Battery heat dissipation decision-making method and system based on meta-learning

Abstract

The invention relates to the technical field of battery heat dissipation decisions, and provides a battery heat dissipation decision method and system based on meta-learning. The method comprises the following steps: performing clustering division based on historical charge-discharge rate records of a first battery pack to obtain a plurality of preset charge-discharge rates; obtaining a plurality of battery temperature sequence sets; obtaining a temperature distribution deviation coefficient; generating contact area information; the activation element learner performs heat dissipation parameter fitting on the optimized first battery heat dissipation device to obtain a heat dissipation parameter decision model; real data are input into the heat dissipation parameter decision model for heat dissipation parameter fitting, and real-time heat dissipation parameters are output; and controlling the first battery heat dissipation device according to the real-time heat dissipation parameters. The technical problem that in the prior art, battery heat dissipation treatment is difficult under different charging multiplying powers is solved, and the technical effects that the battery temperature is effectively managed, the heat dissipation device is intelligently regulated and controlled, safe operation of the battery is ensured, and the service life of the battery is prolonged are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of battery heat dissipation decision-making, and in particular to a battery heat dissipation decision-making method and system based on meta-learning. Background Art

[0002] Nowadays, the use of batteries is increasing, but the battery pack cannot dissipate heat in time, resulting in the battery pack system temperature being too high or unevenly distributed. As a result, the battery charge and discharge cycle efficiency will be reduced, affecting the power and energy of the battery. In severe cases, it will also lead to thermal runaway, affecting the safety and reliability of the system. In addition, due to the dense placement of the heat-generating battery bodies, more heat will inevitably accumulate in the middle area, and less heat will accumulate in the edge area, which increases the temperature imbalance between the units in the battery pack, which will cause uneven performance of the battery modules and cells, and ultimately affect the consistency of battery performance and the accuracy of battery charge state estimation.

[0003] In summary, the prior art has the technical problem of difficulty in heat dissipation of batteries at different charging rates. Summary of the invention

[0004] Based on this, it is necessary to provide a meta-learning-based battery heat dissipation decision method and system that can effectively manage the battery temperature and intelligently regulate the heat dissipation device to ensure the safe operation of the battery and extend its service life in response to the above technical problems.

[0005] In a first aspect, the present application provides a battery heat dissipation decision method based on meta-learning, the method comprising: obtaining a plurality of preset charge and discharge rates based on the historical charge and discharge rate records of a first battery group and performing clustering division; traversing the plurality of preset charge and discharge rates to collect historical charge and discharge temperatures of M battery cells of the first battery group to obtain a plurality of battery temperature sequence sets; traversing the plurality of battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis to obtain M temperature distribution deviation coefficients; analyzing the contact area between the flat heat pipe in the first battery heat dissipation device and the M battery cells using the M temperature distribution deviation coefficients to generate M contact area information; optimizing the first battery heat dissipation device using the M contact area information, activating the meta-learner to perform heat dissipation parameter fitting on the optimized first battery heat dissipation device to obtain a heat dissipation parameter decision model; collecting the real-time charge and discharge rate and the real-time ambient temperature of the first battery group and inputting them into the heat dissipation parameter decision model to perform heat dissipation parameter fitting and output real-time heat dissipation parameters; controlling the first battery heat dissipation device using the real-time heat dissipation parameters.

[0006] In a second aspect, the present application provides a battery heat dissipation decision system based on meta-learning, the system comprising: a preset charge and discharge rate acquisition module, the preset charge and discharge rate acquisition module is used to obtain multiple preset charge and discharge rates based on the historical charge and discharge rate records of the first battery group and perform clustering; a battery temperature sequence set acquisition module, the battery temperature sequence set acquisition module is used to traverse the multiple preset charge and discharge rates to collect historical charge and discharge temperatures of M battery cells of the first battery group, and obtain multiple battery temperature sequence sets; a temperature distribution deviation coefficient acquisition module, the temperature distribution deviation coefficient acquisition module is used to traverse the multiple battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis, and obtain M temperature distribution deviation coefficients; a contact area information generation module, the contact area information generation module is used to collect historical charge and discharge temperatures of M battery cells of the first battery group based on the M preset charge and discharge rates, and obtain multiple battery temperature sequence sets; The temperature distribution deviation coefficient analyzes the contact area between the flat heat pipe in the first battery heat dissipation device and the M battery cells to generate M contact area information; a heat dissipation parameter decision model acquisition module, the heat dissipation parameter decision model acquisition module is used to optimize the first battery heat dissipation device with the M contact area information, activate the meta-learner to fit the heat dissipation parameters of the optimized first battery heat dissipation device, and obtain a heat dissipation parameter decision model; a real-time heat dissipation parameter output module, the real-time heat dissipation parameter output module is used to collect the real-time charge and discharge rate and real-time ambient temperature of the first battery group and input them into the heat dissipation parameter decision model for heat dissipation parameter fitting, and output real-time heat dissipation parameters; a battery heat dissipation device control module, the battery heat dissipation device control module is used to control the first battery heat dissipation device with the real-time heat dissipation parameters.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] First, based on the historical charge and discharge rate records of the first battery group and clustering division, multiple preset charge and discharge rates are obtained; then the historical charge and discharge temperatures of the M battery cells of the first battery group are collected through the multiple preset charge and discharge rates to obtain multiple battery temperature sequence sets; then the charge and discharge temperature distribution deviation analysis is performed through the multiple battery temperature sequence sets to obtain M temperature distribution deviation coefficients; next, the contact area between the flat heat pipe and the M battery cells in the first battery heat dissipation device is analyzed with the M temperature distribution deviation coefficients to generate M contact area information; after optimizing the first battery heat dissipation device with the M contact area information, the meta-learner is activated to fit the heat dissipation parameters of the optimized first battery heat dissipation device to obtain a heat dissipation parameter decision model; then the real-time charge and discharge rate and real-time ambient temperature of the first battery group are collected and input into the heat dissipation parameter decision model to fit the heat dissipation parameters, and the real-time heat dissipation parameters are output; finally, the first battery heat dissipation device is controlled with the real-time heat dissipation parameters. This application solves the technical problem of the difficulty of heat dissipation processing of batteries under different charge rates in the prior art, achieves the technical effect of effective management of battery temperature and intelligent regulation of heat dissipation devices, ensuring the safe operation of batteries and extending their service life.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is a flow chart of a battery heat dissipation decision method based on meta-learning in one embodiment;

[0011] Figure 2 A schematic diagram of a process for obtaining a temperature distribution value of a battery heat dissipation decision method based on meta-learning in one embodiment;

[0012] Figure 3 4 is a structural block diagram of a battery heat dissipation decision system based on meta-learning in one embodiment.

[0013] Explanation of the reference numerals: preset charge and discharge rate acquisition module 11, battery temperature sequence set acquisition module 12, temperature distribution deviation coefficient acquisition module 13, contact area information generation module 14, heat dissipation parameter decision model acquisition module 15, real-time heat dissipation parameter output module 16, battery heat dissipation device control module 17. DETAILED DESCRIPTION

[0014] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0015] like Figure 1 As shown, the present application provides a battery heat dissipation decision method based on meta-learning, the method comprising:

[0016] Based on the historical charge and discharge rate records of the first battery group and clustering, a plurality of preset charge and discharge rates are obtained;

[0017] Meta-learning is a machine learning method that aims to enable machine learning models to quickly adapt to new tasks. Through meta-learning, the model can learn how to learn, and quickly adapt and show good performance when encountering new tasks. It can be applied to various fields to help machines quickly adapt to new tasks and improve performance; uneven temperature distribution in the battery pack will reduce performance consistency, and excessive temperature differences will cause different battery cells to age at different rates. To ensure the balance of each battery cell during control, the heat dissipation device is optimized after temperature distribution analysis of different battery cells at different charge and discharge rates, and then heat dissipation control is performed to enable the battery to quickly adapt to the heat dissipation requirements under different working conditions, improve the efficiency of the battery heat dissipation system, and extend the battery life.

[0018] The first battery group refers to any one selected from the research battery groups for research, recorded as the first battery group. First, it is necessary to collect the historical charge and discharge rate records of the first battery group. These data can come from the battery management system (BMS) or other related monitoring systems. The collected data is cleaned and sorted, including data standardization, removal of outliers, processing of missing values, etc., which can improve the quality of the data for subsequent clustering analysis; clustering algorithms (such as K-means, hierarchical clustering, etc.) are used to classify the historical charge and discharge rates. The purpose of clustering is to group similar charge and discharge rates into a group so that the rates in the same group are similar. The clustering algorithm can better understand the charge and discharge behavior patterns of the battery. According to the results of clustering, one or more representative charge and discharge rates are selected from each cluster as the preset charge and discharge rates. These preset rates represent the charge and discharge behavior patterns of the first battery pack, wherein the preset charge and discharge rates mean that the battery pack will perform charge and discharge at different rates in historical operation. Different charge and discharge rates will result in different degrees of battery temperature increase. The preset charge and discharge rates include rates with different numerical values. Through the above method, multiple preset charge and discharge rates are obtained. These preset rates can be used as a reference to guide battery charge and discharge operations, which will help to better manage and optimize battery performance and help improve battery life, safety and efficiency.

[0019] Traversing the plurality of preset charge and discharge rates to collect historical charge and discharge temperatures of the M battery cells of the first battery group to obtain a plurality of battery temperature sequence sets;

[0020] According to the preset charge and discharge ratio, traverse each preset ratio, where one preset charge and discharge ratio corresponds to M temperature sequences of M cells; for each preset charge and discharge ratio, use a temperature sensor or other related equipment to collect historical charge and discharge temperatures of the M battery cells of the first battery group. Ensure that the collected temperature data is accurate and reliable, and organize the collected temperature data into a sequence set, with each battery cell corresponding to a temperature sequence. Ensure that the data is arranged in chronological order and process any missing or abnormal temperature data; obtain multiple battery temperature sequence sets, each set corresponding to a preset charge and discharge ratio. Obtain multiple battery temperature sequence sets, these data help to better understand the temperature variation law of the battery at different charge and discharge ratios, thereby optimizing the battery's charge and discharge strategy and improving the battery's safety, life, and efficiency.

[0021] Traversing the plurality of battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis to obtain M temperature distribution deviation coefficients;

[0022] The multiple battery temperature sequence sets are obtained, and for each battery temperature sequence set, the temperature distribution deviation is analyzed at different charge and discharge rates. Statistical methods can be used to calculate the mean, variance and other statistics of each sequence to evaluate the stability of the temperature distribution; the temperature distribution deviation coefficient of each battery cell is calculated. This coefficient can reflect the temperature consistency of the battery during the charge and discharge process. The smaller the deviation coefficient, the more uniform the temperature distribution; conversely, it indicates that there is a large deviation in the temperature distribution; the temperature distribution deviation coefficients of the M battery cells are organized into a table or graphic form for subsequent analysis and comparison. M temperature distribution deviation coefficients are obtained, and these data provide a basis for further optimizing the charge and discharge management of the battery pack. By adjusting and managing these parameters, the efficiency, life and safety of the battery pack can be improved.

[0023] Extracting a first battery temperature sequence set according to the multiple battery temperature sequence sets, wherein the first battery temperature sequence set includes temperature sequences of M battery cells;

[0024] Performing temperature balance analysis on the temperature sequences of the M battery cells in sequence to obtain M temperature balance indicators;

[0025] Comparing the M temperature balance indicators with a predetermined temperature balance indicator, and calculating a temperature distribution value based on the comparison result to obtain M temperature distribution values;

[0026] Calculating a ratio of the M temperature distribution values ​​to the sum of the M temperature distribution values ​​to obtain a first initial deviation coefficient set;

[0027] Similarly, the multiple battery temperature sequence sets are traversed to obtain multiple first initial deviation coefficient sets;

[0028] Based on a plurality of preset charge and discharge rates, weighted calculations are performed on the initial deviation coefficients of the M battery cells in the plurality of first initial deviation coefficient sets in sequence to obtain the M temperature distribution deviation coefficients.

[0029] From multiple battery temperature sequence sets, a temperature sequence set of the first battery is screened out. This set should contain temperature sequence data of M battery cells; the screened temperature sequence data is sorted to ensure the integrity and accuracy of the data. Outliers can be removed, missing values ​​can be processed, etc. to obtain more reliable analysis results; a temperature sequence set of the first battery is obtained, which includes temperature sequence data of M battery cells. The temperature sequence of each battery cell in the first battery temperature sequence set is balanced and analyzed, and statistical methods, frequency domain analysis or other related technologies can be used to evaluate the balance of temperature distribution, wherein temperature balance analysis refers to the stability, variance, standard deviation, etc. of temperature changes. Based on the results of the balance analysis, the temperature balance index of each battery cell is calculated. These indicators can reflect the degree of temperature balance of the battery cell at different charge and discharge rates. According to the actual application requirements and battery performance standards, one or more predetermined temperature balance indicators are set as reference standards, the temperature balance index of each battery cell is compared with the predetermined temperature balance index, and the temperature distribution value is calculated based on the comparison result to obtain M temperature distribution values. Calculate the ratio of the temperature distribution value of each battery cell to the sum of the temperature distribution values ​​of all battery cells to obtain the initial deviation coefficient of each battery cell, and organize the initial deviation coefficients of M battery cells into a set to obtain the first initial deviation coefficient set. This set contains the initial deviation coefficients of M battery cells, and so on, traverse the multiple battery temperature sequence sets. Calculate the first initial deviation coefficient set under the preset charge and discharge rate to obtain the first initial deviation coefficient set, the first initial deviation coefficient set corresponds to a preset charge and discharge rate, and assign corresponding weights to each preset charge and discharge rate according to actual needs and data analysis results. The weight allocation should take into account the degree of influence of the charge and discharge rate on the battery performance; according to the result of weight allocation, perform weighted calculation on the initial deviation coefficients in each first initial deviation coefficient set, organize the weighted deviation coefficients of M battery cells, and analyze and evaluate them. These data reflect the temperature distribution deviation of the battery under different charge and discharge rates. Through continuous optimization and management, the efficiency, life and safety of the battery pack can be improved.

[0030] like Figure 2As shown, P temperature sequences whose temperature balance index is less than or equal to the predetermined temperature balance index are extracted, and the temperature mean is calculated to obtain P temperature balance values, where P is an integer greater than 0 and less than or equal to M;

[0031] Extract Q temperature sequences whose temperature balance index is greater than a predetermined temperature balance index, perform temperature concentration value analysis, and obtain Q temperature concentration values, where Q+P=M;

[0032] The P temperature balance values ​​and the Q temperature concentration values ​​are integrated to obtain the M temperature distribution values ​​corresponding to the first battery temperature sequence set.

[0033] From multiple battery temperature sequence sets, temperature sequences that meet the conditions are screened out. The temperature balance index of these sequences should be less than or equal to the predetermined temperature balance index, where the predetermined temperature balance index is set by the staff. The screened temperature sequence data is sorted to ensure the integrity and accuracy of the data. The temperature mean of the screened P temperature sequences is calculated. A mathematical model or statistical method can be used to calculate the mean of each sequence to obtain P temperature balance values, where P is an integer greater than 0 and less than or equal to M. From multiple battery temperature sequence sets, temperature sequences that meet the conditions are screened out. The temperature balance index of these sequences should be greater than the predetermined temperature balance index. The screened temperature sequence data is sorted out, and the temperature concentration value of the screened Q temperature sequences is calculated and analyzed to obtain the temperature concentration degree of each sequence, where Q+P=M. The P temperature balance values ​​and the Q temperature concentration values ​​are integrated and analyzed to obtain the M temperature distribution values ​​corresponding to the first battery temperature sequence set, that is, the specific heating situation in the first battery sequence set is obtained. The obtained temperature distribution value can be used to guide the optimization of the battery charging and discharging strategy and improve the efficiency, life and safety of the battery pack.

[0034] Extracting a first temperature sequence from the Q temperature sequences, and updating the sequence in descending order of temperature to obtain a first updated sequence;

[0035] Dividing the first update sequence into equal distances to obtain a plurality of subsequences;

[0036] Extracting multiple data distribution quantities of the multiple subsequences, and performing a data total amount difference analysis between adjacent subsequences to obtain data difference quantities of the multiple adjacent subsequences;

[0037] Extracting a subsequence whose data difference amount meets a predetermined threshold and whose data distribution quantity meets a predetermined data amount, and performing temperature mean calculation to obtain a first temperature concentration value;

[0038] In this way, the Q temperature sequences are traversed to obtain the Q temperature concentration values.

[0039] From the Q temperature sequences, select any one temperature sequence as the first temperature sequence, sort the temperatures in the first temperature sequence from large to small through a sorting operation, update the first temperature sequence, obtain the first updated sequence, and divide the first updated sequence into equal distances according to the actual needs and the accuracy requirements of data analysis. Determine the location of the division and the number of subsequences, and divide the first updated sequence into multiple subsequences according to the setting of equal distance division, each subsequence contains a certain number of data points; by counting and analyzing the data in each subsequence, extract the data distribution quantity in each subsequence, and perform a difference analysis on the total amount of data between adjacent subsequences, that is, calculate the difference, ratio or other relevant indicators between adjacent subsequences to evaluate the degree of difference between them, find multiple subsequences with similar continuous data volume, and determine the concentration value. If the data volume of both sequences is large, but there are many other sequences between the two, it does not meet the requirements to ensure data accuracy. The data difference amount that meets a predetermined threshold and the subsequences that meet a predetermined data distribution number are screened out from multiple adjacent subsequences, and the temperature mean is calculated for the screened subsequences to obtain a first temperature concentration value; according to the above method, the same operation is performed on each temperature sequence in the Q temperature sequences to obtain a temperature concentration value for each temperature sequence, and the obtained Q temperature concentration values ​​are analyzed and evaluated accordingly, which can improve the efficiency, life and safety of the battery pack.

[0040] Analyzing the contact areas between the flat heat pipe in the first battery heat dissipation device and the M battery cells using the M temperature distribution deviation coefficients to generate M contact area information;

[0041] Correlate the temperature distribution deviation coefficient with the contact area between the battery cell and the flat heat pipe. Understand how the temperature distribution deviation coefficient affects the heat dissipation performance of the battery, and the correlation between the heat dissipation performance of the battery and the contact area, and analyze the contact area between each battery cell and the flat heat pipe based on the above correlation. The impact of the temperature distribution deviation coefficient and its demand for heat dissipation effect can be taken into account during the analysis; based on the analysis results, the contact area between each battery cell and the flat heat pipe is calculated, and the design of the first battery heat dissipation device can be further optimized based on the obtained contact area information. For example, adjust the structure of the flat heat pipe, improve the design of the heat dissipation fins, adjust the contact method between the heat pipe and the battery cell, etc., to improve the heat dissipation effect and the overall performance of the battery pack.

[0042] After optimizing the first battery heat dissipation device using the M contact area information, activating the meta-learner to perform heat dissipation parameter fitting on the optimized first battery heat dissipation device to obtain a heat dissipation parameter decision model;

[0043] According to the M contact area information, the first battery heat dissipation device is optimized. This may involve adjusting the structure of the heat pipe, improving the design of the heat dissipation fins, adjusting the contact mode between the heat pipe and the battery cell, etc. The goal of the optimization is to improve the heat dissipation performance according to the contact area information, improve the efficiency, life and safety of the battery pack, evaluate the performance of the optimized first battery heat dissipation device; according to the evaluation results, adjust the parameters of the optimized heat dissipation device. This may include adjusting the structural parameters of the heat pipe, the design parameters of the fins, etc. to achieve the best heat dissipation effect. Select a suitable meta-learner, such as a neural network, a support vector machine, etc., to fit the heat dissipation parameters of the optimized first battery heat dissipation device, use the meta-learner to fit the data to establish a heat dissipation parameter decision model, use a machine learning algorithm to train the meta-learner so that it can predict the heat dissipation performance based on the input contact area information and other parameters; evaluate the performance of the obtained heat dissipation parameter decision model. The prediction accuracy and generalization ability of the model can be evaluated by cross-validation, test set evaluation or other methods. The first battery heat dissipation device is optimized based on the M contact area information, and a heat dissipation parameter decision model is established using a meta-learner. The model can predict the heat dissipation performance based on the input contact area information and other parameters, providing guidance for further optimization of the battery heat dissipation device.

[0044] Collecting the type of battery surface material of the first battery group;

[0045] Performing heat conduction analysis based on the battery surface material type and the flat heat pipe to generate a heat conduction coefficient;

[0046] Based on the matching of the thermal conductivity coefficient, a matching test material is obtained;

[0047] Establishing multiple sets of reference ambient temperatures based on the operating environment of the first battery pack;

[0048] With the goal of making the surface stable temperature of the matching test material lower than the calibrated stable temperature of the first battery pack, collecting multiple sets of historical battery operating temperatures, testing the first battery heat dissipation device in combination with the multiple sets of reference ambient temperatures, and recording multiple sets of heat dissipation control parameters;

[0049] The meta-learner is trained and fitted using the multiple groups of historical battery operating temperatures, the multiple groups of reference ambient temperatures, and the multiple groups of heat dissipation control parameters to obtain the heat dissipation parameter decision model.

[0050] The surface material type of each battery cell in the first battery pack is collected through experiments or measurement methods, including identification, classification and recording of battery surface materials; heat conduction analysis is performed based on the known battery surface material type and flat heat pipe material properties, and the thermal conductivity coefficient between the battery surface material and the flat heat pipe is calculated based on the results of the heat conduction analysis. The thermal conductivity coefficient is a scalar value that represents the thermal conductivity between the two materials; a matching strategy is formulated based on the known thermal conductivity coefficient and battery surface material type (such as material properties, heat transfer requirements, etc.), and related parameters are adjusted. An optimal or suitable material is determined through experimental verification, simulation analysis and matching strategies; a detailed analysis of the operating environment of the first battery pack is performed to understand its temperature variation range under different conditions; and multiple sets of reference ambient temperatures are established based on the results of the operating environment analysis. These temperature values ​​should be representative of various temperature conditions that the first battery pack may encounter in actual operation; the test goal is set to ensure that the surface stable temperature of the matching test material is less than the calibrated stable temperature of the first battery pack, ensuring that the battery operates within a safe and stable temperature range; collect historical operating temperature data of the first battery pack under various conditions; combine the collected historical temperature data with multiple sets of reference ambient temperatures to test the first battery heat dissipation device. Various heat dissipation control parameters, such as fan speed, coolant flow, etc., should be recorded during the test; during the test, the values ​​of various heat dissipation control parameters are recorded in detail. The parameters are very important for evaluating the performance of the heat dissipation device; multiple sets of historical battery operating temperatures, multiple sets of reference ambient temperatures, and heat dissipation control parameters are integrated to form a comprehensive data set. These data are used to train the meta-learner and fit the heat dissipation parameter decision model; the meta-learner is trained using the integrated data set. Select a suitable meta-learner model (such as a neural network, a support vector machine, etc.) and train it using the training data to fit the heat dissipation parameter decision model. Evaluate and optimize the trained heat dissipation parameter decision model. The prediction accuracy and generalization ability of the model are tested through cross-validation, test set evaluation or other methods, and the model is adjusted and improved based on the evaluation results; the best heat dissipation effect and battery performance optimization are achieved by real-time monitoring of battery operating temperature and ambient temperature, and adjusting heat dissipation control parameters.

[0051] After controlling the first battery heat dissipation device with the real-time heat dissipation parameter, collecting M controlled stable temperatures of the M battery cells in a preset monitoring window;

[0052] Determining whether the M controlled stable temperatures are less than or equal to the calibrated stable temperature;

[0053] If not, a control update instruction is generated to iteratively update and adjust the heat dissipation parameters of the first battery heat dissipation device in unit steps within a predetermined adjustment range until the total adjustment step reaches the critical value of the predetermined adjustment range or the M control stable temperatures are less than or equal to the calibrated stable temperature.

[0054] According to the heat dissipation parameters collected in real time, the first battery heat dissipation device is controlled in real time, such as adjusting the heat dissipation control parameters such as fan speed and coolant flow rate to achieve the best heat dissipation effect. During the control process, the temperature changes of the battery cells are continuously monitored to evaluate the control effect of the heat dissipation device, such as by comparing the control stable temperature with the calibrated stable temperature; a preset monitoring window is set to collect the temperature data of the battery cells. This window should be able to reflect the stable temperature state of the battery over a period of time. Within the monitoring window, the temperature data of M battery cells are collected. The temperature data of each battery cell should be able to reflect its control stable state; the M control stable temperatures are compared with the calibrated stable temperatures. It is judged whether these temperature values ​​meet the preset safety and performance requirements, and based on the comparison results, it is judged whether the temperature is within an acceptable range. If all the control stable temperatures are greater than the calibrated stable temperature, the heat dissipation device needs to be adjusted. If the judgment result is no, a control update instruction is generated. These instructions are used to guide the heat dissipation device to adjust parameters. Within the predetermined adjustment range, the heat dissipation parameters of the first battery heat dissipation device are iteratively updated and adjusted in unit steps. The adjustment amount of the parameters is gradually increased, and only a small range is adjusted each time to ensure the stability of the adjustment process. During the parameter adjustment process, the temperature changes of the M battery cells are continuously monitored to evaluate the control effect of the heat dissipation device. When the total adjustment step reaches the critical value of the predetermined adjustment range, or all controlled stable temperatures are less than or equal to the calibrated stable temperature, the parameter adjustment process is terminated. At this time, the parameters of the heat dissipation device have reached the optimal or acceptable range. The adjusted heat dissipation device parameters will provide a reference for subsequent optimization and maintenance.

[0055] If the total adjustment step reaches the critical value of the predetermined adjustment range, and the M controlled stable temperatures are greater than the calibrated stable temperatures, temperature adjustment abnormality information is generated;

[0056] The first battery pack and the first battery heat sink are inspected and repaired for abnormalities based on the temperature regulation abnormality information.

[0057] During the adjustment process, if the total adjustment step reaches the critical value of the predetermined adjustment range, but the M control stable temperatures are still greater than the calibrated stable temperature, it is judged that the temperature adjustment is abnormal, and the temperature adjustment abnormality information is generated based on the abnormal detection result. This information should include the time when the abnormality occurred, the specific description of the abnormality, the battery cells and heat sink parameters involved, such as heat sink failure, battery temperature is too high, etc.; the temperature adjustment abnormality information is notified to the relevant personnel and recorded. The record should include the specific situation of the abnormality, the treatment measures and results, etc., for subsequent analysis and processing. According to the temperature adjustment abnormality information, the first battery pack and the first battery heat sink are analyzed in detail to determine the cause of the abnormality. It is obtained by inspection and analysis of hardware equipment, control systems, environmental conditions, etc. According to the abnormal analysis results, a corresponding maintenance plan is formulated. The plan should include the parts that need to be inspected and replaced, the maintenance steps and time schedule, etc.; the first battery pack and the first battery heat sink are repaired according to the maintenance plan.

[0058] Collecting the real-time charge and discharge rate and the real-time ambient temperature of the first battery pack and inputting them into the heat dissipation parameter decision model to perform heat dissipation parameter fitting, and outputting the real-time heat dissipation parameters;

[0059] The charge and discharge rate and ambient temperature of the first battery pack are collected in real time through sensors or monitoring systems. These data will be used as input parameters for the analysis and fitting of heat dissipation parameters; the collected real-time charge and discharge rate and real-time ambient temperature are input into the heat dissipation parameter decision model; the heat dissipation parameter decision model fits the heat dissipation parameters through fitting calculations based on the input real-time data. After the heat dissipation parameters are fitted, the model will output real-time heat dissipation parameters. These parameters may include but are not limited to heat dissipation efficiency, heat flux density, temperature distribution, etc. They can reflect the performance of the battery heat dissipation device under current conditions. Applying the output real-time heat dissipation parameters to actual batteries is of great significance for optimizing the performance of the battery heat dissipation device and improving the safety and stability of the battery pack.

[0060] The first battery heat dissipation device is controlled according to the real-time heat dissipation parameter.

[0061] The current heat dissipation parameters of the first battery heat dissipation device are obtained in real time, and the output of the heat dissipation parameter decision model is obtained. Based on the heat dissipation parameters obtained in real time, a control decision is made. According to the result of the control decision, the first battery heat dissipation device is adjusted in real time to ensure that the battery temperature remains within the range of safety and performance optimization; during the control process, the temperature change of the battery is continuously monitored to evaluate the control effect of the heat dissipation device. This is achieved by comparing the controlled stable temperature with the calibrated stable temperature. Controlling the first battery heat dissipation device with the real-time heat dissipation parameters can achieve effective management of the battery temperature and intelligent regulation of the heat dissipation device, ensuring the safe operation of the battery and extending its service life. The present application solves the technical problem of the difficulty of heat dissipation processing of batteries at different charging rates in the prior art, and achieves the technical effect of effective management of the battery temperature and intelligent regulation of the heat dissipation device to ensure the safe operation of the battery and extend its service life.

[0062] like Figure 3 As shown, a battery heat dissipation decision system based on meta-learning includes:

[0063] A preset charge and discharge rate acquisition module 11, wherein the preset charge and discharge rate acquisition module 11 is used to obtain a plurality of preset charge and discharge rates based on the historical charge and discharge rate records of the first battery group and clustering division;

[0064] A battery temperature sequence set acquisition module 12, the battery temperature sequence set acquisition module 12 is used to collect historical charge and discharge temperatures of the M battery cells of the first battery group through the multiple preset charge and discharge rates to obtain multiple battery temperature sequence sets;

[0065] A temperature distribution deviation coefficient obtaining module 13, wherein the temperature distribution deviation coefficient obtaining module 13 is used to traverse the plurality of battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis to obtain M temperature distribution deviation coefficients;

[0066] A contact area information generating module 14, the contact area information generating module 14 is used to analyze the contact area between the flat heat pipe in the first battery heat dissipation device and the M battery cells by using the M temperature distribution deviation coefficients to generate M contact area information;

[0067] a heat dissipation parameter decision model acquisition module 15, wherein the heat dissipation parameter decision model acquisition module 15 is used to optimize the first battery heat dissipation device with the M contact area information, activate the meta-learner to perform heat dissipation parameter fitting on the optimized first battery heat dissipation device, and obtain a heat dissipation parameter decision model;

[0068] A real-time heat dissipation parameter output module 16, which is used to collect the real-time charge and discharge rate and the real-time ambient temperature of the first battery pack and input them into the heat dissipation parameter decision model to perform heat dissipation parameter fitting and output the real-time heat dissipation parameters;

[0069] The battery cooling device control module 17 is used to control the first battery cooling device according to the real-time cooling parameters.

[0070] Furthermore, the embodiment of the present application also includes:

[0071] a battery temperature sequence set extraction module, the battery temperature sequence set extraction module being used to extract a first battery temperature sequence set according to the multiple battery temperature sequence sets, the first battery temperature sequence set comprising temperature sequences of M battery cells;

[0072] A temperature balance index obtaining module, wherein the temperature balance index obtaining module is used to perform temperature balance analysis on the temperature sequences of the M battery cells in sequence to obtain M temperature balance indexes;

[0073] a temperature distribution value obtaining module, the temperature distribution value obtaining module being used to compare the M temperature balance indicators with a predetermined temperature balance indicator, and to calculate the temperature distribution value based on the comparison result to obtain M temperature distribution values;

[0074] a ratio calculation module, the ratio calculation module being used to calculate a ratio of the M temperature distribution values ​​to a sum of the M temperature distribution values ​​to obtain a first initial deviation coefficient set;

[0075] A battery temperature sequence set traversal module, the battery temperature sequence set traversal module is used to traverse the multiple battery temperature sequence sets by analogy to obtain multiple first initial deviation coefficient sets;

[0076] The deviation coefficient weighted calculation module is used to perform weighted calculation on the initial deviation coefficients of the M battery cells in the plurality of first initial deviation coefficient sets in sequence based on a plurality of preset charge and discharge rates to obtain the M temperature distribution deviation coefficients.

[0077] Furthermore, the embodiment of the present application also includes:

[0078] A temperature balance value acquisition module, the temperature balance value acquisition module is used to extract P temperature sequences whose temperature balance index is less than or equal to a predetermined temperature balance index, perform temperature mean calculation, and obtain P temperature balance values, where P is an integer greater than 0 and less than or equal to M;

[0079] A temperature concentration value analysis module, wherein the temperature concentration value analysis module is used to extract Q temperature sequences whose temperature balance index is greater than a predetermined temperature balance index, perform temperature concentration value analysis, and obtain Q temperature concentration values, wherein Q+P=M;

[0080] A temperature distribution value acquisition module is used to integrate the P temperature balance values ​​and the Q temperature concentration values ​​to obtain the M temperature distribution values ​​corresponding to the first battery temperature sequence set.

[0081] Furthermore, the embodiment of the present application also includes:

[0082] An update sequence acquisition module, the update sequence acquisition module is used to extract the first temperature sequence from the Q temperature sequences, and update the sequence in descending order of temperature to obtain a first update sequence;

[0083] A subsequence obtaining module, the subsequence obtaining module is used to divide the first update sequence into equal distances to obtain a plurality of subsequences;

[0084] An adjacent subsequence data difference amount obtaining module, the adjacent subsequence data difference amount obtaining module is used to extract multiple data distribution quantities of the multiple subsequences, and perform a total data amount difference analysis between adjacent subsequences to obtain data difference amounts of multiple adjacent subsequences;

[0085] A temperature mean value calculation module, the temperature mean value calculation module is used to extract the subsequences whose data difference amount meets the predetermined threshold and whose data distribution quantity meets the predetermined data quantity, perform temperature mean value calculation, and obtain a first temperature concentration value;

[0086] A temperature concentration value obtaining module is used to traverse the Q temperature sequences and obtain the Q temperature concentration values.

[0087] Furthermore, the embodiment of the present application also includes:

[0088] A battery surface material type collection module, the battery surface material type collection module is used to collect the battery surface material type of the first battery pack;

[0089] A heat conductivity coefficient generation module, the heat conductivity coefficient generation module is used to perform heat conduction analysis based on the battery surface material type and the flat heat pipe to generate a heat conductivity coefficient;

[0090] A matching test material obtaining module, the matching test material obtaining module is used to obtain matching test materials based on the matching of the thermal conductivity coefficient;

[0091] a reference ambient temperature establishing module, the reference ambient temperature establishing module being used to establish a plurality of groups of reference ambient temperatures based on the operating environment of the first battery pack;

[0092] a heat dissipation control parameter recording module, the heat dissipation control parameter recording module being used to collect multiple sets of historical battery operating temperatures with the goal of making the surface stable temperature of the matching test material lower than the calibrated stable temperature of the first battery group, and to test the first battery heat dissipation device in combination with the multiple sets of reference ambient temperatures, and to record multiple sets of heat dissipation control parameters;

[0093] A heat dissipation parameter decision model acquisition module is used to perform training and fitting of a meta-learner using the multiple groups of historical battery operating temperatures, the multiple groups of reference ambient temperatures and the multiple groups of heat dissipation control parameters to obtain the heat dissipation parameter decision model.

[0094] Furthermore, the embodiment of the present application also includes:

[0095] a control stable temperature acquisition module, the control stable temperature acquisition module being used to acquire M control stable temperatures of the M battery cells in a preset monitoring window after controlling the first battery heat dissipation device with the real-time heat dissipation parameter;

[0096] A control stable temperature judgment module, the control stable temperature judgment module is used to judge whether the M control stable temperatures are less than or equal to the calibrated stable temperature;

[0097] A control update instruction generation module, wherein the control update instruction generation module is used to generate a control update instruction if no, to iteratively update and adjust the heat dissipation parameters of the first battery heat dissipation device in a unit step within a predetermined adjustment range until the total adjustment step reaches a critical value of the predetermined adjustment range or the M control stable temperatures are less than or equal to the calibrated stable temperature.

[0098] Furthermore, the embodiment of the present application also includes:

[0099] A temperature regulation abnormality information generating module, wherein the temperature regulation abnormality information generating module is used to generate temperature regulation abnormality information if the total adjustment step reaches a critical value of the predetermined adjustment range and the M controlled stable temperatures are greater than the calibrated stable temperature;

[0100] An abnormality inspection module is used to perform abnormality inspection on the first battery pack and the first battery heat dissipation device based on the temperature regulation abnormality information.

[0101] For the specific implementation of the battery heat dissipation decision system based on meta-learning, please refer to the implementation of the battery heat dissipation decision method based on meta-learning above, which will not be repeated here. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0102] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0103] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A battery heat dissipation decision method based on meta-learning, characterized in that: The method comprises: Based on the historical charge and discharge rate records of the first battery group and clustering, a plurality of preset charge and discharge rates are obtained; Traversing the plurality of preset charge and discharge rates to collect historical charge and discharge temperatures of the M battery cells of the first battery group to obtain a plurality of battery temperature sequence sets; Traversing the plurality of battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis to obtain M temperature distribution deviation coefficients; Analyzing the contact areas between the flat heat pipe in the first battery heat dissipation device and the M battery cells using the M temperature distribution deviation coefficients to generate M contact area information; After optimizing the first battery heat dissipation device using the M contact area information, activating the meta-learner to perform heat dissipation parameter fitting on the optimized first battery heat dissipation device to obtain a heat dissipation parameter decision model; Collecting the real-time charge and discharge rate and the real-time ambient temperature of the first battery pack and inputting them into the heat dissipation parameter decision model to perform heat dissipation parameter fitting, and outputting the real-time heat dissipation parameters; The first battery heat dissipation device is controlled according to the real-time heat dissipation parameter.

2. The method according to claim 1, characterized in that The traversing the multiple battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis to obtain M temperature distribution deviation coefficients includes: Extracting a first battery temperature sequence set according to the multiple battery temperature sequence sets, wherein the first battery temperature sequence set includes temperature sequences of M battery cells; Performing temperature balance analysis on the temperature sequences of the M battery cells in sequence to obtain M temperature balance indicators; Comparing the M temperature balance indicators with a predetermined temperature balance indicator, and calculating a temperature distribution value based on the comparison result to obtain M temperature distribution values; Calculating a ratio of the M temperature distribution values ​​to the sum of the M temperature distribution values ​​to obtain a first initial deviation coefficient set; Similarly, the multiple battery temperature sequence sets are traversed to obtain multiple first initial deviation coefficient sets; Based on a plurality of preset charge and discharge rates, weighted calculations are performed on the initial deviation coefficients of the M battery cells in the plurality of first initial deviation coefficient sets in sequence to obtain the M temperature distribution deviation coefficients.

3. The method according to claim 2, characterized in that The step of comparing the M temperature balance indicators with a predetermined temperature balance indicator and calculating the temperature distribution value based on the comparison result to obtain the M temperature distribution values ​​comprises: Extract P temperature sequences whose temperature balance index is less than or equal to the predetermined temperature balance index, calculate the temperature mean, and obtain P temperature balance values, where P is an integer greater than 0 and less than or equal to M; Extract Q temperature sequences whose temperature balance index is greater than a predetermined temperature balance index, perform temperature concentration value analysis, and obtain Q temperature concentration values, where Q+P=M; The P temperature balance values ​​and the Q temperature concentration values ​​are integrated to obtain the M temperature distribution values ​​corresponding to the first battery temperature sequence set.

4. The method according to claim 3, characterized in that The extracting Q temperature sequences whose temperature balance index is greater than a predetermined temperature balance index and performing temperature concentration value analysis comprises: Extracting a first temperature sequence from the Q temperature sequences, and updating the sequence in descending order of temperature to obtain a first updated sequence; Dividing the first update sequence into equal distances to obtain a plurality of subsequences; Extracting multiple data distribution quantities of the multiple subsequences, and performing a data total amount difference analysis between adjacent subsequences to obtain data difference quantities of the multiple adjacent subsequences; Extracting a subsequence whose data difference amount meets a predetermined threshold and whose data distribution quantity meets a predetermined data amount, and performing temperature mean calculation to obtain a first temperature concentration value; In this way, the Q temperature sequences are traversed to obtain the Q temperature concentration values.

5. The method according to claim 1, characterized in that The activation meta-learner performs heat dissipation parameter fitting on the optimized first battery heat dissipation device to obtain a heat dissipation parameter decision model, including: Collecting the type of battery surface material of the first battery group; Performing heat conduction analysis based on the battery surface material type and the flat heat pipe to generate a heat conduction coefficient; Based on the matching of the thermal conductivity coefficient, a matching test material is obtained; Establishing multiple sets of reference ambient temperatures based on the operating environment of the first battery pack; With the goal of making the surface stable temperature of the matching test material lower than the calibrated stable temperature of the first battery pack, collecting multiple sets of historical battery operating temperatures, testing the first battery heat dissipation device in combination with the multiple sets of reference ambient temperatures, and recording multiple sets of heat dissipation control parameters; The meta-learner is trained and fitted using the multiple groups of historical battery operating temperatures, the multiple groups of reference ambient temperatures, and the multiple groups of heat dissipation control parameters to obtain the heat dissipation parameter decision model.

6. The method according to claim 5, characterized in that The method further comprises: After controlling the first battery heat dissipation device with the real-time heat dissipation parameter, collecting M controlled stable temperatures of the M battery cells in a preset monitoring window; Determining whether the M controlled stable temperatures are less than or equal to the calibrated stable temperature; If not, a control update instruction is generated to iteratively update and adjust the heat dissipation parameters of the first battery heat dissipation device in unit steps within a predetermined adjustment range until the total adjustment step reaches the critical value of the predetermined adjustment range or the M control stable temperatures are less than or equal to the calibrated stable temperature.

7. The method according to claim 6, characterized in that The method further comprises: If the total adjustment step reaches the critical value of the predetermined adjustment range, and the M controlled stable temperatures are greater than the calibrated stable temperatures, temperature adjustment abnormality information is generated; The first battery pack and the first battery heat sink are inspected and repaired for abnormalities based on the temperature regulation abnormality information.

8. A battery cooling decision system based on meta-learning, characterized in that: The system comprises: A preset charge and discharge rate acquisition module, the preset charge and discharge rate acquisition module is used to obtain a plurality of preset charge and discharge rates based on the historical charge and discharge rate records of the first battery group and clustering; a battery temperature sequence set acquisition module, the battery temperature sequence set acquisition module being used to collect historical charge and discharge temperatures of the M battery cells of the first battery group through the multiple preset charge and discharge rates to obtain multiple battery temperature sequence sets; A temperature distribution deviation coefficient obtaining module, wherein the temperature distribution deviation coefficient obtaining module is used to traverse the plurality of battery temperature sequence sets to perform charge and discharge temperature distribution deviation analysis to obtain M temperature distribution deviation coefficients; a contact area information generating module, the contact area information generating module being used to analyze the contact area between the flat heat pipe in the first battery heat dissipation device and the M battery cells using the M temperature distribution deviation coefficients to generate M contact area information; a heat dissipation parameter decision model acquisition module, wherein the heat dissipation parameter decision model acquisition module is used to optimize the first battery heat dissipation device with the M contact area information, activate the meta-learner to perform heat dissipation parameter fitting on the optimized first battery heat dissipation device, and obtain a heat dissipation parameter decision model; A real-time heat dissipation parameter output module, the real-time heat dissipation parameter output module is used to collect the real-time charge and discharge rate and the real-time ambient temperature of the first battery pack and input them into the heat dissipation parameter decision model to perform heat dissipation parameter fitting and output the real-time heat dissipation parameters; A battery heat sink control module is used to control the first battery heat sink with the real-time heat dissipation parameters.

Images