Intelligent temperature control system of new energy battery pack

Abstract

The application relates to the technical field of battery temperature control, and discloses an intelligent temperature control system for a new energy battery pack; the system comprises a battery division module, a digital modeling module, a parameter prediction module, a state simulation module and a temperature control execution module; the battery division module is used for dividing the battery pack into battery units; the digital modeling module is used for constructing a digital battery model; the parameter prediction module is used for predicting temperature parameters in a target period; the state simulation module is used for simulating temperature control states in the target period; and the temperature control execution module is used for formulating intelligent temperature control measures; the battery pack is divided into scattered battery units, and the scattered battery units are fused and mapped into a digital battery model, so that the collection amount of relevant data in the battery units can be reduced, independent and accurate position positioning functions can be provided for temperature control simulation analysis of local positions in the battery pack, the phenomenon of inaccurate position positioning during overall analysis of the battery pack is avoided, and it is ensured that the battery cells at different positions in the battery pack can maintain independent temperature control analysis states.

Description

Technical Field

[0001] This invention relates to the field of battery temperature control technology, and more specifically, to an intelligent temperature control system for a new energy battery pack. Background Technology

[0002] As the core component of electric vehicles, the operating temperature of new energy battery packs directly affects the safety of the battery pack. During the charging and discharging process of the battery pack, the cells in different locations will have large temperature differences due to factors such as heat dissipation conditions, cooling channel layout, uneven current distribution and heat accumulation effect. When the temperature is too high, it is easy to cause thermal runaway, and when the temperature is too low, it is easy to cause capacity decay and power reduction. Therefore, it is necessary to carry out precise and intelligent temperature control management of the battery pack.

[0003] Reference patent application CN120999197A discloses an intelligent temperature control system for new energy battery packs, including: a sensing module for real-time acquisition of dynamic temperature field data of individual battery cells and modules; and an analysis module for receiving dynamic temperature field data of individual battery cells and modules, integrating historical temperature control data to construct a temperature change prediction model, analyzing the currently received dynamic temperature field data to predict and output the predicted values ​​of standardized temperature characteristic parameters within a preset time interval in the future. It accurately acquires battery temperature information in real time, integrates historical data to construct a prediction model to predict future temperature changes in advance, and rationally sets temperature control priorities based on battery aging, user fast charging, and range requirements. Existing new energy battery pack temperature control systems typically employ a holistic approach to data acquisition and analysis within the battery pack. Due to the large number of cells within the battery pack, their dense distribution, and significant differences in thermal characteristics, the amount of data acquired in this holistic approach is enormous, increasing the workload of single data analysis. Furthermore, because cells in different locations exhibit variations in heat dissipation conditions and cooling channel layouts, this holistic approach overlooks subtle differences between individual cells. Consequently, the battery pack temperature control decisions fail to match actual thermal demands, reducing the accuracy of intelligent temperature control.

[0004] In view of this, the present invention proposes an intelligent temperature control system for new energy battery packs to solve the above problems. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an intelligent temperature control system for a new energy battery pack, applied to an intelligent temperature controller, comprising: The battery segmentation module is used to extract the structural and positional features of the battery pack, using structural features as the basis for morphological segmentation and positional features as the basis for spatial segmentation, to divide the battery pack into battery cells. The digital modeling module is used to acquire multi-dimensional temperature control data of battery cells, construct a virtual structural model based on structural characteristics, and fuse and map the virtual structural model with the multi-dimensional temperature control data to construct a digital battery model. The parameter prediction module is used to determine the target time period based on the time period planning criteria, and predict the temperature parameters of the digital battery model during the target time period through the time series parameter prediction model. The temperature parameters include temperature peak, temperature valley, temperature change rate and the proportion of stabilization time. The state simulation module is used to import the temperature parameters of the target time period into the digital battery model, set the simulation mechanism, and simulate the temperature control state of the battery cell during the target time period. The temperature control execution module is used to identify the runaway unit from the battery cells based on the temperature control status, perform timing analysis on the runaway unit, and formulate intelligent temperature control measures.

[0006] Furthermore, structural features include geometric dimensions, cell arrangement, cell spacing, and cooling channel features; positional features include the battery pack reference coordinates and cell center coordinates.

[0007] Furthermore, the method for dividing battery cells is as follows: Cells with the same structural features within the battery pack are grouped into one structural family, resulting in B structural families. The Euclidean distance between the center coordinates of A cells and the reference coordinates of the battery pack is calculated one by one using the Euclidean distance formula. The Euclidean distance is then normalized to obtain A relative position weights. Using relative position weights as the dividing criterion, within the same structural group, the K-means clustering algorithm is used to divide the cells with the same weight level in B structural groups into a battery cell, resulting in C array-distributed battery cells.

[0008] Furthermore, the multi-dimensional temperature control data includes surface temperature, internal temperature, operating voltage, charging and discharging current, and internal resistance; The method for constructing a digital battery model is as follows: Based on the structural characteristics of a single battery cell, a three-dimensional battery cell model is constructed using three-dimensional modeling technology. Then, according to the actual spatial relationship of the battery cells, the three-dimensional battery cell models belonging to the same battery unit are sequentially spliced ​​together to generate a virtual structure model. Create a blank dataset on the battery cell, set the number of data slots in the blank dataset to match the number of battery cells, and fill the data slots in the blank dataset with the multidimensional temperature control data from the C battery cells one by one to generate the temperature control dataset. By using digital twin technology, two unidirectional transmission mapping channels are established between the virtual structural model and the temperature control dataset, and a clock-synchronized time axis is set on both mapping channels to fuse and map the virtual structural model into a digital battery model.

[0009] Furthermore, the time-period planning criteria are as follows: the proportion of temperature rise is inversely proportional to the first weighting factor, and the proportion of temperature drop is inversely proportional to the second weighting factor.

[0010] Furthermore, the method for determining the target time period is as follows: The rated temperature range and optimal temperature value of the battery pack are queried through the battery management system. The minimum and maximum values ​​of the rated temperature range are recorded as the lower limit and upper limit of the temperature range, respectively. The temperature rise is calculated by subtracting the upper limit of the temperature from the optimal temperature value, and the temperature drop is calculated by subtracting the optimal temperature from the lower limit of the temperature. The temperature range is calculated by subtracting the upper temperature limit from the lower temperature limit. The temperature drop and temperature rise are then divided by the temperature range to calculate the temperature drop percentage and temperature rise percentage, respectively. The average duration of a temperature drop and a temperature rise in the battery pack were retrieved from the database and denoted as the temperature drop duration and temperature rise duration, respectively. The proportion of temperature drop is used as the second weighting factor, and the proportion of temperature rise is used as the first weighting factor. The duration of temperature drop and the duration of temperature rise are weighted, summed and averaged to calculate the duration of the period. Starting from the current moment, the endpoint is marked after delaying the timeline by one time period, and the time period between the starting point and the endpoint is recorded as the target time period.

[0011] Furthermore, the temperature control state includes a temperature controllable state, a temperature rise runaway state, and a temperature drop runaway state; The simulation mechanism is as follows: simulation is performed based on a single time group.

[0012] Furthermore, the simulation method for temperature control is as follows: The duration of the target period is compared with the proportion of the stability maintenance period to calculate the dynamic duration. The duration of the target period is then divided by the dynamic duration to calculate the duration multiple. When the duration multiple is an integer, the target time period is divided into D time groups based on the duration span of the dynamic duration; When the duration multiple is not an integer, the duration of two consecutive moments is used as a reduction increment, and the dynamic duration is continuously reduced until the duration multiple of the reduced dynamic duration is an integer. Then, the target time period is divided into D time groups based on the duration span of the reduced dynamic duration. The temperature peak, temperature valley, temperature change rate, and stability duration percentage of all times within the D time groups are summarized to form D simulation data sets. The D simulation data sets are then imported into the digital battery model in chronological order to simulate the temperature values ​​of the digital battery model in the D time groups. Plot a temperature distribution map with time on the horizontal axis and temperature on the vertical axis, including the temperature distribution curve. Then, plot the lower boundary line corresponding to the lower limit value and the upper boundary line corresponding to the upper limit value on the temperature distribution map. Using the standard that one time group corresponds to one time curve, the temperature distribution curve is divided into D time curves, and the positional relationship between the D time curves and the lower and upper boundary lines is observed. When all time-matter curves are between the lower and upper boundary lines, the temperature control is in a controllable state; when some time-matter curves are below the lower boundary line and some time-matter curves are between the lower and upper boundary lines, the temperature control is in a cooling runaway state; when some time-matter curves are above the upper boundary line and some time-matter curves are between the lower and upper boundary lines, the temperature control is in a heating runaway state. When the time curve is simultaneously above the upper boundary line, between the lower boundary line and the upper boundary line, and below the lower boundary line, if the temperature value at the last time point is above the upper boundary line, the temperature control state is a temperature rise runaway state; if the temperature value at the last time point is below the lower boundary line, the temperature control state is a temperature drop runaway state; if the temperature value at the last time point is between the lower boundary line and the upper boundary line, the temperature control state is a controllable state.

[0013] Furthermore, during the identification of out-of-control units, the number of battery units in a controllable temperature control state within the target time period is counted. The number of controllable states is divided by the number of states at all times within the target time period to calculate the controllable percentage. Battery units with a controllable percentage less than a preset controllable threshold are recorded as out-of-control units, resulting in E out-of-control units.

[0014] Furthermore, the method for formulating intelligent temperature control measures is as follows: Arrange the temperature control status of D time groups in E out-of-control units in chronological order to form a status queue. Based on the cell center coordinates of the cells in the runaway unit, a heating system that provides heating and temperature rise function and a cooling system that provides cooling and temperature drop function for the runaway unit are selected. When the temperature control status of the time group is in a state of uncontrolled temperature rise, the intelligent temperature controller formulates timing measures to control the cooling system to execute cooling and temperature reduction commands; when the temperature control status of the time group is in a state of uncontrolled temperature drop, the intelligent temperature controller formulates timing measures to control the heating system to execute heating and temperature rise commands. Traverse the temperature control states of D time groups in the state queue, generate D time measures, and then arrange and summarize the D time measures in sequence to formulate intelligent temperature control measures.

[0015] Technical advantages of the intelligent temperature control system for new energy battery packs of the present invention: This invention divides an integrated battery pack into discrete battery cells based on structural and positional features, and then fuses these battery cells with multidimensional temperature control data to create a digital battery model. This approach reduces the amount of data collected from individual battery cells by breaking down the entire battery pack into smaller units. Furthermore, by incorporating digital twin technology, it provides independent and precise location analysis for localized temperature control within the battery pack, avoiding inaccurate location analysis common in overall battery pack analysis. This also ensures that cells in different locations within the battery pack maintain independent temperature control analysis states, thereby improving the accuracy of the intelligent temperature control analysis results for the battery pack. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the architecture of an intelligent temperature control system for a new energy battery pack provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of a module of an intelligent temperature control system for a new energy battery pack provided in Embodiment 1 of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Example 1: Please refer to Figure 1-2 As shown in this embodiment, an intelligent temperature control system for a new energy battery pack is applied to an intelligent temperature controller and includes: The battery segmentation module extracts the structural and positional features of the battery pack, using structural features as the basis for morphological segmentation and positional features as the basis for spatial segmentation, dividing the battery pack into array-distributed battery cells. The battery pack of a new energy vehicle consists of multiple cells connected in a series-parallel hybrid configuration. Each cell can function as an independent entity. Due to the large number of cells in the battery pack and their integrated structure, performing overall data collection and analysis on the battery pack not only results in a large amount of data being collected each time, increasing the burden of single data collection, but also makes it impossible to accurately locate the temperature control position during subsequent temperature control processing of the battery pack, thus reducing the effectiveness of intelligent temperature control. Therefore, it is necessary to divide the overall battery pack into independently distributed battery units.

[0019] When dividing a battery pack into battery cells, in order to ensure the rationality and accuracy of the division of battery cells, it is necessary to divide them based on structural and positional characteristics. Structural features refer to the inherent physical architecture and geometric arrangement of the battery pack, reflecting the specific information about the spatial connection, boundary, and pathway layout of the battery cells; Specifically, structural features include geometric dimensions, cell arrangement, cell spacing, and cooling channel characteristics.

[0020] Geometric dimensional characteristics refer to the inherent dimensional and structural information of the battery cell itself. Geometric dimensional characteristics include, but are not limited to, length, width, and thickness.

[0021] Cell arrangement characteristics refer to the specific information about the spatial arrangement of cells within a battery pack. Cell arrangement characteristics include, but are not limited to, the number of rows, columns, layers, and number of cells.

[0022] Cell spacing characteristics refer to specific information about the gap between adjacent cells; cell spacing characteristics include, but are not limited to, the spacing between two adjacent cells within the same module, the spacing between two adjacent modules, and the boundary gap between the battery pack edge and the cells.

[0023] Cooling channel characteristics refer to the specific location and direction of the cooling plates and coolant within the cooling mechanism; cooling channel information includes, but is not limited to, the flow direction of the liquid cooling plate, the location of branch points, the location of inlet and outlet points, and the distribution density of the heat dissipation fins.

[0024] In this embodiment, the structural features are automatically read from the three-dimensional CAD model of the battery pack.

[0025] Positional characteristics refer to the absolute and relative positions of each cell in the battery pack in the three-dimensional space of the battery pack, reflecting the specific information of the cell's spatial position in the battery pack. Specifically, the location features include the battery pack's reference coordinates and the cell's center coordinates; The battery pack reference coordinates refer to the three-dimensional coordinates of the geometric center point of the battery pack in a three-dimensional coordinate system, which serve as the basis for subsequent analysis and calculation of the cell positions.

[0026] Cell center coordinates refer to the three-dimensional coordinates of the geometric center of each cell in a three-dimensional coordinate system, which serve as the basis for identifying and locating the battery cell.

[0027] When extracting the battery pack reference coordinates, the 3D CAD model of the battery pack is imported into the 3D space. The lower left corner of the 3D CAD model is used as the origin of the coordinate system, the length direction of the battery pack is used as the X-axis, the width direction of the battery pack is used as the Y-axis, and the height direction of the battery pack is used as the Z-axis. The geometric center of the battery pack and the geometric center of A cells are measured one by one to obtain the reference coordinates of the battery pack and the center coordinates of A cells.

[0028] After obtaining the structural and positional features, the A cells in the battery pack can be divided in terms of space and form, and the integrated battery pack can be divided into multiple arrayed battery units. In this embodiment, a battery cell refers to a collection of cells that includes at least one battery cell and can individually represent the structure and position of a certain local area in the battery pack. It also serves as the direct object for subsequent intelligent temperature control analysis and control of the battery pack.

[0029] The method for dividing battery cells is as follows: The structural features of A cells in the battery pack are compared for consistency. Cells with the same structural features are grouped into one structural group to obtain B structural groups. The Euclidean distance between the center coordinates of A individual cells and the reference coordinates of the battery pack is calculated one by one using the Euclidean distance formula. The Euclidean distance is then normalized to obtain A relative position weights. The relative position weights range from 0 to 1. When the cell is closer to the center point of the battery pack, the single-cell heat accumulation effect of the cell is more obvious, and the relative position weight is closer to 1; conversely, the opposite is true. Using relative position weights as the dividing criterion, within the same structural group, the K-means clustering algorithm is used to divide the cells with the same weight level in B structural groups into a battery cell, resulting in C array-distributed battery cells.

[0030] The weight level is the range of weight values ​​obtained through K-means clustering algorithm analysis, and is used as the relative position weight of all cells in the battery cell. For example, the weight level of the first battery cell is 0-0.15, that is, cells with a relative position weight of 0-0.15 can be assigned to the same battery cell. This ensures that each battery cell contains several thermally coupled cells and becomes the smallest management unit for subsequent temperature control.

[0031] The digital modeling module acquires multi-dimensional temperature control data of the battery cell, constructs a virtual structural model of the battery cell, and fuses and maps the virtual structural model with the multi-dimensional temperature control data to construct a digital battery model. Multidimensional temperature control data is a dataset of physical quantities used to represent the thermal and electrochemical states of battery cells. Specifically, the multi-dimensional temperature control data includes surface temperature, internal temperature, operating voltage, charging and discharging current, and internal resistance.

[0032] Surface temperature refers to the temperature of the outer surface of each cell in the battery cell; internal temperature refers to the temperature inside each cell in the battery cell; operating voltage refers to the terminal voltage between the positive and negative terminals of each cell in the battery cell; charging and discharging current refers to the current during the charging or discharging process of each cell in the battery cell; internal resistance refers to the sum of the ohmic internal resistance and polarization internal resistance of each cell in the battery cell. In this embodiment, the multidimensional temperature control data is obtained by querying the battery management system.

[0033] After obtaining the multidimensional temperature control data, it is necessary to summarize the multidimensional temperature control data belonging to the same cell and the multidimensional temperature control data belonging to the same battery unit.

[0034] After obtaining multidimensional temperature control data, the multidimensional temperature control data can only represent the battery cell at the physical entity level. It cannot dynamically map the real-time dynamic changes of the battery cell at the digital virtual level, resulting in an inability to intuitively and accurately understand the specific temperature changes of the battery cell. In order to achieve real-time dynamic understanding of the temperature changes of the battery cell, it is necessary to establish a digital battery model that can be fused and mapped with the multidimensional temperature control data. In this embodiment, the digital battery model is a virtual model based on digital twin technology, which can represent the three-dimensional physical model of the battery cell and real-time multi-dimensional temperature control data in a real-time and comprehensive manner. It also serves as a prerequisite for analyzing and predicting the temperature changes of the battery cell.

[0035] The method for constructing a digital battery model is as follows: Based on the geometric dimensions, cell arrangement, cell spacing, and cooling channel characteristics of a single cell, a three-dimensional cell model of a single cell is constructed using three-dimensional modeling technology. Then, according to the actual spatial relationship of the cells, the three-dimensional cell models belonging to the same battery unit are sequentially spliced ​​together to generate a virtual structure model. The number of cells in each of the C battery cells is counted one by one. A blank dataset is created on the battery cell, and data slots with the same number of cells are set in the blank dataset. The multidimensional temperature control data from the C battery cells are then filled into the data slots of the blank dataset one by one to generate a temperature control dataset. The blank dataset is a collection of data used to summarize the multidimensional temperature control data as a whole, and the data slots are the locations where specific multidimensional temperature control data is imported to ensure that the multidimensional temperature control data of the same battery cell can be summarized together. By using digital twin technology, two unidirectional transmission mapping channels are established between the virtual structural model and the temperature control dataset. A clock-synchronized timeline is set on both mapping channels to perform real-time fusion mapping between the temperature control dataset and the virtual structural model, thereby converting the virtual structural model into a digital battery model. The clock-synchronized timeline ensures that the two mapping channels remain synchronized in time, thus guaranteeing the synchronicity of the fusion mapping operation.

[0036] By constructing two unidirectional transmission mapping channels, on the one hand, real-time multidimensional temperature control data in the temperature control dataset can be dynamically mapped to the corresponding spatial coordinates of the virtual structure model to generate a digital battery model, realizing a visualization of "virtual" versus "real". On the other hand, the analysis results of the digital battery model can also be written back to the temperature control dataset to provide a theoretical basis for guiding the temperature control actuator of the physical entity. At the same time, the two independently distributed mapping channels also ensure the independence and stability of the "virtual" and "real" mapping operations, avoiding the phenomenon of mutual interference between the "virtual" and "real" mappings.

[0037] The parameter prediction module, based on the time period planning criteria, determines the target time period and predicts the temperature parameters of the digital battery model during the target time period through the time series parameter prediction model; The target time period is a specific time period used to determine whether the temperature change of the battery cell has caused temperature runaway.

[0038] Since the time-series parameter prediction model is used to predict the temperature of battery cells at future times, the target time period needs to be determined in conjunction with time period planning criteria. Specifically, the time-period planning criteria are as follows: the proportion of temperature rise is inversely proportional to the first weighting factor, and the proportion of temperature drop is inversely proportional to the second weighting factor.

[0039] The method for determining the target time period is as follows: The rated temperature range and optimal temperature value of the battery pack are retrieved through the battery management system, and the minimum and maximum values ​​of the rated temperature range are recorded as the lower limit and upper limit values, respectively. The temperature rise is calculated by subtracting the upper limit temperature value from the optimal temperature value, and the temperature drop is calculated by subtracting the lower limit temperature value from the optimal temperature value. The temperature range is calculated by subtracting the upper temperature limit from the lower temperature limit. The temperature drop and temperature rise are then divided by the temperature range to calculate the temperature drop percentage and temperature rise percentage, respectively. The average duration of a temperature drop and a temperature rise in the battery pack were retrieved from the database and denoted as the temperature drop duration and temperature rise duration, respectively. The temperature drop ratio is used as the second weighting factor, and the temperature rise ratio is used as the first weighting factor. The duration of the temperature drop and the duration of the temperature rise are weighted, summed, and averaged to calculate the duration of the time period. In this embodiment, the sum of the temperature drop ratio and the temperature rise ratio is 1. For example, when the temperature drop ratio is 0.6 and the temperature rise ratio is 0.4, the second weighting factor is 0.4 and the first weighting factor is 0.6, thereby ensuring that the temperature rise ratio is inversely proportional to the first weighting factor, and the temperature drop ratio is inversely proportional to the second weighting factor. The formula for calculating the duration of a time period is: ; In the formula, For the duration of the period, For the duration of temperature drop, For the duration of temperature rise, As the first weighting factor, As the second weighting factor; Starting from the current moment, the endpoint is marked after delaying the timeline by one time period, and the time period between the starting point and the endpoint is recorded as the target time period.

[0040] The target time period is a time span that includes multiple consecutive moments, and each specific moment has one and only one temperature parameter.

[0041] Digital battery models can only dynamically represent the real-time temperature changes of battery cells. Because they lack the specific temperature changes of battery cells at future moments, digital battery models cannot represent the specific state of battery cells at future moments.

[0042] To ensure that the digital battery model can fully and in advance identify the temperature changes of the battery cells in the future, it is necessary to first predict the temperature parameters of the battery cells in the future using a time-series parameter prediction model. In this embodiment, the time series parameter prediction model is an artificial intelligence model trained based on deep learning technology, combined with a large amount of multidimensional temperature control data and corresponding temperature parameters as training data. The time series parameter prediction model is a model obtained by fusing deep learning technology with time series model. Therefore, the time series parameter prediction model needs to be trained in advance. When training the time series parameter prediction model, a large amount of historical multidimensional temperature control data and corresponding temperature parameters of battery cells are collected in advance. A set of multi-source data and corresponding risk parameters are bound into data samples. The window length N and step size 1 of the sliding window are set. After traversing all the data, multiple data samples are generated. The data samples are divided into training samples and test samples in chronological order at a ratio of 80%:20%, ensuring that the time period of the test samples is after that of the training samples to avoid future data leakage. Set the model parameters of the deep learning model, including the learning rate, batch size, and maximum number of epochs. Use cross-validation to optimize the deep learning model and avoid overfitting. The deep learning model learns the temperature changes at each moment in the multidimensional temperature control data and outputs the corresponding temperature parameters; The deep learning model is trained using training samples and tested and validated using test samples until the identification accuracy of risk parameters in the test samples reaches a preset accuracy threshold, thus obtaining the required time series parameter prediction model.

[0043] Temperature parameters are parameters that can characterize the specific temperature changes of a battery cell over a certain period of time, and are used as input data to the digital battery model. Specifically, temperature parameters include peak temperature, trough temperature, rate of temperature change, and percentage of time the temperature remains stable.

[0044] Peak temperature refers to the highest temperature value that a battery cell can reach within a target time period, while valley temperature refers to the lowest temperature value that a battery cell can reach within a target time period.

[0045] Temperature change rate refers to the rate at which the temperature of a battery cell changes over a target period of time, and is used to express how quickly the temperature rises or falls within the battery cell.

[0046] The stabilization time percentage refers to the proportion of the time during which the temperature change of a battery cell remains within the equilibrium range at adjacent time points within the target time period, compared to the total duration of the target time period. The larger the stabilization time percentage, the longer the battery cell's temperature remains within the equilibrium range. The equilibrium range is a temperature range, for example, the temperature value of the equilibrium range is 35 degrees Celsius to 50 degrees Celsius.

[0047] By extracting multidimensional temperature control data from the digital battery model and inputting the multidimensional temperature control data into the time series parameter prediction model, the temperature parameters of the digital battery model at each moment within the target time period can be predicted.

[0048] The state simulation module imports the temperature parameters for the target time period into the digital battery model, sets up the simulation mechanism, and simulates the temperature control state of the battery cell during the target time period. Once the temperature parameters for the target time period are predicted, they can be imported into the digital battery model, allowing the temperature parameters to be integrated with the digital battery model for simulation, thus simulating the temperature control state of the battery cell during the target time period. Temperature control status is used to measure the overall temperature change of the battery cell within a target period, and also serves as the theoretical basis for whether the battery cell will undergo intelligent temperature control in the future. Specifically, the temperature control state includes a controllable temperature state, a temperature rise runaway state, and a temperature drop runaway state. Among them, the controllable temperature state means that the temperature change of the battery cell within the target period is safe and controllable, the temperature rise runaway state means that the temperature rise of the battery cell within the target period is uncontrollable, and the temperature drop runaway state means that the temperature drop of the battery cell within the target period is uncontrollable.

[0049] When simulating temperature control conditions, a simulation mechanism needs to be set up to ensure the orderliness and rationality of the digital battery model in the simulation process and to improve the accuracy of temperature control conditions. The simulation mechanism is as follows: simulation is performed by simulating a single time group as the standard; this allows for continuous and distributed simulation of various time periods within a long target time period, thereby improving the accuracy of temperature control simulation within the target time period.

[0050] The method for simulating temperature control is as follows: The dynamic duration is calculated by comparing the duration of the target period with the proportion of the stability maintenance period. The formula for calculating dynamic duration is: ; In the formula, For dynamic duration, For the proportion of time spent maintaining stability; Calculate the duration multiple by dividing the target time period by the dynamic time period; When the duration multiple is an integer, the target time period is divided into D time groups based on the duration span of the dynamic duration; When the duration multiple is not an integer, the duration of two consecutive moments is used as a reduction increment, and the dynamic duration is continuously reduced until the duration multiple of the reduced dynamic duration is an integer. Then, the target time period is divided into D time groups based on the duration span of the reduced dynamic duration. The temperature peak, temperature valley, temperature change rate, and stability duration percentage of all times within the D time groups are summarized to form D simulation data sets. The D simulation data sets are then imported into the digital battery model in chronological order to simulate the temperature values ​​of the digital battery model in the D time groups. Plot a temperature distribution map with time on the horizontal axis and temperature on the vertical axis, including the temperature distribution curve. Then, plot the lower boundary line corresponding to the lower limit value and the upper boundary line corresponding to the upper limit value on the temperature distribution map. Using the standard that one time group corresponds to one time curve, the temperature distribution curve is divided into D time curves, and the positional relationship between the D time curves and the lower and upper boundary lines is observed. When the time curve is entirely between the lower and upper boundary lines, it indicates that the temperature of the battery cell remains stable during the target time period, and the temperature control state is determined to be controllable. When the curve is below the lower boundary line for some time points and between the lower and upper boundary lines for other time points, it indicates that the temperature of the battery cell group is below the lower limit of the temperature at that time point, and the temperature control state is determined to be a cooling runaway state. When the curve is above the upper boundary line for some time points and between the lower boundary line and the upper boundary line for some time points, it indicates that the temperature of the battery cell group is higher than the upper temperature limit at that time point, and the temperature control state is determined to be a temperature runaway state. When the time curve is simultaneously above the upper boundary line, between the lower boundary line and the upper boundary line, and below the lower boundary line, the temperature value of the last time step in the time curve is analyzed. If the temperature value of the last time step is above the upper boundary line, the temperature control state is determined to be a temperature rise runaway state; if the temperature value of the last time step is below the lower boundary line, the temperature control state is determined to be a temperature drop runaway state; if the temperature value of the last time step is between the lower boundary line and the upper boundary line, the temperature control state is determined to be a controllable state.

[0051] The type of temperature control state of the battery cell during the target period is not fixed. There is a minimum of one and a maximum of three, which is determined by the temperature value of a specific time group in the target period. That is, a time group in the target period corresponds to only one temperature control state.

[0052] The temperature control execution module identifies the out-of-control units from the battery cells based on the temperature control status, arranges the out-of-control units in a time sequence, and formulates intelligent temperature control measures for the out-of-control units. A runaway unit refers to a battery cell whose temperature control status has changed significantly, making the runaway unit the direct target for subsequent intelligent temperature control operations. When identifying out-of-control units, the number of battery units in a controllable temperature state within the target time period is counted. The number of controllable states is divided by the number of states at all times within the target time period to calculate the controllable percentage. Battery units with a controllable percentage less than a preset controllable threshold are recorded as out-of-control units, resulting in E out-of-control units, where E is less than or equal to D.

[0053] The preset controllable threshold refers to the maximum controllable percentage when a battery cell is recorded as an out-of-control unit, which can be used as a numerical basis for judging whether a battery cell is controllable. The preset controllable threshold is obtained by collecting a large number of historical data on the maximum controllable percentage when a battery cell is recorded as an out-of-control unit and then averaging them. For example, the preset controllable threshold is 0.85.

[0054] When implementing intelligent temperature control for runaway units, it is necessary to formulate corresponding intelligent temperature control measures based on the specific temperature control status and the corresponding time sequence to ensure that the temperature of the runaway unit can remain stable during the target period, improve the temperature controllability of the battery unit in the future, avoid the phenomenon of temperature runaway of the battery pack in the future, and improve the charging and discharging safety of the battery pack in new energy vehicles. The method for formulating intelligent temperature control measures is as follows: Arrange the temperature control status of D time groups in E out-of-control units in chronological order to form a status queue. Based on the cell center coordinates of the cells in the runaway unit, a heating system that provides heating and temperature rise function and a cooling system that provides cooling and temperature drop function for the runaway unit are selected. When the temperature control status of the time group is in a state of uncontrolled temperature rise, the intelligent temperature controller formulates timing measures to control the cooling system to execute cooling and temperature reduction commands; when the temperature control status of the time group is in a state of uncontrolled temperature drop, the intelligent temperature controller formulates timing measures to control the heating system to execute heating and temperature rise commands. Traverse the temperature control states of D time groups in the state queue, generate D time measures, and then arrange and summarize the D time measures in sequence to formulate intelligent temperature control measures.

[0055] In this embodiment, the intelligent temperature control measure is used as a guide for the battery pack to perform intelligent temperature control operations in advance, thereby ensuring that the heating system and cooling system can perform temperature control operations of heating and cooling within the time period of the previous time group, avoiding the phenomenon of temperature runaway of the battery cells in the future, and thus achieving the intelligent temperature control effect of the battery pack.

[0056] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention 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 the present invention should be included within the scope of protection of the present invention.

Claims

1. An intelligent temperature control system for a new energy battery pack, applied to an intelligent temperature controller, characterized in that, include: The battery segmentation module is used to extract the structural and positional features of the battery pack, using structural features as the basis for morphological segmentation and positional features as the basis for spatial segmentation, to divide the battery pack into battery cells. The digital modeling module is used to acquire multi-dimensional temperature control data of battery cells, construct a virtual structural model based on structural characteristics, and fuse and map the virtual structural model with the multi-dimensional temperature control data to construct a digital battery model. The parameter prediction module is used to determine the target time period based on the time period planning criteria, and predict the temperature parameters of the digital battery model during the target time period through the time series parameter prediction model. The temperature parameters include temperature peak, temperature valley, temperature change rate and the proportion of stabilization time. The state simulation module is used to import the temperature parameters of the target time period into the digital battery model, set the simulation mechanism, and simulate the temperature control state of the battery cell during the target time period. The temperature control execution module is used to identify the runaway unit from the battery cells based on the temperature control status, perform timing analysis on the runaway unit, and formulate intelligent temperature control measures.

2. The intelligent temperature control system for a new energy battery pack according to claim 1, characterized in that, Structural features include geometric dimensions, cell arrangement, cell spacing, and cooling channel characteristics; Location features include the battery pack reference coordinates and the cell center coordinates.

3. The intelligent temperature control system for a new energy battery pack according to claim 2, characterized in that, The method for dividing battery cells is as follows: Cells with the same structural features within the battery pack are grouped into one structural family, resulting in B structural families. The Euclidean distance between the center coordinates of A cells and the reference coordinates of the battery pack is calculated one by one using the Euclidean distance formula. The Euclidean distance is then normalized to obtain A relative position weights. Using relative position weights as the dividing criterion, within the same structural group, the K-means clustering algorithm is used to divide the cells with the same weight level in B structural groups into a battery cell, resulting in C array-distributed battery cells.

4. The intelligent temperature control system for a new energy battery pack according to claim 3, characterized in that, Multidimensional temperature control data includes surface temperature, internal temperature, operating voltage, charging and discharging current, and internal resistance; The method for constructing a digital battery model is as follows: Based on the structural characteristics of a single battery cell, a three-dimensional battery cell model is constructed using three-dimensional modeling technology. Then, according to the actual spatial relationship of the battery cells, the three-dimensional battery cell models belonging to the same battery unit are sequentially spliced ​​together to generate a virtual structure model. Create a blank dataset on the battery cell, set the number of data slots in the blank dataset to match the number of battery cells, and fill the data slots in the blank dataset with the multidimensional temperature control data from the C battery cells one by one to generate the temperature control dataset. By using digital twin technology, two unidirectional transmission mapping channels are established between the virtual structural model and the temperature control dataset, and a clock-synchronized time axis is set on both mapping channels to fuse the virtual structural model into a digital battery model.

5. The intelligent temperature control system for a new energy battery pack according to claim 4, characterized in that, The time-period planning principle is as follows: the proportion of temperature rise is inversely proportional to the first weighting factor, and the proportion of temperature drop is inversely proportional to the second weighting factor.

6. The intelligent temperature control system for a new energy battery pack according to claim 5, characterized in that, The method for determining the target time period is as follows: The rated temperature range and optimal temperature value of the battery pack are queried through the battery management system. The minimum and maximum values ​​of the rated temperature range are recorded as the lower limit and upper limit of the temperature range, respectively. The temperature rise is calculated by subtracting the upper limit of the temperature from the optimal temperature value, and the temperature drop is calculated by subtracting the optimal temperature from the lower limit of the temperature. The temperature range is calculated by subtracting the upper temperature limit from the lower temperature limit. The temperature drop and temperature rise are then divided by the temperature range to calculate the temperature drop percentage and temperature rise percentage, respectively. The average duration of a temperature drop and a temperature rise in the battery pack were retrieved from the database and denoted as the temperature drop duration and temperature rise duration, respectively. The proportion of temperature drop is used as the second weighting factor, and the proportion of temperature rise is used as the first weighting factor. The duration of temperature drop and the duration of temperature rise are weighted, summed and averaged to calculate the duration of the period. Starting from the current moment, the endpoint is marked after delaying the timeline by one time period, and the time period between the starting point and the endpoint is recorded as the target time period.

7. The intelligent temperature control system for a new energy battery pack according to claim 6, characterized in that, Temperature control states include temperature controllable state, temperature rise runaway state, and temperature drop runaway state; The simulation mechanism is as follows: simulation is performed based on a single time group.

8. The intelligent temperature control system for a new energy battery pack according to claim 7, characterized in that, The method for simulating temperature control is as follows: The duration of the target period is compared with the proportion of the stability maintenance period to calculate the dynamic duration. The duration of the target period is then divided by the dynamic duration to calculate the duration multiple. When the duration multiple is an integer, the target time period is divided into D time groups based on the duration span of the dynamic duration; When the duration multiple is not an integer, the duration of two consecutive moments is used as a reduction increment, and the dynamic duration is continuously reduced until the duration multiple of the reduced dynamic duration is an integer. Then, the target time period is divided into D time groups based on the duration span of the reduced dynamic duration. The temperature peak, temperature valley, temperature change rate, and stability duration percentage of all times within the D time groups are summarized to form D simulation data sets. The D simulation data sets are then imported into the digital battery model in chronological order to simulate the temperature values ​​of the digital battery model in the D time groups. Plot a temperature distribution map containing temperature distribution curves with time as the horizontal axis and temperature as the vertical axis. Then, plot the lower boundary line corresponding to the lower limit value and the upper boundary line corresponding to the upper limit value in the temperature distribution map. Using the standard that one time group corresponds to one time curve, the temperature distribution curve is divided into D time curves, and the positional relationship between the D time curves and the lower and upper boundary lines is observed. When all time-matter curves are between the lower and upper boundary lines, the temperature control is in a controllable state; when some time-matter curves are below the lower boundary line and some time-matter curves are between the lower and upper boundary lines, the temperature control is in a cooling runaway state; when some time-matter curves are above the upper boundary line and some time-matter curves are between the lower and upper boundary lines, the temperature control is in a heating runaway state. When the time curve is simultaneously above the upper boundary line, between the lower boundary line and the upper boundary line, and below the lower boundary line, if the temperature value at the last time point is above the upper boundary line, the temperature control state is a temperature rise runaway state; if the temperature value at the last time point is below the lower boundary line, the temperature control state is a temperature drop runaway state; if the temperature value at the last time point is between the lower boundary line and the upper boundary line, the temperature control state is a controllable state.

9. The intelligent temperature control system for a new energy battery pack according to claim 8, characterized in that, When identifying out-of-control units, the number of battery units in a controllable temperature state within the target time period is counted. The number of controllable states is divided by the number of states at all times within the target time period to calculate the controllable percentage. Battery units with a controllable percentage less than a preset controllable threshold are recorded as out-of-control units, resulting in E out-of-control units.

10. The intelligent temperature control system for a new energy battery pack according to claim 9, characterized in that, The method for formulating intelligent temperature control measures is as follows: Arrange the temperature control status of D time groups in E out-of-control units in chronological order to form a status queue. Based on the cell center coordinates of the cells in the runaway unit, a heating system that provides heating and temperature rise function and a cooling system that provides cooling and temperature drop function for the runaway unit are selected. When the temperature control status of the time group is in a state of uncontrolled temperature rise, the intelligent temperature controller formulates timing measures to control the cooling system to execute cooling and temperature reduction commands; when the temperature control status of the time group is in a state of uncontrolled temperature drop, the intelligent temperature controller formulates timing measures to control the heating system to execute heating and temperature rise commands. Traverse the temperature control states of D time groups in the state queue, generate D time measures, and then arrange and summarize the D time measures in sequence to formulate intelligent temperature control measures.

Images