Thermal management method, device and equipment of battery, medium and program product

Abstract

The embodiment of the invention provides a thermal management method, device and equipment of a battery, a medium and a program product, which are applied to a battery management system. The method comprises the following steps: inputting a plurality of operating parameters of a battery collected in real time into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, and after determining that the battery has a thermal failure risk based on the thermal failure risk prediction value, executing a preset risk processing measure. Through the above method, the real-time monitoring and prediction of the thermal failure risk of the battery can be realized, the accuracy and response speed of the identification of the thermal failure risk of the battery are improved, the thermal management effect of the battery is remarkably improved, and the safety of the battery under various operation conditions is ensured.

Description

Technical Field

[0001] The present application relates to the field of battery technology, and in particular to a battery thermal management method, device, equipment, medium and program product. Background Art

[0002] With the rapid development of battery technology, lithium-ion batteries are widely used in electric vehicles, consumer electronics, energy storage systems, etc. due to their high energy density and long cycle life. However, lithium-ion batteries may suffer from thermal failure during use, which not only affects the performance and life of the battery, but may also cause safety hazards such as fire or explosion.

[0003] The existing battery thermal failure prediction method based on reinforcement learning, in a stable and controllable environment, the battery operating conditions such as temperature and current can be kept relatively constant, so that the reinforcement learning model can gradually learn the characteristics and patterns of battery thermal failure through a large amount of training data, thereby improving the accuracy of the prediction. However, it faces challenges such as low learning efficiency and lack of comprehensive consideration of multiple factors, resulting in the prediction accuracy still needs to be improved.

[0004] In summary, providing a battery thermal management method to improve the accuracy of identifying battery thermal failure risks and ensure the safety and reliability of the battery management system are technical issues that need to be solved urgently. Summary of the invention

[0005] The embodiments of the present application provide a battery thermal management method, device, equipment, medium and program product to more accurately predict and handle the thermal failure risk of the battery and ensure the safety and reliability of the battery management system.

[0006] In a first aspect, an embodiment of the present application provides a battery thermal management method, which is applied to a battery management system. The method includes:

[0007] Collect multiple operating parameters of the battery in real time;

[0008] Inputting the multiple operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, wherein the thermal failure risk prediction model is a model for predicting battery thermal failure obtained by training a reinforcement learning model using an adaptive learning mechanism and a multi-factor dynamic prediction method;

[0009] If it is determined that the battery has a thermal failure risk according to the thermal failure risk prediction value, a preset risk handling measure is executed.

[0010] In a possible implementation manner, before inputting the multiple operating parameters into a preset thermal failure risk prediction model to obtain the thermal failure risk prediction value of the battery, the method further includes:

[0011] Obtain a set of historical data samples of the battery, where the set of historical data samples includes: the current, temperature, and corresponding battery thermal failure status of multiple batteries during charging and discharging;

[0012] Based on the set of data samples and the reinforcement learning model, train to obtain a first prediction model;

[0013] Adopt an adaptive learning mechanism to train the first prediction model to obtain a second prediction model;

[0014] According to the multi-factor dynamic prediction method, introduce other factors affecting battery thermal failure into the second prediction model for model training to obtain the thermal failure risk prediction model.

[0015] In a possible implementation manner, the multiple operating parameters include at least three of temperature, current, voltage, internal resistance, state of charge, and charge and discharge rate.

[0016] In a possible implementation manner, the method further includes:

[0017] If the thermal failure risk prediction value is greater than a preset thermal failure risk threshold, determine that the battery has a thermal failure risk;

[0018] Otherwise, determine that the battery does not have a thermal failure risk.

[0019] In a possible implementation manner, the expression of the thermal failure risk prediction model is:

[0020]

[0021] Among them, R(t) represents the thermal failure risk prediction value of the battery at time t; D i (t) represents the value of the i-th factor of the battery at time t, D ref represents the reference value of this factor, w i represents the weight of this factor, g i (t) represents the influence coefficient of the i-th factor of the battery at time t; λ represents a regulation factor; η(t) represents the learning rate of this thermal failure risk prediction model at time t.

[0022] In a possible implementation manner, the execution of the preset risk handling measures includes:

[0023] Push a thermal failure risk warning message, and initiate at least one of a cooling mechanism, reducing the battery charge and discharge rate, and reducing the battery load.

[0024] In a possible implementation manner, the method further includes:

[0025] If the risk of thermal failure of the battery is detected continuously for multiple times, a warning message about the risk of thermal failure is pushed, and a notice message for handling the thermal risk is sent to the terminal device of the maintenance personnel;

[0026] Or,

[0027] If the predicted value of the thermal failure risk is less than the preset lower limit of the thermal failure risk continuously for multiple times, an indication message for model optimization is pushed to the server device.

[0028] In a possible implementation manner, the expression of the adaptive learning mechanism is:

[0029]

[0030] where α represents the benchmark learning rate adjustment factor; β represents the temperature response sensitivity factor; γ represents the attenuation factor of temperature on the learning rate; δ represents the adjustment factor of charge and discharge current on the learning rate; T ref represents the reference value of the battery temperature; C ref represents the reference value of the battery current.

[0031] In a second aspect, an embodiment of the present application provides a thermal management device for a battery, which is applied to a battery management system and includes:

[0032] A first processing module, configured to collect multiple operating parameters of the battery in real time;

[0033] A second processing module, configured to input the multiple operating parameters into a preset thermal failure risk prediction model to obtain a predicted value of the thermal failure risk of the battery, where the thermal failure risk prediction model is a model for predicting battery thermal failure trained by using an adaptive learning mechanism and a multi-factor dynamic prediction method for a reinforcement learning model;

[0034] A third processing module, configured to execute a preset risk handling measure if it is determined according to the predicted value of the thermal failure risk that the battery has a risk of thermal failure.

[0035] In a third aspect, an embodiment of the present application provides an electronic device, including: a memory, a processor;

[0036] The memory stores computer execution instructions;

[0037] The processor executes the computer execution instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementation manners of the first aspect.

[0038] Fourthly, an embodiment of the present application provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the computer-executable instructions are executed by a processor, they are used to implement the first aspect and / or various possible implementation manners of the first aspect as described above.

[0039] Fifthly, an embodiment of the present application provides a computer program product, including a computer program, and when the computer program is executed by a processor, it implements the first aspect and / or various possible implementation manners of the first aspect as described above.

[0040] A thermal management method, device, equipment, medium and program product of a battery provided by an embodiment of the present application are applied to a battery management system. By inputting multiple operating parameters of the battery collected in real time into a preset thermal failure risk prediction model, a thermal failure risk prediction value of the battery is obtained. After determining that the battery has a thermal failure risk based on the thermal failure risk prediction value, a preset risk handling measure is executed. Through the above method, real-time monitoring and prediction of the thermal failure risk of the battery can be realized, the accuracy and response speed of identifying the thermal failure risk of the battery are improved, the thermal management effect of the battery is significantly enhanced, and the safety of the battery under various operating conditions is ensured. Description of the Drawings

[0041] The drawings here are incorporated into the specification and constitute a part of this specification, showing the embodiments consistent with the present application, and are used together with the specification to explain the principles of the present application.

[0042] Figure 1 Schematic flowchart of a thermal management method of a battery provided by the present application Figure 1 ;

[0043] Figure 2 Schematic flowchart of a thermal management method of a battery provided by the present application Figure 2 ;

[0044] Figure 3 Schematic structural diagram of a thermal management device of a battery provided by the present application;

[0045] Figure 4 Schematic structural diagram of an electronic device provided by the present application.

[0046] Through the above drawings, the clear embodiments of the present application have been shown, and there will be more detailed descriptions later. These drawings and text descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. Detailed Embodiments

[0047] Exemplary embodiments will be described in detail herein, and examples thereof are shown in the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present application. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0048] First, the terms related to the present application are explained as follows:

[0049] Battery thermal failure: It refers to the phenomenon of functional failure or damage of the battery caused by excessive temperature during operation. Thermal failure usually occurs when the internal temperature of the battery exceeds its designed safe range, triggering a series of adverse reactions, including electrolyte decomposition, internal short circuit, thermal runaway, etc. Thermal runaway is the most serious situation, which will cause the battery temperature to rise rapidly and may trigger combustion or explosion. The reasons for battery thermal failure may include overcharging or discharging, high ambient temperature, poor heat dissipation, internal defects, etc. To prevent thermal failure, the battery management system usually monitors parameters such as the temperature, current, and voltage of the battery in real time and takes measures when abnormalities are detected, such as reducing the load, starting the cooling system, or stopping charging / discharging, to ensure the safe operation of the battery.

[0050] Next, the application background of the present application is explained as follows:

[0051] With the progress of technology and the growing demand for sustainable energy, lithium-ion batteries have become the mainstream choice in modern battery technology due to their superior performance. The high energy density enables lithium-ion batteries to store a large amount of energy in a relatively small volume, which is crucial for applications such as electric vehicles, consumer electronics such as smartphones and laptops, and large-scale energy storage systems. In addition, the long cycle life of lithium-ion batteries means that they can maintain high performance during multiple charge and discharge cycles, further promoting their wide application in various fields.

[0052] However, lithium-ion batteries face the challenge of possible thermal failure problems during use. Battery thermal failure is usually caused by the accumulation of internal heat in the battery, which may be due to factors such as overcharging, over-discharging, internal short circuit of the battery, too high external ambient temperature, or manufacturing defects. When the heat cannot be effectively dissipated, it will cause the battery temperature to continue to rise, thereby triggering a series of adverse reactions, such as electrolyte decomposition, diaphragm melting, and changes in the structure of electrode materials, resulting in a sharp decline in battery performance and even triggering thermal runaway, ultimately leading to serious safety accidents such as fire or explosion.

[0053] Traditional battery thermal failure prediction methods mainly rely on the monitoring of physical quantities such as temperature and current and simple mathematical models. The prediction accuracy is low, and it is difficult to fully consider the interaction of multiple factors. At the same time, it is also impossible to adjust the prediction strategy according to real-time data. With the development of artificial intelligence and machine learning technology, battery thermal failure prediction methods based on reinforcement learning have gradually been proposed. In the existing battery thermal failure prediction methods based on reinforcement learning, the operating conditions of the battery, such as temperature and current, can remain relatively constant in a stable and controllable environment, so that the reinforcement learning model can gradually learn the characteristics and patterns of battery thermal failure through a large amount of training data, thereby improving the accuracy of the prediction. However, it faces challenges such as low learning efficiency and lack of comprehensive consideration of multiple factors, which makes it impossible to fully analyze the battery status, and the prediction accuracy still needs to be improved.

[0054] In summary, it is a technical problem that needs to be solved urgently to provide a battery thermal management method, improve the accuracy of identifying battery thermal failure risks, ensure the safety and reliability of the battery management system, and thus better meet the needs of modern society for high-performance and safe energy storage battery solutions.

[0055] Based on the above technical problems, the inventors found that by introducing an adaptive learning mechanism into the basic prediction model based on historical battery data and combining it with a multi-factor dynamic prediction method, the prediction model can be adaptively adjusted to comprehensively analyze the battery status and improve the accuracy of the model in identifying the risk of battery thermal failure. Based on this, the present application provides a battery thermal management method, device, equipment, medium and program product.

[0056] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0057] Figure 1 A schematic diagram of a thermal management method for a battery provided in this application Figure 1 , used in battery management systems, such as Figure 1 As shown, the method includes:

[0058] S101: Collect multiple operating parameters of the battery in real time.

[0059] In this step, real-time acquisition of multiple operating parameters of the battery is one of the basic steps of battery thermal management. The multiple operating parameters include at least three of temperature, current, voltage, internal resistance, state of charge, and charge-discharge rate. Temperature is a key parameter for evaluating the thermal state of the battery, unit: degree Celsius. The temperature level of the battery directly affects the performance and safety of the battery. Excessive temperature may cause decomposition of battery materials and thermal runaway, while too low temperature will reduce the efficiency and capacity of the battery. Current is the amount of electric charge passing through the battery during charging or discharging, unit: A. Voltage is a key indicator for measuring the output ability of the battery, unit: V. The voltage level of the battery reflects the remaining energy of the battery and the current charging state. By monitoring the voltage change, the charging and discharging conditions of the battery can be identified, the risks of overcharging or over-discharging, and the voltage imbalance problem between components can be detected. Internal resistance refers to the resistance to the flow of current inside the battery, unit: Ω. A higher resistance may lead to a decrease in battery efficiency and an increase in battery heating. State of Charge (SOC) represents the percentage of the remaining battery charge in the rated capacity of the battery, and is used to estimate the available energy and endurance of the battery. The charge-discharge rate represents the ratio of the current during battery charging and discharging to the nominal capacity of the battery, and is used to measure the speed of battery charging and discharging.

[0060] In addition, the operating parameters of the battery also include, but are not limited to, State of Health (SOH) and Depth of Discharge (DOD). SOH represents the ratio of the capacity discharged from the fully charged state at a certain rate to the cut-off voltage to the corresponding nominal capacity, and is an important indicator for evaluating the degree of battery aging. DOD refers to the percentage of the capacity discharged by the battery to the rated capacity of the battery during the use of the battery. For the same battery, the set DOD depth is inversely proportional to the battery cycle life.

[0061] Specifically, the battery management system uses a monitoring system such as sensors to real-time acquire multiple operating parameters of the battery. The sensors are integrated inside the battery module or installed at key positions of the battery pack. The temperature sensor for measuring the temperature value is usually close to the surface of the battery cell to directly measure the temperature of each cell and monitor the thermal state of the battery. The current sensor for measuring the current value is installed in the circuit path of the battery to measure the current flow during charging and discharging. The voltage sensor for measuring the voltage value is directly connected to the terminals of the battery cell or the battery pack to real-time monitor the voltage level.

[0062] The sensor transmits the collected temperature values, current values, and voltage values to the central processing unit of the battery management system through the data bus, and further calculates the values of other factors. The battery management system calculates the internal resistance value through the changes in voltage and current; calculates the state of charge of the battery by using the voltage value, current value, and temperature value, and combining the coulomb counting method or the open circuit voltage method. The depth of discharge of the battery is a supplementary indicator of the state of charge, and its calculation method depends on the calculation of the state of charge; estimates the health state of the battery by the battery management system's long-term monitoring of parameters such as battery voltage, current, temperature, and internal resistance, and calculates the charge and discharge rate by the ratio of the current value to the rated capacity of the battery. By obtaining multiple operating parameters of the battery, the battery management system can monitor the health state of the battery in real time, predict potential risks using various parameters, and take corresponding protection measures when necessary to ensure the safety and performance optimization of the battery under various working conditions.

[0063] S102: Input multiple operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery. Among them, the thermal failure risk prediction model is a model obtained by training a reinforcement learning model using an adaptive learning mechanism and a multi-factor dynamic prediction method for predicting the thermal failure of the battery.

[0064] In this step, reinforcement learning is a machine learning method. The reinforcement learning model learns the best strategy through interaction with the environment to maximize the cumulative reward. The adaptive learning mechanism means that the model can dynamically adjust its strategy according to new data or environmental changes to improve the accuracy of prediction. The multi-factor dynamic prediction method means that the model considers multiple influencing factors during the prediction process and updates the prediction results in real time according to the changes of these factors. By combining the reinforcement learning model with the adaptive learning mechanism and the multi-factor dynamic prediction method, the preset thermal failure risk prediction model can more accurately predict the thermal failure risk of the battery because the model can comprehensively consider various operating parameters of the battery, such as temperature, voltage, current, internal resistance, and state of charge, etc., and dynamically adjust the prediction strategy to adapt to the changes in the battery state, providing a more reliable thermal failure risk prediction so as to take measures in advance to prevent battery failures.

[0065] In a possible implementation manner, the expression of the preset thermal failure risk prediction model obtained by combining the reinforcement learning model with the adaptive learning mechanism and the multi-factor dynamic prediction method is:

[0066]

[0067] Among them, R(t) represents the thermal failure risk prediction value of the battery at time t; D i (t) represents the value of the i-th factor of the battery at time t, D ref represents the reference value of this factor, w i represents the weight of this factor, gi (t) represents the influence coefficient of the i-th factor of the battery at time t; λ represents the adjustment factor; η(t) represents the learning rate of the thermal failure risk prediction model at time t.

[0068] Therefore, the battery management system will collect multiple operating parameters of the battery in real time, such as at least three of temperature, current, voltage, internal resistance, state of charge, and charge and discharge rate, and input them into a preset thermal failure risk prediction model to obtain the thermal failure risk prediction value of the battery. By obtaining the thermal failure risk prediction value of the battery, the health state of the battery can be managed more intelligently, the real-time monitoring and preventive maintenance of the battery state can be realized, so that the system can take necessary measures before the thermal failure occurs, thereby improving the safety of the battery system. At the same time, the battery management system can avoid downtime and maintenance costs caused by battery failures, and thus extend the service life of the battery and improve the reliability of the overall system.

[0069] S103: If it is determined that the battery has a thermal failure risk based on the thermal failure risk prediction value, then execute a preset risk handling measure.

[0070] In one case, if the thermal failure risk prediction value is greater than a preset thermal failure risk threshold, it is determined that the battery has a thermal failure risk.

[0071] Specifically, the thermal failure risk threshold is a critical value used to determine whether the battery is at risk of thermal failure. Analyzing the historical data of the battery is an important reference basis for the preset risk threshold. Identify the typical parameter values of the battery when thermal failure occurs under different operating conditions in the historical data, and based on a mathematical statistical model, determine under what parameter combinations the battery is more likely to have thermal failure, providing a reference for the preset threshold. In addition, the physical and chemical characteristics of the battery are also one of the essential reference bases. For example, the requirements for thermal management of lithium-ion batteries and lead-acid batteries are different, so their thermal failure risk thresholds are also different.

[0072] In a possible implementation, the preset thermal failure risk threshold includes a thermal failure risk upper limit value and a thermal failure risk lower limit value. Exemplarily, if the normal operating temperature of a conventional battery is 25°C - 45°C, then the thermal failure risk upper limit value can be set to be about 5°C higher than the highest normal operating temperature of the battery, for example, 50°C, and the thermal failure risk lower limit value can be set to be about 5°C higher than the lowest normal operating temperature of the battery, for example, 30°C. It should be noted that the thermal failure risk upper limit value of 50°C and the thermal failure risk lower limit value of 30°C here are only for illustrative purposes, and the present application does not make specific limitations. In actual work, when setting the thermal failure risk threshold, it is necessary to ensure the safety of the battery, optimize its performance and service life, comprehensively consider the historical data, physical and chemical properties, application scenarios, use environment of the battery, and refer to industry standards and regulatory requirements. In addition, the setting of the thermal failure risk threshold should also consider the monitoring and response capabilities of the battery management system to ensure that when the risk prediction value approaches or exceeds the preset threshold, the system can promptly take appropriate protection measures to maximize the performance and service life of the battery while ensuring safety.

[0073] Therefore, if the thermal failure risk prediction value is greater than the preset thermal failure risk threshold, it is determined that the battery has a thermal failure risk, and then the preset risk handling measures are executed, including: pushing a thermal failure risk warning message and initiating at least one of the following measures: a cooling mechanism, reducing the battery charge and discharge rate, and reducing the battery load. Among them, initiating the cooling mechanism is an effective means to directly reduce the battery temperature, which can be achieved by increasing heat dissipation, starting the cooling system, or adjusting the ambient temperature, etc. Reducing the battery temperature helps prevent material decomposition and thermal runaway. Reducing the battery charge and discharge rate can reduce the heat generation inside the battery because a higher charge and discharge rate usually leads to a greater current flow, thus generating more heat. By reducing the rate, heat accumulation can be reduced, and the risk of thermal failure can be lowered. Reducing the battery load means reducing the output power of the battery or restricting it to operate under lower load conditions, which can reduce the overall heat generation of the battery and relieve the thermal stress of the battery. Through the preset risk handling measures, the temperature and thermal load of the battery can be effectively reduced, thereby controlling the risk of thermal failure and ensuring the safe operation of the battery.

[0074] In another possible implementation, if the battery is detected to have a thermal failure risk continuously for multiple times, a thermal failure risk warning message is pushed, and a thermal risk handling notification message is sent to the terminal device of the maintenance personnel.

[0075] Specifically, if the predicted value of the thermal failure risk is greater than the preset risk threshold for N consecutive times, that is, the battery management system pushes thermal failure risk warning messages continuously for multiple times, indicating that the battery has been in a high-risk state for a period of time, a thermal risk handling notice message is sent to the terminal device of the maintenance personnel. It should be noted that the preset number of times N is a parameter for evaluating the stability of the thermal failure risk prediction model, used to determine whether the thermal failure risk prediction model is too conservative or sensitive. Generally, it can be set to 5 - 10 times. In actual applications, it can be appropriately adjusted according to the specific application scenario, usage environment of the battery, as well as the response ability and maintenance strategy of the battery management system. Exemplarily, for high-risk or critical mission application scenarios, a smaller N value can be selected to respond more quickly to potential thermal failure risks and ensure timely measures are taken. In relatively stable or low-risk application scenarios, a larger N value can be selected to reduce unnecessary alarms and maintenance interventions.

[0076] In another case, if the predicted value of the thermal failure risk is less than or equal to the preset thermal failure risk threshold, it is determined that the battery has no thermal failure risk.

[0077] In a possible implementation manner, if the predicted value of the thermal failure risk is less than the preset lower limit value of the thermal failure risk for multiple consecutive times, a model optimization indication message is pushed to the server device.

[0078] Specifically, if the predicted value of the thermal failure risk is less than the preset lower limit value of the thermal failure risk for N consecutive times, although it indicates that the battery has no thermal failure risk, it also reflects that the prediction result of the thermal failure risk prediction model is too conservative, and the model parameters can be appropriately adjusted. Therefore, the battery management system pushes a model optimization indication message to the server device.

[0079] In a possible implementation manner, model optimization can consider increasing the learning rate of the thermal failure risk prediction model to make it respond to environmental changes faster and improve the prediction ability.

[0080] Specifically, by introducing a sensitivity factor ε, the learning rate of the model can be further increased. The learning rate of the optimized model can be expressed as:

[0081] η(t + 1) = η(t)·(1 + ε)

[0082] Where ε represents the sensitivity factor dynamically adjusted according to the prediction results of the model in the past N times, and ε is positive.

[0083] By introducing the sensitivity factor, the learning rate of the thermal failure risk prediction model is increased, making the model easier to identify slight temperature changes or load changes of the battery, thereby enhancing the early warning ability of the model and improving the safety and efficiency of the battery management system.

[0084] A thermal management method for a battery provided by an embodiment of the present application inputs multiple operating parameters of the battery collected in real time into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery. Based on the thermal failure risk prediction value of the battery and a preset risk threshold, it is determined whether the battery has a thermal failure risk, and corresponding processing measures are executed. If the prediction value exceeds the preset risk threshold, the system will execute the preset risk processing measures, such as pushing warning information, starting a cooling mechanism, reducing the charge and discharge rate, or reducing the load, to effectively control the thermal failure risk. If a high-risk state is detected continuously for multiple times, a notification will be sent to the maintenance personnel for timely intervention. In addition, if the prediction value continues to be lower than the lower limit threshold, the system will push model optimization indication information to adjust the model parameters and improve its response ability. Through the above method, the battery management system can optimize its performance and service life while ensuring the safety of the battery, reduce downtime and maintenance costs, and improve the reliability of the overall system. It not only improves the safety and management efficiency of the battery, but also enhances the system's adaptability to environmental changes, ensuring the stable operation of the battery under various working conditions.

[0085] Figure 2 Schematic flow of a thermal management method for a battery provided by this application Figure 2 , such as Figure 2 shown, on the basis of the Figure 2 embodiment, before inputting multiple operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, the thermal management method of the battery further includes:

[0086] S201: Obtain a historical data sample set of the battery, where the historical data sample set includes: currents, temperatures of multiple batteries during charge and discharge, and corresponding battery thermal failure states.

[0087] In this step, the battery management system collects the historical data of the battery in real time through monitoring systems such as current sensors and temperature sensors, and obtains the currents, temperatures of multiple batteries during charge and discharge, and corresponding battery thermal failure states. The current reflects the charge flow situation of the battery during charge and discharge, and is an important indicator for evaluating the performance and health status of the battery. The change of the current can affect the temperature and internal resistance of the battery, thus having a direct impact on the thermal management of the battery. Temperature is another key parameter. Excessive temperature may cause material decomposition and thermal runaway, while too low temperature may reduce the efficiency and capacity of the battery.

[0088] Exemplarily, historical data samples are obtained: At time t, the current of the battery is recorded as 3.5 A and the temperature is 37 °C. At this time, the battery is in a normal working state and no thermal failure occurs. At time t+1, the current gradually increases to 5 A and the temperature rises to 45 °C. At this time, the battery is in a normal working state and no thermal failure occurs. At time t+3, the current further increases to 6 A and the temperature reaches 50 °C. At this time, an obvious thermal failure of the battery occurs. Through the obtained historical data samples, it shows how the changes in current and temperature affect the thermal failure state of the battery, providing an important reference basis for developing and validating the thermal failure risk prediction model and helping to identify the key parameters and conditions leading to thermal failure.

[0089] S202: Based on the data sample set and the reinforcement learning model, a first prediction model is trained.

[0090] In this step, the reinforcement learning model is used to train and develop the first prediction model for the thermal failure risk of the battery. Reinforcement learning is a machine learning method. Through interaction with the environment, the model can learn the optimal strategy to maximize the cumulative reward.

[0091] Specifically, first, the obtained data sample set is preprocessed, including operations such as denoising, missing value filling, and data standardization. Denoising can effectively remove random noise and outliers in the data, thereby improving the accuracy and reliability of the data; missing value filling ensures the integrity of the data and avoids analysis biases or model training interruptions caused by missing data; data standardization converts data with different dimensions to the same scale, making the model's sensitivity to each feature balanced during training and avoiding model bias caused by too large differences in the range of feature values. The preprocessed data samples are cleaner, more complete, and more consistent, providing a solid foundation for subsequent model training and helping to improve the prediction accuracy and stability of the model.

[0092] During the training process, the reinforcement learning model uses the current, temperature, and thermal failure state in the historical data sample set to gradually adjust its strategy to improve the prediction accuracy. Through continuous experimentation and feedback, the model can identify the behavior patterns and risk states of the battery under different parameter combinations. Finally, the first prediction model based on reinforcement learning can comprehensively consider the current value and temperature value of the battery to provide a preliminary thermal failure risk prediction.

[0093] S203: The first prediction model is trained using an adaptive learning mechanism to obtain a second prediction model.

[0094] In this step, the adaptive learning mechanism is a method for dynamically adjusting the model learning process, enabling the model to update its strategies and parameters in real time according to new data or environmental changes. This mechanism allows the model to automatically adjust its learning rate and weight allocation when facing different battery operating conditions and state changes, so as to improve the prediction accuracy and response speed. Through adaptive learning, the model can identify subtle changes in the battery operating environment and adjust its prediction strategy accordingly. For example, when there are abnormal fluctuations in the battery temperature or current, the model can quickly adapt to these changes and optimize its prediction ability. In addition, the adaptive learning mechanism can also help the model avoid overfitting or underfitting problems when the data distribution changes, thus maintaining the stability and robustness of the model.

[0095] Specifically, the battery management system adopts an adaptive learning mechanism. By obtaining the real-time temperature value and current value of the battery, it dynamically adjusts the learning rate and trains the first prediction model. In a possible implementation, the expression of the learning rate of the first prediction model at time t is:

[0096]

[0097] where α represents the reference learning rate adjustment factor; β represents the temperature response sensitivity factor; γ represents the attenuation factor of temperature on the learning rate; δ represents the adjustment factor of charge and discharge current on the learning rate; T ref represents the battery temperature reference value; C ref represents the battery current reference value.

[0098] In this expression, the integral term fully considers the influence of temperature changes on the learning rate. The exponential decay function exp(-γ·(T e (t)-T ref ) 2 ) is used to smooth the changes in battery temperature, making the influence of temperature changes on the learning rate gradually weaken. The changes in current and internal resistance during the charge and discharge process of the battery are weighted by the square difference term to ensure that the learning rate can affect the changes in battery state.

[0099] By introducing the adaptive learning mechanism, the second prediction model can effectively improve the adaptability of the model, enabling it to be optimized for different battery operating states and improving the accuracy of thermal failure prediction.

[0100] S204: According to the multi-factor dynamic prediction method, other factors affecting battery thermal failure are introduced into the second prediction model for model training to obtain a thermal failure risk prediction model.

[0101] In this step, the multi-factor dynamic prediction method is a prediction method that considers multiple influencing factors simultaneously during the model training process, which is particularly suitable for complex systems such as battery thermal management. In this method, the model not only relies on a single parameter (such as temperature or current) for prediction, but also comprehensively considers the impact of multiple key factors and their interactions on battery thermal failure, including battery voltage, internal resistance, SOC, DOD, and SOH.

[0102] Specifically, the battery management system sets m factors as input variables of the thermal failure risk prediction model. For each factor, the influence function of the factor is constructed as f i (t) = w i ·g i (t), where g i (t) represents the influence coefficient of the i-th factor, w i Represents the weight of this factor. Based on the first prediction model and combined with the influence function of multiple factors, a thermal failure risk prediction model is trained and obtained. The model expression is as follows:

[0103]

[0104] The meaning of each parameter is as follows Figure 1 The above is described in the embodiments and will not be repeated here.

[0105] In this thermal failure risk prediction model, The weighted approach takes into account multiple influencing factors. It is used to adjust the combined impact of various factors to avoid certain extreme input values ​​that make the model overly sensitive and lead to high risk prediction values.

[0106] Ultimately, the thermal failure risk prediction model trained by the multi-factor dynamic prediction method can more accurately assess the thermal failure risk of the battery under various working conditions, provide more effective monitoring and preventive measures for the battery management system, and ensure battery safety and performance optimization.

[0107] A battery thermal management method provided by an embodiment of the present application collects historical data samples of a battery in real time through current and temperature sensors, including current, temperature, and thermal failure status, providing a basis for model development. Then, a reinforcement learning model is used to train the preprocessed data to form a first prediction model. Subsequently, an adaptive learning mechanism is introduced to dynamically adjust the first prediction model so that it can update the strategy according to real-time data, improving the prediction accuracy and response speed. Finally, through a multi-factor dynamic prediction method, multiple key factors such as voltage, internal resistance, SOC, DOD, and SOH are incorporated into the model to form a more comprehensive thermal failure risk prediction model. Through the above method, the thermal failure risk prediction model can more accurately evaluate the thermal failure risk of the battery under different working conditions, not only improving the monitoring ability of the battery management system, but also enhancing its adaptability to environmental changes, ensuring the safety and performance optimization of the battery.

[0108] Figure 3 The structural schematic diagram of a battery thermal management device provided by this application is as follows Figure 3 As shown, the battery thermal management device 30 provided in this embodiment includes:

[0109] A first processing module 301, configured to collect multiple operating parameters of the battery in real time;

[0110] A second processing module 302, configured to input the multiple operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, where the thermal failure risk prediction model is a model for predicting battery thermal failure obtained by training a reinforcement learning model using an adaptive learning mechanism and a multi-factor dynamic prediction method;

[0111] A third processing module 303, configured to execute a preset risk handling measure if it is determined that the battery has a thermal failure risk according to the thermal failure risk prediction value.

[0112] In a possible implementation manner, the battery thermal management device 30 further includes a fourth processing module 304, configured to:

[0113] Obtain a historical data sample set of the battery, where the historical data sample set includes: currents, temperatures, and corresponding battery thermal failure statuses of multiple batteries during charge and discharge;

[0114] Based on the data sample set and the reinforcement learning model, train to obtain a first prediction model;

[0115] Train the first prediction model using an adaptive learning mechanism to obtain a second prediction model;

[0116] According to the multi-factor dynamic prediction method, introduce other factors affecting battery thermal failure into the second prediction model for model training to obtain a thermal failure risk prediction model.

[0117] In a possible implementation, the first processing module 301 is specifically configured to indicate that:

[0118] The multiple operating parameters include at least three of temperature, current, voltage, internal resistance, state of charge, and charge and discharge rate.

[0119] In a possible implementation, the thermal management device 30 of the battery further includes a fifth processing module 305 for:

[0120] If the predicted value of the thermal failure risk is greater than a preset thermal failure risk threshold, it is determined that the battery has a thermal failure risk;

[0121] Otherwise, it is determined that the battery does not have a thermal failure risk.

[0122] In a possible implementation, the second processing module 302 is specifically configured to indicate that:

[0123] The expression of the thermal failure risk prediction model is:

[0124]

[0125] where R(t) represents the predicted value of the thermal failure risk of the battery at time t; D i (t) represents the value of the i-th factor of the battery at time t, D ref represents the reference value of this factor, w i represents the weight of this factor, g i (t) represents the influence coefficient of the i-th factor of the battery at time t; λ represents the adjustment factor; η(t) represents the learning rate of this thermal failure risk prediction model at time t.

[0126] In a possible implementation, the third processing module 303 is specifically configured to indicate that:

[0127] Execute a preset risk handling measure, including: pushing a thermal failure risk warning message, and starting at least one of a temperature reduction mechanism, reducing the battery charge and discharge rate, and reducing the battery load.

[0128] In a possible implementation, the thermal management device 30 of the battery further includes a sixth processing module 306 for:

[0129] If the battery is detected to have a thermal failure risk continuously for multiple times, then push a thermal failure risk warning message and send a thermal risk handling notification message to the terminal device of the maintenance personnel;

[0130] Or,

[0131] If the predicted value of the thermal failure risk is less than the preset lower limit of the thermal failure risk for multiple consecutive times, an indication information for model optimization is pushed to the server device.

[0132] In a possible implementation manner, the fourth processing module 304 is specifically configured to indicate:

[0133] The expression of the adaptive learning mechanism is:

[0134]

[0135] where α represents the reference learning rate adjustment factor; β represents the temperature response sensitivity factor; γ represents the decay factor of temperature on the learning rate; v represents the adjustment factor of the charge and discharge current on the learning rate; T ref represents the battery temperature reference value; C ref represents the battery current reference value.

[0136] The thermal management device of a battery provided in this embodiment can execute the method provided in the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0137] Figure 4 This is a schematic structural diagram of an electronic device provided by this application. As Figure 4 shown, the electronic device 40 provided in this embodiment includes: at least one processor 401 and a memory 402. Optionally, the device 40 further includes a communication component 403. Among them, the processor 401, the memory 402, and the communication component 403 are connected through a bus 404.

[0138] In the specific implementation process, at least one processor 401 executes the computer-executable instructions stored in the memory 402, so that at least one processor 401 executes the above method.

[0139] For the specific implementation process of the processor 401, reference can be made to the above method embodiment, and its implementation principle and technical effect are similar, which will not be elaborated here in this embodiment.

[0140] In the above embodiment, it should be understood that the processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in combination with the invention can be directly embodied as being executed by the hardware processor, or executed by a combination of hardware and software modules in the processor.

[0141] The memory may include a high-speed memory (Random Access Memory, RAM), and may also include a non-volatile memory (Non-volatile Memory, NVM), such as at least one disk memory.

[0142] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the buses in the drawings of this application are not limited to only one bus or one type of bus.

[0143] This application also provides a computer program product, including a computer program, which implements the above method when executed by a processor.

[0144] This application also provides a computer-readable storage medium, in which computer-executable instructions are stored, and when the processor executes the computer-executable instructions, the above method is implemented.

[0145] The above-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a Static Random-Access Memory (SRAM), an Electrically Erasable Programmable Read Only Memory (EEPROM), an Erasable Programmable Read-Only Memory (EPROM), a Programmable Read-Only Memory (PROM), a Read-Only Memory (ROM), a magnetic memory, a flash memory, a magnetic disk, or an optical disk. The readable storage medium can be any available medium accessible by a general-purpose or special-purpose computer.

[0146] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be part of the processor. The processor and the readable storage medium can be located in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium can also exist as discrete components in a device.

[0147] The division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Additionally, the couplings or direct couplings or communication connections shown or discussed among each other can be indirect couplings or communication connections through some interfaces, devices or units, and can be in electrical, mechanical or other forms.

[0148] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0149] Furthermore, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0150] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in each embodiment of the present invention. And the aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory ROM, random access memory (RAM), magnetic disks or optical discs that can store program codes.

[0151] Those of ordinary skill in the art will understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it performs the steps including those of the above method embodiments; and the aforementioned storage medium includes: various media such as ROM, RAM, magnetic disk, or optical disk that can store program code.

[0152] Finally, it should be noted that those skilled in the art will readily conceive of other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses, or adaptations of the present invention, which follow the general principles of the present invention and include well-known common general knowledge or conventional technical means in the technical field not disclosed by the present invention. It is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present invention is only limited by the appended claims.

Claims

1. A battery thermal management method, characterized in that: Applied to a battery management system, the method comprises: Collect multiple operating parameters of the battery in real time; Inputting the multiple operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, wherein the thermal failure risk prediction model is a model for predicting battery thermal failure obtained by training a reinforcement learning model using an adaptive learning mechanism and a multi-factor dynamic prediction method; If it is determined that the battery has a thermal failure risk according to the thermal failure risk prediction value, a preset risk handling measure is executed.

2. The method according to claim 1, characterized in that Before inputting the plurality of operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, the method further includes: Acquire a battery historical data sample set, wherein the historical data sample set includes: current, temperature and corresponding battery thermal failure status of multiple batteries during charging and discharging; Based on the data sample set and the reinforcement learning model, a first prediction model is trained; Using an adaptive learning mechanism to train the first prediction model to obtain a second prediction model; According to the multi-factor dynamic prediction method, other factors that affect battery thermal failure are introduced into the second prediction model for model training to obtain the thermal failure risk prediction model.

3. The method according to claim 1 or 2, characterized in that: The multiple operating parameters include: at least three of temperature, current, voltage, internal resistance, state of charge, and charge and discharge rate.

4. The method according to claim 1 or 2, characterized in that: The method further comprises: If the thermal failure risk prediction value is greater than a preset thermal failure risk threshold, determining that the battery has a thermal failure risk; Otherwise, it is determined that the battery does not have a thermal failure risk.

5. The method according to claim 1 or 2, characterized in that: The expression of the thermal failure risk prediction model is: Where R(t) represents the predicted value of thermal failure risk of the battery at time t; D i (t) represents the value of the i-th factor of the battery at time t, D ref represents the reference value of the factor, w i represents the weight of the factor, g i (t) represents the influence coefficient of the i-th factor of the battery at time t; λ represents the adjustment factor; η(t) represents the learning rate of the thermal failure risk prediction model at time t.

6. The method according to claim 1 or 2, characterized in that: The implementation of preset risk treatment measures includes: Push thermal failure risk warning information, and start at least one of the following measures: cooling mechanism, reducing battery charge and discharge rate, and reducing battery load.

7. The method according to claim 1 or 2, characterized in that: The method further comprises: If the battery is detected to have a thermal failure risk for multiple consecutive times, a thermal failure risk warning message is pushed, and a thermal risk handling notification message is sent to the terminal device of the maintenance personnel; or, If the thermal failure risk prediction value is less than a preset thermal failure risk lower limit value for multiple times in a row, model optimization indication information is pushed to the server device.

8. The method according to claim 2, characterized in that: The expression of the adaptive learning mechanism is: Among them, α represents the reference learning rate adjustment factor; β represents the temperature response sensitivity factor; γ represents the attenuation factor of temperature on the learning rate; δ represents the adjustment factor of charge and discharge current on the learning rate; T ref Indicates the battery temperature reference value; C ref Indicates the battery current reference value.

9. A thermal management device for a battery, characterized in that: Applications in battery management systems, including: A first processing module, used for collecting multiple operating parameters of the battery in real time; A second processing module is used to input the multiple operating parameters into a preset thermal failure risk prediction model to obtain a thermal failure risk prediction value of the battery, wherein the thermal failure risk prediction model is a model for predicting battery thermal failure obtained by training a reinforcement learning model using an adaptive learning mechanism and a multi-factor dynamic prediction method; The third processing module is configured to execute a preset risk processing measure if it is determined that the battery has a thermal failure risk according to the thermal failure risk prediction value.

10. An electronic device, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 8 when executed by a processor.

12. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 8 when being executed by a processor.

Images