Battery thermal management optimization method and device

Abstract

The invention provides a battery thermal management optimization method and device. The method is applied to a collaboration platform, and the collaboration platform comprises a physical simulation platform, an intelligent optimization platform, an intelligent processing platform and an acquisition control platform. The method comprises the steps that a physical simulation platform takes preprocessed data as simulation parameters, battery thermoelectric coupling simulation is executed, and a simulation result of battery thermoelectric coupling simulation is obtained; the intelligent optimization platform adjusts simulation parameters according to the simulation result to obtain adjusted simulation parameters, and the intelligent optimization platform calls a physical simulation platform to perform one or more battery thermoelectric coupling iterative simulation according to the adjusted simulation parameters to obtain optimization parameters; the intelligent processing platform performs post-processing on the optimized parameters to obtain control parameters; and the acquisition control platform performs logic verification on the control parameters and transmits the control parameters to the test system to execute a battery charging test, so that the overall efficiency of battery thermal management optimization is improved.

Description

Technical Field

[0001] This application relates to the field of battery thermal management, and more specifically, to a battery thermal management optimization method and apparatus. Background Technology

[0002] With the rapid development of the new energy vehicle industry, the thermal management of batteries, as the core energy storage unit of new energy vehicles, has become a key focus of industry attention. During charging, the chemical reactions inside the battery generate a large amount of heat. If the temperature is not controlled in a timely and effective manner, it will affect the battery's charging efficiency, shorten its lifespan, and even cause safety hazards.

[0003] Solutions to battery thermal management problems often involve analyzing and optimizing data using modeling and simulation tools. For example, Simulink's physics-based modeling method accurately describes the internal heat conduction and convection processes of a battery, giving it an advantage in building thermoelectric coupling models. While this approach—using a physics simulation platform like Simulink to model and analyze the battery's thermoelectric coupling, followed by testing and verification—allows for adjusting thermal management optimization parameters based on simulation results, the overall efficiency of the optimization process is low because it still relies on repeated manual adjustments. Summary of the Invention

[0004] This application provides a method and apparatus for optimizing battery thermal management. The various aspects covered in this application are described below.

[0005] Firstly, this application provides a battery thermal management optimization method. This method is applied to a collaborative platform, which includes a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, and a data acquisition and control platform. The method includes: the physical simulation platform using preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain simulation results; the intelligent optimization platform adjusting the simulation parameters based on the simulation results to obtain adjusted simulation parameters; the intelligent optimization platform calling the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling based on the adjusted simulation parameters to obtain optimized parameters; the intelligent processing platform post-processing the optimized parameters to obtain control parameters; and the data acquisition and control platform performing logical verification of the control parameters and transmitting the control parameters to a test system to perform battery charging tests.

[0006] According to the first aspect, before the intelligent optimization platform calls the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling based on the adjusted simulation parameters, the intelligent optimization platform constructs an optimization objective function. The optimization objective function is used to determine whether the simulation results of one or more iterative simulations of battery thermoelectric coupling meet the convergence conditions.

[0007] According to the first aspect, or any implementation of the first aspect above, if the simulation results of the iterative simulation do not meet the convergence condition, the intelligent optimization platform continues to call the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling until the simulation results of one or more iterative simulations of battery thermoelectric coupling meet the convergence condition; if the simulation results of the iterative simulation meet the convergence condition, the intelligent optimization platform searches the simulation results of one or more iterative simulations of battery thermoelectric coupling to obtain the optimization parameters.

[0008] According to the first aspect, or any implementation of the first aspect above, before the physical simulation platform uses the preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain the simulation results of battery thermoelectric coupling simulation, the acquisition and control platform collects the test data in the battery charging test performed by the test system to obtain the raw test data; the intelligent processing platform preprocesses the raw test data to obtain the preprocessed data.

[0009] According to the first aspect, or any implementation of the first aspect above, the intelligent processing platform preprocesses the raw test data by one or more of the following: outlier removal, missing value filling, filtering, current boundary value solving, and requested water temperature boundary value solving.

[0010] According to the first aspect, or any implementation of the first aspect above, the data acquisition and control platform performs logical verification of the control parameters, including one or more of the following: numerical range check; rate of change limit; physical rationality verification.

[0011] According to the first aspect, or any implementation of the first aspect above, the collaborative platform also includes an intelligent management platform and a data storage platform. The method further includes: the data storage platform archives and stores one or more of the following according to the archiving and storage rules of the intelligent management platform: simulation parameters, simulation results, adjusted simulation parameters, optimized parameters, control parameters, raw test data, and preprocessed data.

[0012] According to the first aspect, or any implementation of the first aspect above, the archiving storage rules include one or more of the following: data storage is performed according to a three-level directory structure of experimental batch-time series-data type; the signal table records the collected raw values ​​and processed values; the optimization parameter table records the iteration parameters and boundary constraints; the model state table records the key information of the thermoelectric coupling model in the physical simulation platform; and the iteration details table records the decision log of the intelligent optimization platform.

[0013] According to the first aspect, or any of the above implementation methods of the first aspect, the acquisition and control platform is implemented through CANoe; one or more of the intelligent processing platform and intelligent optimization platform are implemented through Python; and the physical simulation platform is implemented through Simulink.

[0014] Secondly, this application provides a battery thermal management optimization device, which is located on a collaborative platform. The collaborative platform includes a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, and a data acquisition and control platform. The device includes: a physical simulation module, used to use preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain simulation results; an intelligent optimization module, used to adjust simulation parameters according to the simulation results and obtain adjusted simulation parameters; the intelligent optimization module calls the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling according to the adjusted simulation parameters to obtain optimized parameters; an intelligent processing module, used to post-process the optimized parameters to obtain control parameters; and a data acquisition and control module, used to perform logical verification of the control parameters and transmit the control parameters to the test system to perform battery charging tests.

[0015] Thirdly, this application provides a computer-readable storage medium storing program code for computer execution, the program code including any possible battery thermal management optimization method for performing the first aspect and any possible implementation of the first aspect.

[0016] The battery thermal management optimization method provided in this application can be applied to a collaborative platform. This collaborative platform can include a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, and a data acquisition and control platform. Through this collaborative platform, the physical simulation platform (e.g., Simulink) can intelligently iteratively solve the battery thermoelectric coupling model and achieve intelligent collaboration with the testing system, thus overcoming the shortcomings of the physical simulation platform in terms of intelligence and improving the overall efficiency of battery thermal management optimization. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a battery thermal management optimization method provided in an embodiment of this application.

[0018] Figure 2 This is an example flowchart of the overall process for optimizing battery thermal management, provided in an embodiment of this application.

[0019] Figure 3 This is a schematic diagram of a battery thermal management device provided in an embodiment of this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0021] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0022] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0023] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0024] With the rapid development of the new energy vehicle industry, the overall performance of the battery, as the core energy storage unit of new energy vehicles, directly affects the vehicle's range, performance, and safety. In particular, the balance between battery charging technology and temperature control has become a key focus of the industry. With the increasing popularity of electric vehicles and the promotion of related policies, the new energy vehicle industry has placed higher performance requirements on battery systems. During battery charging, intense electrochemical reactions occur inside the battery, leading to a rapid accumulation of heat. If the battery's heat dissipation capacity is insufficient, the temperature will rise rapidly, significantly reducing charging efficiency, prolonging charging time, and potentially accelerating electrode material aging, shortening the overall battery life, or even triggering thermal runaway, causing serious safety hazards such as fires. Therefore, how to maintain the battery's optimal operating temperature range while ensuring charging speed has become a crucial challenge in the design of battery thermal management systems.

[0025] The most commonly used batteries in new energy vehicles are lithium-ion batteries. Lithium-ion batteries primarily rely on the insertion and extraction of lithium ions between the positive and negative electrodes to achieve the interconversion of electrical and chemical energy. When a lithium-ion battery is charged, lithium ions in the positive electrode material are extracted from the positive electrode lattice under the influence of an electric field and move towards the negative electrode through the electrolyte. Once at the negative electrode, the lithium ions are inserted into the interlayer structure or pores of the negative electrode material. Simultaneously, an equal number of electrons flow from the positive electrode to the negative electrode through an external circuit, forming an electric current and storing electrical energy. When a lithium-ion battery is used, lithium ions in the negative electrode are extracted from the negative electrode material under the influence of an electric field and move towards the positive electrode through the electrolyte, inserting into the lattice of the positive electrode material, restoring the positive electrode material to its pre-charging state. At the same time, electrons flow from the negative electrode to the positive electrode through an external circuit, combining with the lithium ions embedded in the positive electrode, thus forming an electric current and releasing electrical energy. During the charging and discharging process of a lithium-ion battery, internal temperature changes affect the lithium-ion migration rate and the stability of the electrode materials. If the temperature is too high, the electrolyte is prone to decomposition, and the solid electrolyte interphase (SEI) film will rupture more rapidly, leading to irreversible capacity loss. If the temperature is too low, lithium-ion diffusion kinetics will be sluggish, charging efficiency will decrease, and lithium metal deposition may even occur, causing a short circuit risk. Therefore, thermal management strategies for lithium-ion batteries are crucial to ensuring their efficient and safe operation.

[0026] Currently, in the field of battery thermal management, modeling and analysis using physical simulation platforms has become an important method. Taking Simulink as an example, as a widely used modeling tool in engineering, Simulink's physics-based modeling methods can accurately describe the internal heat conduction and convection processes of a battery, giving it an advantage in constructing thermoelectric coupling models. Furthermore, Simulink provides various interfaces and tools for interacting with external data. For instance, it supports reading test data from common file formats (such as Excel, MATLAB's .mat files, etc.). In battery thermal management research, measured temperature, voltage, current, and other data of the battery under different operating conditions (such as different charge / discharge rates, ambient temperature, etc.) can be stored in these file formats. Then, the data can be imported into the Simulink model through corresponding modules, serving as simulation input or for comparison and verification with simulation results. It is not difficult to see that the battery thermal management optimization method, which uses physical simulation platforms such as Simulink to model and analyze the battery thermoelectric coupling and then combines it with a test system for testing and verification, can adjust the thermal management optimization parameters based on the simulation results. However, the efficiency of the entire battery thermal management optimization process is low because it still relies on manual adjustment of parameters repeatedly throughout the optimization process.

[0027] To address the aforementioned issues, this application introduces an intelligent platform to enable intelligent iterative solving of the battery thermoelectric coupling model by the physical simulation platform and intelligent collaboration with the testing system. This battery thermal management optimization method compensates for the shortcomings of the physical simulation platform in terms of intelligence, thereby improving the overall efficiency of the battery thermal management optimization method.

[0028] The following is combined with Figure 1 The embodiments of this application will be described in detail below.

[0029] Figure 1 This is a flowchart illustrating a battery thermal management optimization method provided in an embodiment of this application. This battery thermal management optimization method can be applied to a collaborative platform. It should be noted that the collaborative platform includes a multi-platform system that combines multiple platforms to achieve collaborative operation.

[0030] In some embodiments, the collaborative platform may include: a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, an acquisition and control platform, an intelligent management platform, and a data storage platform. As an example, the acquisition and control platform can be implemented using CANoe. One or more of the intelligent processing platform, intelligent optimization platform, and intelligent management platform can be implemented using Python. The physical simulation platform can be implemented using Simulink. The data storage platform can be implemented using ClickHouse.

[0031] In some embodiments, data interaction and workflow collaboration between platforms can be achieved through application programming interfaces (APIs). For example, Python and CANoe can interact and collaborate via the COM API, Python and Simulink can interact and collaborate via the MATLAB Engine API, and Python and ClickHouse can interact and collaborate via the Python-ClickHouse data interface. It should be noted that this is only an illustrative example of the interface types between platforms; other interface types that can be used to implement inter-platform communication should be included within the scope of this application.

[0032] Figure 1 The method shown includes steps S110 to S140.

[0033] In step S110, the physical simulation platform can use the preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain the simulation results of battery thermoelectric coupling simulation.

[0034] In some embodiments, before the physical simulation platform uses the preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation, the test data of the test system in performing battery charging test can be collected through the acquisition and control platform to obtain the raw test data.

[0035] In some embodiments, the raw test data may include one or more of the following: battery inlet water temperature, battery outlet water temperature, battery maximum temperature, battery minimum temperature, battery average temperature, current value, and state of charge (SOC) value.

[0036] As an example, the data acquisition and control platform can collect one or more of the following data during the battery charging process through a real-time communication link with the testing system: battery inlet water temperature, battery outlet water temperature, battery maximum temperature, battery minimum temperature, battery average temperature, current value, requested water temperature value, and SOC value. The collected test data can then be used as the raw test data.

[0037] In some embodiments, after the acquisition and control platform collects the raw test data, the raw test data can be transmitted to the intelligent processing platform, where it is preprocessed to obtain preprocessed data. For example, the test data can be systematically configured as variables to be transmitted in both the acquisition and control platform and the intelligent processing platform, including setting variable names and data types.

[0038] Table 1 shows a possible example of the systematic variable configuration for the data acquisition and control platform and the intelligent processing platform. In Table 1, variable names and data types can be defined for the test data. For example, the highest battery temperature can be defined as Tmax, with a double data type; the lowest battery temperature can be defined as Tmin, with a double data type; and the requested water temperature value can be defined as T_req, with an integer data type. It should be noted that the systematic variable configuration shown in Table 1 is only an example; test data not listed in Table 1 also requires variable names and data types to be defined.

[0039] Table 1 It is important to note that when defining variable names and data types for test data in the acquisition and control platform and the intelligent processing platform, the principle of maintaining consistency in naming across different platforms should be followed. This ensures accurate mapping and identification of test data when transmitted across different platforms.

[0040] In some embodiments, after completing the systematic configuration of variables, the intelligent processing platform can use specific functions to obtain the raw test data collected by the acquisition control platform and load the collected raw test data into the intelligent processing platform. For example, the intelligent processing platform can use the get_SysVar function to obtain the raw test data collected by the acquisition control platform and load it into the intelligent processing platform for subsequent preprocessing.

[0041] In some embodiments, after the intelligent processing platform acquires the raw test data, it can preprocess the raw test data to obtain preprocessed data.

[0042] In some embodiments, preprocessing may include one or more of the following: outlier removal, missing value filling, filtering, current boundary value solving, and requested water temperature boundary value solving.

[0043] For example, an intelligent processing platform can identify and remove outliers from the acquired raw test data based on an offline-trained outlier identification model.

[0044] For example, the intelligent processing platform can intelligently fill in missing values ​​in the acquired raw test data based on the missing value imputation model.

[0045] It should be noted that the outlier identification model or missing value imputation model here can be a machine learning model trained on historical data by an intelligent processing platform. This machine learning model can automatically identify and process abnormal or missing test data based on the data distribution characteristics.

[0046] For example, intelligent processing platforms can use filtering algorithms (such as the Hodrick-Prescott decomposition algorithm) to filter data such as water temperature in the original test data to obtain trend and cycle terms. Taking the battery inlet water temperature Mtc_temp_in at a certain sampling time (represented by t) as an example, the Hodrick-Prescott decomposition algorithm can be used to obtain the trend and cycle terms according to the formula. The battery inlet water temperature Mtc_temp_in is filtered. In the formula, t is the sampling time. The trend value after decomposition at time t. This represents the periodic term after decomposition at time t. Here, we retain... The value is the filtered value of the battery inlet water temperature Mtc_temp_in.

[0047] For example, when solving for current boundary values, the intelligent processing platform can use lookup functions and the SOC-Tmax table, which consists of the SOC value and battery temperature, for lookup. For instance, the maximum current value Imax can be looked up using a lookup function and by consulting the real-time SOC and the highest battery temperature in the SOC-Tmax table. It's important to note that this SOC-Tmax table can be a table generated from offline testing; it possesses high accuracy and reliability and can be directly used when solving for current boundary values.

[0048] For example, when solving for the boundary value of the requested water temperature, taking the sampling time t as an example, the intelligent processing platform can follow the formula... Calculate the maximum boundary value Treq_max and the minimum boundary value Treq_min of the requested water temperature. Where, in the formula... Let be the highest temperature of the battery at sampling time t. Let be the lowest temperature of the battery at sampling time t, 'a' be an adjustable value, and 'b' be the highest water temperature the battery can request. 'a' is not greater than 'b', and both 'a' and 'b' are positive numbers. It should be noted that the values ​​of 'a' and 'b' depend on the battery type and / or model. For example, given batteries A, B, and C, when solving for the requested water temperature boundary value for battery A, a = 30 and b = 53. For battery B, a = 20 and b = 60. For battery C, a = 30 and b = 60.

[0049] In some embodiments, after the original test data is preprocessed by the intelligent processing platform and the preprocessed data is obtained, the physical simulation platform can use the preprocessed data as simulation parameters to run the thermoelectric coupling simulation of the battery and obtain the simulation results of the battery thermoelectric coupling simulation.

[0050] As can be seen from the above, the embodiments of this application preprocess the original test data to ensure the validity of the data in the subsequent battery thermoelectric coupling simulation process, thereby improving the accuracy of the battery thermoelectric coupling simulation.

[0051] In step S120, the intelligent optimization platform adjusts the simulation parameters according to the simulation results to obtain the adjusted simulation parameters, and calls the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling according to the adjusted simulation parameters to obtain the optimized parameters.

[0052] In some embodiments, before the intelligent optimization platform executes step S120, the intelligent optimization platform can construct an optimization objective function based on the preprocessed data. As an example, the intelligent optimization platform can construct the optimization objective function based on the current boundary value and the requested water temperature boundary value, and its expression is: In the formula, J represents the optimization target value. The magnitude of the optimization target value reflects the degree of optimization of the simulation results. For example, in this embodiment, a larger optimization target value indicates a better degree of optimization of the simulation results.

[0053] In some embodiments, the intelligent optimization platform can establish a bidirectional data channel with the physical simulation platform by setting up corresponding API interfaces. Taking the intelligent optimization platform implemented in Python and the physical simulation platform implemented in Simulink as an example, a bidirectional data channel between Python and Simulink can be established through the MATLAB Engine API, thereby realizing data interaction between Python and Simulink.

[0054] In some embodiments, the intelligent optimization platform can convert optimization parameters into a format recognizable by the battery thermoelectric coupling model of the physical simulation platform and transmit them through the corresponding API interface.

[0055] In some embodiments, the intelligent optimization platform can define one or more extraction rules from the simulation results obtained by the physical simulation platform through iterative simulation based on the optimization objective function, and set the parameters of the fixed-step solver and the simulation duration in the physical simulation platform. Taking the intelligent optimization platform implemented in Python and the physical simulation platform implemented in Simulink as an example, Python can define extraction rules for the SOC value, the maximum battery temperature Tmax, and the minimum battery temperature Tmin in the iterative simulation results of Simulink based on the optimization objective function, and set the parameters of the fixed-step solver and the simulation duration of the Simulink model, thereby ensuring that Simulink maintains both computational accuracy and meets real-time requirements during simulation.

[0056] In some embodiments, the optimization objective function can be used to determine whether the simulation results of one or more battery thermoelectric coupling iterative simulations meet the convergence condition. For example, after performing one or more battery thermoelectric coupling iterative simulations on a physical simulation platform, the intelligent optimization platform can extract the SOC value, the battery maximum temperature Tmax, and the battery minimum temperature Tmin from the simulation results of the iterative simulation, and combine the optimization objective function and intelligent optimization algorithm (such as the improved NSGA-II multi-objective genetic algorithm) to determine whether the simulation results meet the convergence condition.

[0057] In some embodiments, the convergence condition may include one or more of the following: the change in the target value is less than a specific threshold D; the change in one or more of the SOC value, the maximum battery temperature Tmax, and the minimum battery temperature Tmin in the simulation results is less than a specific threshold H in one or more iterations; and the number of iterations for the iterative simulation reaches a set iteration number value E. Here, D and H can both be positive numbers, and E can be a positive integer.

[0058] For example, taking a threshold D=0.001 as an example, if the change in the optimization objective value J is less than 0.001 after one or more battery thermoelectric coupling iterative simulations, then the simulation results after one or more battery thermoelectric coupling iterative simulations are considered to meet the convergence condition.

[0059] For example, taking a threshold H=0.01 as an example, if the change of one or more of the SOC value, the maximum battery temperature Tmax and the minimum battery temperature Tmin in the simulation results is less than 0.01 in one or more iterations after one or more battery thermoelectric coupling iterative simulations, then the simulation results after one or more battery thermoelectric coupling iterative simulations are considered to meet the convergence condition.

[0060] For example, taking the iteration number E=200 as an example, after the number of iterations of the battery thermoelectric coupling iterative simulation reaches 200, if one or more of the optimization target value and / or the SOC value, the battery maximum temperature Tmax and the battery minimum temperature Tmin in the simulation result are less than their corresponding specific thresholds, then the simulation result after 200 battery thermoelectric coupling iterative simulations is considered to meet the convergence condition.

[0061] In some embodiments, if the simulation results of the iterative simulation do not meet the convergence condition, the intelligent optimization platform can continue to call the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling until the simulation results of one or more iterative simulations of battery thermoelectric coupling meet the convergence condition. For example, when the simulation results of one or more iterative simulations do not meet the convergence condition, the intelligent optimization platform can dynamically adjust the simulation parameters according to the intelligent optimization algorithm (e.g., the improved NSGA-II multi-objective genetic algorithm) and resubmit the simulation solution, calling the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling until the simulation results of one or more iterative simulations of battery thermoelectric coupling meet the convergence condition.

[0062] In some embodiments, when the simulation results of one or more battery thermoelectric coupling iterative simulations meet the convergence condition, the intelligent optimization platform searches the simulation results of one or more battery thermoelectric coupling iterative simulations to obtain optimized parameters. For example, when the simulation results of one or more battery thermoelectric coupling iterative simulations meet the convergence condition, the intelligent optimization platform can calculate the optimization target value based on the SOC value, the battery maximum temperature Tmax, and the battery minimum temperature Tmin in the simulation results of one or more battery thermoelectric coupling iterative simulations, and dynamically adjust the search strategy in conjunction with an intelligent optimization algorithm (e.g., an improved NSGA-II multi-objective genetic algorithm) to shrink the parameter search range in real time until the optimized parameters are obtained.

[0063] In some embodiments, the optimization parameters include the current optimization value I_req_op and the requested water temperature optimization value T_req_op.

[0064] As can be seen from the above, through step S120, the embodiments of this application can realize that the physical simulation platform can perform simulation iteration through intelligent algorithms, eliminating the need for manual adjustment of simulation parameters, realizing intelligent optimization of parameters by the physical simulation platform, and improving the efficiency of simulation iteration.

[0065] In step S130, the intelligent processing platform performs post-processing on the optimization parameters to obtain the control parameters.

[0066] In some embodiments, after obtaining the optimization parameters through the intelligent optimization platform, the intelligent processing platform can encode the optimization parameters into a format recognizable by the acquisition and control platform, and transmit the data through the API interface between them. For example, the intelligent optimization platform can convert the current optimization value I_req_op to a Double type, convert the requested water temperature optimization value T_req_op to an Int type, and pass them to the set_SysVar function. The acquisition and control platform can read the set_SysVar function transmitted through the API interface to obtain the control parameters.

[0067] Step S140: The acquisition and control platform performs logical verification of the control parameters and transmits the control parameters to the test system to perform battery charging test.

[0068] In some embodiments, after the acquisition and control platform obtains the control parameters, it can perform logical verification on the control parameters. Logical verification includes one or more of the following: numerical range checking; rate of change limitation; and physical rationality verification. For example, the acquisition and control platform can check the numerical range of the current optimization value I_req_op (e.g., I_req_op needs to be below 790A). As another example, the acquisition and control platform can limit the rate of change of the current optimization value I_req_op (e.g., the change amplitude of I_req_op must be ≤100A). Furthermore, the acquisition and control platform can perform rationality verification on the battery's SOC value (e.g., charging needs to be stopped when SOC > 97%).

[0069] As can be seen from the above, through steps S130 to S140, the embodiments of this application can perform post-processing and logical verification of the optimized parameters, which effectively improves the efficiency and reliability of transmitting simulation results to the actual test environment.

[0070] In some embodiments, the acquisition and control platform performs one or more logical verifications on the control parameters, including numerical range checks, rate of change limits, and physical rationality verifications. After the logical verifications are passed, the acquisition and control platform can transmit the control parameters to the test system to perform battery charging tests.

[0071] In some embodiments, the intelligent management platform can formulate data archiving and storage rules. For example, the intelligent management platform can formulate archiving and storage rules for one or more of the simulation parameters, simulation results, adjusted simulation parameters, optimized parameters, control parameters, raw test data, and preprocessed data involved in the embodiments of this application.

[0072] In some embodiments, the archiving and storage rules established by the intelligent management platform may include one or more of the following: data storage according to a three-level directory structure of experimental batch-time series-data type; a signal table recording the collected raw values ​​and processed values; an optimization parameter table recording the iteration parameters and boundary constraints; a model state table recording key information of the thermoelectric coupling model in the physical simulation platform; and an iteration details table recording the decision log of the intelligent optimization platform. Specifically, the signal table may adopt a time-series database structure to record the collected raw values ​​and processed values; the optimization parameter table may adopt a wide table model to record the iteration parameters and their boundary constraints; the model state table may adopt versioned storage to record key information of the thermoelectric coupling model in the physical simulation platform; and the iteration details table may adopt an event flow model to record the algorithm's decision log.

[0073] In some embodiments, the data storage platform can archive and store one or more of the following data according to the archiving and storage rules defined by the intelligent management platform: simulation parameters, simulation results, adjusted simulation parameters, optimized parameters, control parameters, raw test data, and preprocessed data. For example, the data storage platform can obtain the data archiving and storage rules through an API interface with the intelligent management platform, classify the data according to these rules, and select target tables for data writing. Based on this archiving and storage mechanism, the integrity and traceability of the data can be ensured, facilitating subsequent data querying and result analysis and statistics.

[0074] In some embodiments, the data storage platform can record all transmission anomaly events through an intelligent management platform to provide subsequent tracing and analysis capabilities for these events. For example, it can quickly statistically analyze the distribution of anomaly events and locate faults.

[0075] As can be seen from the above, through the intelligent management platform and data parameter platform, the embodiments of this application can effectively solve the problems of local limitations and insufficient sharing in data storage of physical simulation platforms. This improves the intelligence of data storage, facilitating subsequent on-demand retrieval and sharing of data.

[0076] To facilitate understanding, the following examples (Examples 1-4) illustrate the battery thermal management optimization method described in this application, using examples 1-4, where the intelligent processing platform is implemented using Python; the acquisition and control platform is implemented using CANoe; one or more of the intelligent processing platform, intelligent optimization platform, and intelligent management platform are implemented using Python; the physical simulation platform is implemented using Simulink; and the data storage platform is implemented using ClickHouse.

[0077] Example 1 Example 1 illustrates the process by which CANoe collects raw test data and preprocesses it using Python.

[0078] In Example 1, CANoe acquires raw test data in real time during battery charging tests. This raw test data may include one or more of the following: battery maximum temperature, battery minimum temperature, battery average temperature, battery inlet temperature, battery outlet temperature, current value, and SOC value. A sliding window analysis module is then built using Python to perform outlier identification, missing value imputation, and filtering smoothing on the acquired raw test data. Finally, Python is used to optimize the boundary conditions of the processed data. The process in Example 1 is explained in detail through steps 1.1 to 1.6.

[0079] Step 1.1: Establish a real-time communication link with the battery test in CANoe and systematically configure variables, including setting variable names and data types. Create a Python group and define the above variables. For example, the battery's highest temperature is Tmax (Double type); the battery's lowest temperature is Tmin (Double type); the battery's inlet temperature is Mtc_temp_in (Double type); the battery's outlet temperature is Mtc_temp_out (Double type); the current value is I_req (Double type); and the requested water temperature value is T_req (Int type).

[0080] Step 1.2: Define the CANoe project in the Python environment. The variable names and data types should be consistent with those in Step 1.1. When interacting with variables, follow the principle of name consistency.

[0081] Step 1.3: Start CANoe using Python, retrieve the data from the Python group variables set in Step 1.1 in CANoe using the get_SysVar function, and then load them into the Python environment.

[0082] Step 1.4: Load the outlier detection and missing value imputation models that have been trained offline using Python, and perform real-time online processing on the parameter signals obtained in Step 1.3. Both the outlier detection and missing value imputation models are machine learning models trained on historical data.

[0083] Step 1.5: Filter the water temperature from Step 1.3. Using the Hodrick-Prescott decomposition algorithm, obtain the trend and cycle terms. Taking the battery inlet temperature Mtc_temp_in at sampling time t as an example, the filtering formula is as follows: In the formula, t is the sampling time. The trend value after decomposition at time t. This represents the periodic term after decomposition at time t. During the filtering process of the battery inlet temperature Mtc_temp_in, only the periodic term is retained. The value is used as the filtered battery inlet water temperature.

[0084] Step 1.6: Using the data processed in Steps 1.4 and 1.5, further solve for the current boundary values ​​and the requested water temperature boundary values ​​in the Python environment. For the current boundary values, the maximum current value Imax is retrieved from the SOC-Tmax table using a lookup function, the real-time SOC value, and the battery's highest temperature. The SOC-Tmax table can be a table completed offline and can be used directly. The minimum current value Imin = 0. The formulas for calculating the maximum and minimum boundary values ​​of the requested water temperature, Treq_max and Treq_min, are as follows: Example 2 Example 2 illustrates the process of intelligent iterative simulation using Simulink and Python based on the preprocessed data from Example 1.

[0085] In Example 2, Python can provide initial parameter boundaries and an optimization objective function based on the preprocessed data from Example 1, and then drive the Simulink thermoelectric coupling model for iterative solution via the MATLAB Engine API. Each time Simulink is called to complete the temperature field and SOC calculations for the current parameter combination, the optimization algorithm on the Python side can dynamically adjust the parameters based on the model output and resubmit the solution, iterating in this loop until the convergence condition is met, thus achieving dynamic parameter optimization. The process in Example 2 is explained in detail through steps 2.1 to 2.3.

[0086] Step 2.1: Construct the basic structure of the optimization framework in the Python environment, and define the multi-objective optimization function J, with the following expression: In the formula, J represents the optimization target value, and the larger the J value, the better the optimization of the simulation results.

[0087] Step 2.2: Establish a bidirectional data channel between Python and Simulink through the MATLAB Engine API, convert the optimization variables of Python into signals that can be recognized by the Simulink model, define the extraction rules for extracting SOC values, Tmax and Tmin from the Simulink simulation results, set the fixed step size solver parameters and simulation duration, and ensure that the model solution guarantees both accuracy and real-time requirements.

[0088] Step 2.3: Execute the intelligent optimization iteration process in Simulink using Python. Submit a set of parameter combinations generated in each iteration to Simulink for simulation. Calculate the optimization target value J based on the SOC, Tmax, and Tmin data extracted from the Simulink simulation results. Then, use an improved NSGA-II multi-objective genetic algorithm to dynamically adjust the search strategy, narrowing the parameter search range in real time to improve the efficiency of later optimization stages. Finally, obtain the optimized parameters: the current optimization value I_req_op and the requested water temperature optimization value T_req_op.

[0089] Example 3 Example 3 illustrates the process by which CANoe and Python obtain optimized parameters based on the intelligent iterative simulation in Example 2, then post-process the parameters and transmit them to the test system for testing.

[0090] In Example 3, Python encodes the optimization parameters obtained in Example 2 into a format recognizable by CANoe to obtain control parameters. Then, it establishes a real-time data channel with CANoe via the COM API to transmit these control parameters. CANoe can perform logical verification on the control parameters and send the verified parameters to the actual test system for execution. The process in Example 3 is explained in detail in steps 3.1 to 3.2.

[0091] Step 3.1: In the Python environment, process the current optimization value I_req_op and the requested water temperature optimization value T_req_op into a standardized format. Specifically, convert the current optimization value I_req_op to a Double type and the requested water temperature optimization value T_req_op to an Int type. Then, pass the current optimization value I_req_op and the requested water temperature optimization value T_req_op to the set_SysVar function to generate control parameters.

[0092] Step 3.2: Establish a data channel between Python and CANoe via the COM API interface, and transfer the set_SysVar function to CANoe. CANoe obtains the control parameters through the set_SysVar function. CANoe performs logical verification on the control parameters, including numerical range checks (e.g., the optimized current value I_req_op is below 790A), rate of change limits (e.g., the adjustment range of the optimized current value I_req_op is ≤100A), and physical rationality verification (e.g., stop charging when SOC > 97%). If the verification is successful, the verified control parameters are sent to the actual test system for execution, enabling real-time dynamic control of the test system.

[0093] Example 4 Example 4 illustrates the process by which ClickHouse's Python-based intelligent algorithm archives and stores one or more of the data from Examples 1-3 above.

[0094] In Example 4, Python can use an asynchronous write interface to persistently store the raw test data and preprocessed data (e.g., optimization boundary values) from Example 1; intermediate variables, detailed information of iterative simulations, objective function values, and optimization parameters from Simulink simulations in Example 2; and control parameters and one or more data points from abnormal events in Example 3, according to a three-level directory structure of experimental batch-time series-data type, to ClickHouse, achieving end-to-end data traceability and sharing. The process of Example 4 is explained in detail through steps 4.1 to 4.3.

[0095] Step 4.1: Define the data archiving and storage rules in Python. Specifically, the signal table uses a time-series database structure to record the raw and processed values; the optimization parameter table uses a wide table model to record the iteration parameters and boundary constraints; the model state table uses versioned storage to record key information about the Simulink model; and the iteration details table uses an event stream model to record the algorithm's decision log.

[0096] Step 4.2: Establish a data channel between Python and ClickHouse through the Python-ClickHouse data interface. ClickHouse obtains the archiving and storage rules specified by Python, classifies the data according to the archiving and storage rules, automatically selects the target table according to the range of the data, and performs the writing process.

[0097] Step 4.3: Python records all transmission exception events in ClickHouse in real time, and uses Structured Query Language (SQL) to quickly analyze the distribution of exception events and locate faults.

[0098] To make it easier to understand, the following will be explained... Figure 2 The battery thermal management optimization method described in this application is illustrated by taking as an example one of the following: the intelligent processing platform is implemented in Python, the acquisition and control platform is implemented in CANoe, one or more of the intelligent processing platform, intelligent optimization platform, and intelligent management platform are implemented in Python, the physical simulation platform is implemented in Simulink, and the data storage platform can be implemented in ClickHouse.

[0099] Figure 2 This is an example flowchart of an overall process for a battery thermal management optimization method provided in an embodiment of this application.

[0100] Figure 2 The method shown may include steps S201-S205.

[0101] In step S201, Python can obtain the raw test data collected in CANoe through the COM API.

[0102] After collecting the raw test data, Python can preprocess the raw test data using intelligent processing algorithms to obtain preprocessed data.

[0103] In step S202, Python can use the MATLAB Engine API interface to transmit the preprocessed data as simulation parameters to Simulink and call Simulink to perform iterative simulation of battery thermoelectric coupling.

[0104] In step S203, after completing one or more iterative simulations of battery thermoelectric coupling in Simulink, the optimized parameters can be transferred to Python via the MATLAB Engine API interface.

[0105] It is important to note that steps S202 and S203 enable the interaction between simulation parameters and the Simulink battery thermoelectric coupling model and the Python intelligent optimization algorithm. The thermoelectric coupling model built in Simulink can provide optimization results for solving the objective function value, while Simulink can receive simulation parameters from Python. The intelligent optimization algorithm deployed in Python can dynamically adjust the simulation parameters based on the iterative simulation results from Simulink and call Simulink again to perform one or more iterative simulations of battery thermoelectric coupling until optimized parameters that meet the convergence conditions are found.

[0106] In step S204, Python can post-process the optimization parameters using an intelligent processing algorithm to obtain control parameters, and then transmit the control parameters to CANoe via the COM API interface.

[0107] After receiving the control parameters, CANoe can perform logical verification on the control parameters and transmit the control parameters to the test system to perform battery charging tests.

[0108] In step S205, Python can use the Python-ClickHouse data channel interface to transfer all data from steps S201 to S204 to ClickHouse for data storage.

[0109] ClickHouse provides real-time storage support for the battery thermal management optimization process and receives iterative parameters, intermediate calculation results, and optimization logs from Python.

[0110] Optionally, Figure 2 It may also include step S206, whereby Python can access and / or analyze (historical analysis) the data stored in ClickHouse at any time through the Python-ClickHouse data channel interface.

[0111] The inventors of this application discovered that, based on Examples 1 to 4 and... Figure 2 By leveraging Python as the intelligent hub, this approach deeply integrates the model prediction accuracy of Simulink, the real-time execution capabilities of CANoe, and the real-time data storage capabilities of ClickHouse. This creates a closed loop that balances the accuracy of the physical model, the adaptability of the intelligent algorithm, the real-time nature of the hardware interface, and the shared nature of data storage. It overcomes the limitations of single platforms while ensuring the accuracy of battery temperature and SOC predictions during charging, as well as the control response speed of the test system, through the complementary capabilities of each platform. This makes the battery thermal management optimization process more efficient and intelligent, improving overall optimization efficiency and reliability.

[0112] The method embodiments of this application have been described in detail above. The apparatus embodiments of this application are described in detail below. It should be understood that the descriptions of the method embodiments correspond to the descriptions of the apparatus embodiments. Therefore, any parts not described in detail can be referred to the foregoing method embodiments.

[0113] Figure 3 This is a schematic diagram of a battery thermal management device provided in an embodiment of this application. Figure 3 The device shown is located on the collaborative platform. The collaborative platform includes: a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, and a data acquisition and control platform.

[0114] Figure 3 The battery thermal management optimization device 300 shown includes: a physical simulation module 310, an intelligent optimization module 320, an intelligent processing module 330, and a data acquisition and control module 340.

[0115] The physical simulation module 310 is used to use the preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain the simulation results of battery thermoelectric coupling simulation.

[0116] The intelligent optimization module 320 is used to adjust the simulation parameters according to the simulation results, obtain the adjusted simulation parameters, and call the physical simulation platform to perform one or more iterative simulations of battery thermoelectric coupling based on the adjusted simulation parameters to obtain the optimized parameters.

[0117] The intelligent processing module 330 is used to post-process the optimization parameters to obtain the control parameters.

[0118] The acquisition and control module 340 performs logical verification on the control parameters and transmits the control parameters to the test system to perform battery charging tests.

[0119] Furthermore, this application also proposes a computer-readable storage medium storing a computer program. When the computer program is executed by a computer, it implements the operations in the battery thermal management optimization method provided in the above embodiments. The specific steps will not be described in detail here.

[0120] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity / operation / object from another, and do not necessarily require or imply any such actual relationship or order between these entities / operations / objects; the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0121] In the embodiments of this application, the term "correspondence" can indicate a direct or indirect correspondence between two things, or an association between two things, or a relationship such as instruction and being instructed, configuration and being configured.

[0122] In the embodiments of this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0123] In the embodiments of this application, "comprising" can refer to direct inclusion or indirect inclusion. Optionally, "comprising" mentioned in the embodiments of this application can be replaced with "indicating" or "used to determine". For example, "A includes B" can be replaced with "A indicates B" or "A is used to determine B".

[0124] In the various embodiments of this application, the order of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0125] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0126] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0127] For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and relevant details can be found in the description of the method embodiments. The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separate. Some or all of the modules can be selected according to actual needs to achieve the purpose of this application. Those skilled in the art can understand and implement this without creative effort.

[0128] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can read or a data storage device such as a server or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0129] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0130] The above are merely embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A battery thermal management optimization method, characterized in that, The method is applied to a collaborative platform, which includes: a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, and a data acquisition and control platform. The method includes: The physical simulation platform uses the preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain the simulation results of the battery thermoelectric coupling simulation. The intelligent optimization platform adjusts the simulation parameters based on the simulation results to obtain the adjusted simulation parameters. The intelligent optimization platform then calls the physical simulation platform to perform one or more iterative simulations of the battery thermoelectric coupling based on the adjusted simulation parameters to obtain the optimized parameters. The intelligent processing platform performs post-processing on the optimized parameters to obtain control parameters; The acquisition and control platform performs logical verification on the control parameters and transmits the control parameters to the test system to perform battery charging tests.

2. The method according to claim 1, characterized in that, Before the intelligent optimization platform calls the physical simulation platform to perform one or more iterative simulations of the battery thermoelectric coupling based on the adjusted simulation parameters to obtain the optimized parameters, the method includes: The intelligent optimization platform constructs an optimization objective function, which is used to determine whether the simulation results of one or more battery thermoelectric coupling iterative simulations meet the convergence condition.

3. The method according to claim 2, characterized in that, The method further includes: If the simulation results of the iterative simulation do not meet the convergence condition, the intelligent optimization platform continues to call the physical simulation platform to perform one or more iterative simulations of the battery thermoelectric coupling until the simulation results of one or more iterative simulations of the battery thermoelectric coupling meet the convergence condition. If the simulation results of the iterative simulation meet the convergence condition, the intelligent optimization platform searches the simulation results of one or more battery thermoelectric coupling iterative simulations to obtain optimization parameters.

4. The method according to claim 1, characterized in that, Before the physical simulation platform uses the preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain the simulation results, the method includes: The data acquisition and control platform collects test data from the battery charging test performed by the test system to obtain raw test data. The intelligent processing platform preprocesses the original test data to obtain the preprocessed data.

5. The method according to claim 4, characterized in that, The intelligent processing platform preprocesses the raw test data by one or more of the following: Outlier removal, missing value filling, filtering, current boundary value solving, and request water temperature boundary value solving.

6. The method according to claim 1, characterized in that, The data acquisition and control platform performs logical verification of the control parameters, including one or more of the following: Numerical range check; rate of change limit; physical rationality verification.

7. The method according to claim 4, characterized in that, The collaborative platform also includes an intelligent management platform and a data storage platform. The method further includes: The data storage platform archives and stores one or more of the following according to the archiving and storage rules of the intelligent management platform: the simulation parameters, the simulation results, the adjusted simulation parameters, the optimized parameters, the control parameters, the original test data, and the preprocessed data.

8. The method according to claim 7, characterized in that, The archive storage rules include one or more of the following: Data is stored according to a three-level directory structure of experimental batch-time series-data type; the signal table records the collected raw values ​​and processed values; the optimization parameter table records the iteration parameters and boundary constraints; and the model state table records the key information of the thermoelectric coupling model in the physical simulation platform. The iteration details table records the decision logs of the intelligent optimization platform.

9. The method according to claim 1, characterized in that, The acquisition and control platform is implemented using CANoe; one or more of the intelligent processing platform and the intelligent optimization platform are implemented using Python; and the physical simulation platform is implemented using Simulink.

10. A battery thermal management optimization device, characterized in that, The device is located on a collaborative platform, which includes: a physical simulation platform, an intelligent optimization platform, an intelligent processing platform, and a data acquisition and control platform. The device includes: The physical simulation module is used to use preprocessed data as simulation parameters to perform battery thermoelectric coupling simulation and obtain the simulation results of the battery thermoelectric coupling simulation. The intelligent optimization module is used to adjust the simulation parameters according to the simulation results to obtain the adjusted simulation parameters; the intelligent optimization module calls the physical simulation platform to perform one or more iterative simulations of the battery thermoelectric coupling according to the adjusted simulation parameters to obtain the optimized parameters; The intelligent processing module is used to post-process the optimized parameters to obtain control parameters; The acquisition and control module is used to perform logical verification on the control parameters and transmit the control parameters to the test system to perform battery charging tests.

Images