logic for determining the final Green's function value of the final temperature measurement point is as follows: extract the Green's function value of the final temperature measurement point at this time under this working condition, and directly use it as the final Green's function value at this time under this working condition.

[0080] Specifically, for the final sample point set at any time under any working condition, the logic for determining the final Green's function value of the high-confidence location point is as follows: determine all neighborhoods containing the high-confidence location point, and extract the Green's function fitting value of the high-confidence location point at this time under this working condition from the distribution of Green's function fitting values ​​of each neighborhood at this time under this working condition, and take the average value as its final Green's function value at this time under this working condition. The setting of taking the average value avoids the influence of randomness and accidental errors, making the obtained final Green's function value more reliable.

[0081] The calculation logic for the distribution of the Green's function value on the front side of a battery cell at any given moment under any operating condition is as follows: The final set of sample points at this moment under any operating condition is taken as the target set of sample points. The final temperature measurement point and high-confidence location points within the target set are considered as sample location points. All other location points on the front side of the battery cell, excluding the sample location points, are considered as locations to be fitted. For any location to be fitted, the distance weight of each sample location point relative to itself is calculated. Based on the type of each sample location point, the adaptive confidence weight of each sample location point is determined. The types of sample location points include final temperature measurement points and high-confidence location points. For any sample location point, its distance weight relative to the location to be fitted is multiplied by its adaptive confidence weight to obtain the corrected weight relative to the location to be fitted. Based on the corrected weight, the final Green's function values ​​of each sample location point are weighted and averaged to determine the Green's function value of the location to be fitted at this moment under any operating condition, thereby determining the distribution of the Green's function value of the battery cell at this moment under any operating condition.

[0082] In this embodiment, for any sample location, its distance weight relative to the location to be fitted is negatively correlated with the distance between the two locations. That is, the greater the distance between the sample location and the location to be fitted, the smaller the distance weight of the sample location relative to the location to be fitted. Therefore, when fitting the location to be fitted, the sample location closer to it is given priority. In this embodiment, a typical weighting formula is used to calculate the distance weight of the sample location relative to the location to be fitted. The specific mathematical expression is as follows:

[0083]

[0084] In the formula, For the first The sample location point and the first The distance between the points to be fitted is taken here as the Euclidean distance between their coordinates in a two-dimensional coordinate system. The specific formula for calculating the Euclidean distance will not be elaborated here. The preset power exponent is used to characterize the influence of distance on the fitting results. Because the Green's function values ​​of individual battery cells exhibit strong smoothness and continuity, the power exponent is set to a small value, generally between 1 and 2. The specific value can be set by the staff according to the actual situation. For the first The sample location point compared to the first The distance weight of the i-th position to be fitted, the i-th The sample location point and the first The larger the distance between the points to be fitted, the greater the... The sample location point pair The lower the reference value of each point to be fitted, the better. The smaller it is, For the index of the sample location point, The index of the location to be fitted;

[0085] Among them, for the sample location points belonging to the final temperature measurement point, since their final Green function value is obtained through experimental calculation and has a high degree of reliability, their adaptive confidence weight is set to 1 to ensure that their final Green function value is effectively used when calculating the Green function value at the location point to be fitted.

[0086] Among them, for sample location points belonging to high confidence location points, since their final Green's function value is obtained by interpolation based on a small number of local temperature measurement points, the reliability is questionable. Therefore, their adaptive confidence weight is set between 0 and 1, and is negatively correlated with the coefficient of variation and positively correlated with the number of neighborhoods containing them.

[0087] As one implementation method, the specific mathematical expression for the adaptive confidence weight is as follows:

[0088]

[0089] In the formula, For the first Adaptive confidence weights for each sample location point Indicates the final temperature measurement point. Indicates the first Each sample location is used as the final temperature measurement point. This indicates a high-confidence location. Indicates the first Each sample location is a high-confidence location. For the first The coefficient of variation is calculated at each sample location point. The specific calculation logic is as follows: from the distribution of Green's function fitted values ​​of each neighborhood containing the sample location point, the Green's function fitted value of the sample location point at this moment under this working condition is extracted one by one, and the coefficient of variation is calculated. The specific calculation process has been given above and will not be repeated here. The larger the value, the more likely it is to contain the first... Within each neighborhood of the nth sample location point, for the nth sample location point... The Green's function fitting values ​​for each sample location point vary significantly, resulting in a low reliability of the final calculated Green's function value. Therefore, we use [a different approach]. To the first The reliability of the final Green's function value for each sample location is measured once.

[0090] In the formula, For including the first The larger the neighborhood number of the i-th sample location, the better the calculation of the i-th... When calculating the final Green's function value for each sample location, the more Green's function fitting values ​​are considered, the better the influence of randomness and accidental errors can be avoided. The higher the reliability of the final Green's function value for each sample location, the better. The neighborhood size is the baseline value used for... To avoid over-amplifying the adaptive confidence weights, a normalization process is performed. The logic for obtaining the weights is as follows: for each high-confidence location, the number of its neighbors is counted. The number of neighbors corresponding to each high-confidence location is then sorted in descending order, and the median is taken as the baseline value for the number of neighbors. for The relative size provides a benchmark. Greater than This indicates that in calculating the first... When calculating the final Green's function value for the nth sample location, if the Green's function fit value included in the calculation exceeds the baseline, its adaptive confidence weight can be appropriately increased; otherwise, it indicates that the calculation of the nth sample location point is flawed. When calculating the final Green's function value for each sample location, if the Green's function fit value included in the calculation does not exceed the baseline, its adaptive confidence weight can be appropriately reduced. Therefore, this is done by... To the first The reliability of the final Green's function value for each sample location is measured a second time, and then... The form constitutes the interaction item, representing the interaction with the first... The reliability of the final Green's function value for each sample location is comprehensively evaluated. The settings include setting a cutoff value (i.e., 1) for the adaptive confidence weight to avoid the problem that the adaptive confidence weight of high confidence points exceeds the final temperature measurement point;

[0091] Based on the above, the mathematical expression for fitting the Green's function value of the location to be fitted, based on the final Green's function value of the sample location, is as follows:

[0092]

[0093] In the formula, For the first The sample location point compared to the first Corrected weights for each position to be fitted For the first The final Green's function value for each sample location point. For the first Green's function values ​​for the points to be fitted. The number of sample locations is given. Based on the final Green's function value of each sample location, a weighted average is used to calculate the Green's function value of each location to be fitted.

[0094] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0095] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0096] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

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

Claims

1. A Green's function experimental calibration system based on battery thermal response, characterized in that, include: The operating condition setting module is used to set the internal heat source operating condition driven by the internal heat source, the lateral conduction operating condition driven by the front and rear side thermal excitation, and the bottom forced convection operating condition driven by the forced convection heat transfer of the liquid cooling system. The Green function calibration module is used to analyze the heat source input power of a battery cell under various operating conditions and the transient temperature rise at each temperature measuring point on the battery cell, and to obtain the Green function value of each temperature measuring point under various operating conditions. The sample point addition / reduction module is used to analyze other temperature measuring points in the neighborhood of any given temperature measuring point based on the Kriging interpolation method. The interpolation is used to obtain the Green's function fitting value of the temperature measuring point under each operating condition, and it is compared with the true Green's function value of the temperature measuring point under each operating condition to determine the final temperature measuring point under each operating condition. For the overlapping parts of the neighborhood of each temperature measuring point, the coefficient of variation is calculated and judged for each location point, and high confidence location points under each operating condition are selected. The final temperature measuring points and high confidence location points are summarized for each operating condition to determine the final sample point set under each operating condition. The Green's function extension module analyzes the final sample point set under various operating conditions based on inverse distance weighted interpolation to determine the distribution of Green's function values ​​of individual cells under various operating conditions. In the inverse distance weighted interpolation, an adaptive confidence weight is introduced to correct the distance weight of each point in the final sample point set. The adaptive confidence weight of high-confidence location points is not higher than that of the final temperature measurement point and is positively correlated with the number of its neighbors and negatively correlated with its corresponding coefficient of variation.

2. The Green's function experimental calibration system based on battery thermal response according to claim 1, characterized in that, The internal heat source operating conditions are set as follows: the battery cell operates under constant current charging, the coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature, and the front and rear sides and other surfaces of the battery cell are treated as thermally insulating boundaries. The lateral conduction condition is set as follows: the battery cell is not powered, and a constant heat source input power is applied to the front and rear sides of the battery cell. The coolant temperature in the liquid cooling system at the bottom of the battery cell is set to be the same as the ambient temperature, and the remaining surfaces of the battery cell, except for the front and rear sides, are treated as thermally insulating boundaries. The bottom forced convection condition is set as follows: the battery cell is not powered, and the coolant temperature in the liquid cooling system at the bottom of the battery cell maintains a constant temperature difference with the ambient temperature, and the front and rear sides and other surfaces of the battery cell are treated as thermally insulating boundaries.

3. The Green's function experimental calibration system based on battery thermal response according to claim 2, characterized in that: A thermocouple array is arranged on the front side of the battery cell to form a 5*5 temperature measuring point array. For internal heat source conditions, the heat source input power of the battery cell is calculated at each moment during the duration of the condition, based on the charging current and state of charge of the battery cell at each moment during the duration of the condition. For lateral conduction conditions, the heat source input power of a single battery cell is a set constant value at each moment during the duration of this condition. For the bottom forced convection condition, the heat source input power of the battery cell at each moment during the duration of this condition is determined based on the constant temperature difference between the coolant temperature and the ambient temperature. For any temperature measuring point, the calculation logic for the transient temperature rise at any moment during the operating period is as follows: extract the temperature value of the temperature measuring point at this moment, and subtract the temperature value of the temperature measuring point at the beginning of the operating period from it, and use it as the transient temperature rise of the temperature measuring point at this moment. For any temperature measuring point, the transient temperature rise at any time under any operating condition and the heat source input power of the battery cell at any time under the same operating condition are jointly solved to obtain the Green's function value of the temperature measuring point at any time under each operating condition.

4. The Green's function experimental calibration system based on battery thermal response according to claim 3, characterized in that, For any temperature measurement point, the logic for setting its neighborhood is as follows: On the front side of the battery cell, a circular area is drawn with the temperature measurement point as the center and a preset length as the radius to serve as the neighborhood of the temperature measurement point, and the preset length satisfies the constraint that there are at least four temperature measurement points in the neighborhood.

5. The Green's function experimental calibration system based on battery thermal response according to claim 3, characterized in that, For any temperature measuring point, the logic for determining the distribution of the Green's function fitting value of its neighborhood under any operating condition at any time is as follows: Construct a two-dimensional coordinate system on the front side of the battery cell, extract temperature measuring points other than the temperature measuring point from the neighborhood of the temperature measuring point as sample measuring points, determine the coordinates of the sample measuring points in the two-dimensional coordinate system, and extract the Green's function value of the sample measuring points at the same time under the same operating condition. Analyze the coordinates and Green's function value of the sample measuring points using Kriging interpolation to determine the distribution of the Green's function fitting value of the neighborhood of the temperature measuring point under this operating condition at this time.

6. The Green's function experimental calibration system based on battery thermal response according to claim 3, characterized in that, The logic for determining the final temperature measurement point is as follows: For any temperature measurement point, extract its Green's function value at each time under each operating condition as the true value of the Green's function, and retrieve the distribution of the Green's function fitting values ​​of its neighborhood at each time under each operating condition. Extract the Green's function fitting value of the temperature measurement point at each time under each operating condition from the Green's function fitting value distribution. If the relative error between the true value of the Green's function and the Green's function fitting value of a temperature measurement point is not greater than a preset threshold at a certain time under a certain operating condition, then at that time under that operating condition, the temperature measurement point is regarded as the final temperature measurement point.

7. The Green's function experimental calibration system based on battery thermal response according to claim 3, characterized in that, For any operating condition at any time, the logic for determining the final sample point set is as follows: Summarize the temperature measurement points determined as the final temperature measurement points at this time under this operating condition, and take the other location points on the front side of the battery cell, excluding the temperature measurement points, as candidate location points. For each candidate location point, determine the number of its neighbors. If only one neighborhood contains the candidate location point, then the candidate location point is not considered a high-confidence location point. Otherwise, extract the Green's function fitting value of the candidate location point at this time under this operating condition from the Green's function fitting value distribution of each neighborhood containing it, and calculate the coefficient of variation. If the calculated coefficient of variation is not greater than the coefficient of variation threshold, then at this time under this operating condition, the candidate location point is considered a high-confidence location point. Summarize the final temperature measurement points and high-confidence location points at this time under this operating condition to form the final sample point set at this time under this operating condition.

8. The Green's function experimental calibration system based on battery thermal response according to claim 7, characterized in that, For any set of final sample points under any operating condition at any time, the logic for determining the final Green's function value of the final temperature measurement point within it is as follows: extract the Green's function value of the final temperature measurement point under this operating condition at this time, and directly use it as the final Green's function value under this operating condition at this time. For the final set of sample points at any time under any working condition, the logic for determining the final Green's function value of the high-confidence location point is as follows: determine all neighborhoods containing the high-confidence location point, and extract the Green's function fitting value of the high-confidence location point at this time under this working condition from the distribution of Green's function fitting values ​​of each neighborhood at this time under this working condition, and take the average value as its final Green's function value at this time under this working condition. The calculation logic for the distribution of the Green's function value on the front side of a battery cell at any given moment under any operating condition is as follows: The final set of sample points at this moment under this operating condition is taken as the target set of sample points. The final temperature measurement point and high-confidence location points within the target set are considered as sample location points. All other location points on the front side of the battery cell, excluding the sample location points, are considered as locations to be fitted. For any location to be fitted, the distance weight of each sample location point relative to itself is calculated. Based on the type of each sample location point (including final temperature measurement points and high-confidence location points), the adaptive confidence weight of each sample location point is determined. For each sample location point, its distance weight relative to the location to be fitted is multiplied by its adaptive confidence weight to obtain the corrected weight relative to the location to be fitted. Based on the corrected weight, the final Green's function values ​​of each sample location point are weighted and averaged to determine the Green's function value of the location to be fitted at this moment under this operating condition, thereby determining the distribution of the Green's function value of the battery cell at this moment under this operating condition.

9. The Green's function experimental calibration system based on battery thermal response according to claim 8, characterized in that, For any sample location, its distance weight relative to the location to be fitted is negatively correlated with the distance between the two locations.

10. The Green's function experimental calibration system based on battery thermal response according to claim 8, characterized in that, For sample location points belonging to the final temperature measurement point, their adaptive confidence weight is constantly set to 1; For sample location points that belong to high-confidence location points, their adaptive confidence weights are set between 0 and 1.

Images